Text

Doe/Eia 2001, Website Reference Used in Chapter 8 NMP FSEIS
	ML060940443
Person / Time
Site:	Nine Mile Point
Issue date:	02/15/2001
From:	; US Dept of Energy, Office of Integrated Analysis & Forecasting
To:	; Office of Nuclear Reactor Regulation
References
	DOE/EIA-0628(2000)
	Download: ML060940443 (116)
	v • d • e

DOE/EIA-0628(2000)

Renewable Energy 2000:

Issues and Trends February 2001 Energy Information Administration Office of Coal, Nuclear, Electric and Alternate Fuels U.S. Department of Energy Washington, DC 20585 This report is available on the Web at:

http://www.eia.doe.gov/cneaf/solar.renewables/rea_issues/rea_issues_sum.html.

This report was prepared by the Energy Information Administration, the independent statistical and analytical agency within the U.S. Department of Energy. The information contained herein should be attributed to the Energy Information Administration and should not be construed as advocating or reflecting any policy of the Department of Energy or any other organization.

Contacts This report was prepared by staff of the Renewable Renewable Information Team (202)287-1750, e-mail Information Team, Coal, Nuclear and Renewables fred.mayes@eia.doe.gov. Questions about the prepar-Division, Office of Coal, Nuclear, Electric and Alternate ation and content of the report should be directed to Fuels. General information regarding this publication Mark Gielecki, project coordinator (202)287-1729, e-mail may be obtained from Fred Mayes, Team Leader, mark.gielecki@eia.doe.gov.

Questions regarding specific information in the report should be directed as follows:

Incentives, Mandates, and Government Mark Gielecki 202/287-1729 mark.gielecki@eia.doe.gov Programs for Promoting Renewable Fred Mayes 202/287-1750 fred.mayes@eia.doe.gov Energy Technology, Manufacturing, and Market Peter Holihan 202/287-1735 james.holihan@eia.doe.gov Trends in the U.S. and International Photovoltaics Industry The Impact of Environmental Regulation John Carlin 202/287-1734 john.carlin@eia.doe.gov on Capital Costs of Municipal Waste Combustion Facilities: 1960-1998" Forces Behind Wind Power Louise Guey-Lee 202/287-1731 louise.guey-lee@eia.doe.gov ii Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Preface Renewable Energy 2000: Issues and Trends waste disposal alternatives such as landfilling. The Impact of Environmental Regulation on Capital Costs of Renewable Energy 2000: Issues and Trends, the second in a Municipal Waste Combustion Facilities: 1960-1998" series of biannual reports, presents four articles that examines the impact of increasingly stringent environ-cover various aspects of renewable energy. The first mental regulations on the capital cost of constructing article covers financial incentives, regulatory mandates, and retrofitting MWC facilities.

and Federal research and development (R&D) programs for renewable energy in general, including renewable There is much interest in the economics of wind energy, transportation fuels. The remaining articles analyze because it is the non-hydroelectric renewable resource issues specific to a particular resource or technology. whose cost of producing electricity is the closest to that of conventional baseload power. A new vintage of wind In a time of electricity deregulation, States and the turbine technology is becoming operational, and the Federal Government are debating the pros and cons of question is how much more efficient are these turbines.

government programs to support renewable energy. Todays turbines are larger and more efficient than their Incentives, Mandates, and Government Programs for predecessors, promising increased production and lower Promoting Renewable Energy examines the role that costs. Forces Behind Wind Power examines the factors these programs have played in the past in these markets, that affect turbine performance, including siting factors and analyzes their characteristics in terms of meeting and their physical and operational characteristics. In their objectives. addition, the article discusses the effects of the restructuring of the electric power industry, and Federal Due to domestic programs like the Federal Million Solar and State incentives on the wind industry. The status of Roofs Initiative and increasing electrification worldwide, State-level wind energy activities is provided in an niche markets are expanding for solar photovoltaic (PV) appendix.

applications. Technology, Manufacturing, and Market Trends in the U.S. and International Photovoltaics The authors gratefully acknowledge the significant Industry presents a comprehensive analysis of the contributions of William King, SAIC, to the Photo-current status and the near-term prospects for global PV voltaic and Wind Power articles and Eileen Berenyi, market growth in terms of both supply and demand. Governmental Advisory Associates, Inc., to the Growth in the municipal waste combustion (MWC) Municipal Waste Combustion article; and the detailed industry leveled-off in the 1990s after rapid growth in technical reviews provided by Kevin Porter, National the 1980s. This trend is partly attributed to unfavorable Renewable Energy Laboratory, of the full report, and economics at MWC facilities relative to less expensive Harry Chernoff, SAIC, of the Incentives article.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends iii

Contents Page Incentives, Mandates, and Government Programs for Promoting Renewable Energy . . . . . . . . . . . . . . . . . . . . . 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Federal Incentives, Mandates, and Programs for Renewable Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 State Incentives, Mandates, and Programs for Renewable Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Effectiveness of Incentives, Mandates, and Government Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Technology, Manufacturing, and Market Trends in the U.S. and International Photovoltaics Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Three Major Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 U.S. and International Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 The Impact of Environmental Regulation on Capital Costs of Municipal Waste Combustion Facilities:

1960-1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Analysis and Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Appendix A. List of Projects Included in Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Appendix B. Rationale for the Use of a Capital Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Forces Behind Wind Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Wind Turbine Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 The Changing World for Wind Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Appendix A. State Wind Profiles: A Compendium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Energy Information Administration/ Renewable Energy 2000: Issues and Trends v

Tables Page Incentives, Mandates, and Government Programs for Promoting Renewable Energy

1. Timeline -- Major Tax Provisions Affecting Renewable Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2. Timeline -- Major Tax Provisions Affecting Renewable Transportation Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3. Nonutility Qualifying Facilities Using Renewable Resources as of December 31, 1998 . . . . . . . . . . . . . . . . . . 10

4. U.S. Electric Power Sector Net Summer Capability, 1989-1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

5. Electricity Generation From Renewable Energy by Energy Source, 1989-1998 . . . . . . . . . . . . . . . . . . . . . . . . . 11

6. California Nonutility Power Plants Installed Capacity, 1980-1996 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

7. Renewable Energy Production Incentive (REPI) Disbursements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

8. Patents Issued to DOE and NREL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Technology, Manufacturing, and Market Trends in the U.S. and International Photovoltaics Industry

1. U.S. Photovoltaic Cell and Module Shipments, 1983-1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

2. Global Corporate Market Share, 1994-1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3. Module and Cell Shipments by Company, 1994-1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

4. Examples of Post-1998 New Manufacturing Capacity Systems for PV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

5. Photovoltaic Module Costs (Wholesale) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

6. U.S. Exports by Country of Destination, 1993-1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

7. Japanese Photovoltaic Cell Exports and Imports, 1996 and 1997 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

8. U.S. National Photovoltaics Program Goals -- 2000-2005 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

9. U.S. Federal Photovoltaic R&D Budget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

10. Research, Development, Demonstration, and Market Stimulation Budget Comparison -- Fiscal Year 1998 . . 30

11. Examples of Photovoltaic Technology Market Development Initiatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

12. Examples of Countries with High Residential Electricity Prices Relative to the United States, 1997 . . . . . . . . 35

13. Funding for Photovoltaics/Wind World Bank China Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 The Impact of Environmental Regulation on Capital Costs of Municipal Waste Combustion Facilities: 1960-1998

1. Years Projects Began and Ceased Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

2. Number and Percent of Projects by Type of Technology and Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

3. Summary Statistics for Total Municipal Waste Combustion Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

4. Number of Firms by Years of Operating Experience and Calendar Year of Operation, Mass Burn Projects . 47

5. Number of Firms by Years of Operating Experience and Calendar Year of Operation, Modular . . . . . . . . . . 48

6. Number of Firms by Years of Operating Experience and Calendar Year of Operation, RDF Projects . . . . . . . 49

7. Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities . . . . . . . . . 55

8. Log Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities . . . . . . 56

9. Log Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities:

Mass Burn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

10. Log Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities:

Modular . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

11. Log Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities:

RDF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

12. Log Linear Regression Results Using Capital Profile Estimates of Municipal Waste Combustion Facilities: Mass Burn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

13. Log Linear Regression Results Using Capital Profile Estimates of Municipal Waste Combustion Facilities: Modular . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

14. Log Linear Regression Results Using Capital Profile Estimates of Municipal Waste Combustion Facilities: RDF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 A1. List of Projects Included in Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 vi Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Tables (continued) Page Forces Behind Wind Power

1. United States Wind Energy Capacity by State, 1998, and New Construction, 1999 and 2000 . . . . . . . . . . . . . 74

2. Definition of Classes of Wind Power Density for 50 Meter (164 Feet) Hub Height . . . . . . . . . . . . . . . . . . . . . . 76

3. Utility-Scale Wind Turbines Performance Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

4. Examples of Wind Farm Capacity Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

5. Renewable Incentives and Support Programs by State and Status of Implementing Electric Power Industry Restructuring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Figures Incentives, Mandates, and Government Programs for Promoting Renewable Energy

1. R&D Funding for Selected Renewable Energy Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Technology, Manufacturing, and Market Trends in the U.S. and International Photovoltaics Industry

1. Decline in Photovoltaic Module Prices, 1975-1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

2. World Photovoltaic Shipments, 1992-1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

3. Photovoltaic Shipments Market Share, 1992-1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

4. U.S. Photovoltaic Cell and Module Shipments, 1983-1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

5. U.S. Shipments by Cell/Module Type, 1993-1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

6. World Shipments by Module Type, 1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

7. U.S. Photovoltaic Cell and Module Shipments by End Use, 1994-1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

8. Federal Photovoltaic R&D Budgets, United States, Japan, and Germany, 1981-1999 . . . . . . . . . . . . . . . . . . . 29 The Impact of Environmental Regulation on Capital Costs of Municipal Waste Combustion Facilities: 1960-1998

1. Initial Capital Costs by Technology Type and Time Period Operations Began . . . . . . . . . . . . . . . . . . . . . . . . . 51

2. Initial Capital Costs in 1999 Dollars per Megawatt by Tons per Day . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

3. Additional Capital Costs Per Ton by Technology Type and Time Period Operations Began . . . . . . . . . . . . . 53

4. Year of Additional Capital Cost by Year Plant Began Operating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

5. Total Additional Capital Costs by EPA Regulatory Time Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

6. Average Total Capital Costs Adjusted for Depreciation by Year: All Projects . . . . . . . . . . . . . . . . . . . . . . . . . 60

7. Average Total Capital Costs Adjusted for Depreciation by Year: Mass Burn . . . . . . . . . . . . . . . . . . . . . . . . . 60

8. Average Total Capital Costs Adjusted for Depreciation by Year: Modular . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

9. Average Total Capital Costs Adjusted for Depreciation by Year: RDF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Forces Behind Wind Power

1. Wind Energy System Schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Energy Information Administration/ Renewable Energy 2000: Issues and Trends vii

Incentives, Mandates, and Government Programs for Promoting Renewable Energy by Mark Gielecki, Fred Mayes, and Lawrence Prete The intended effect of a financial incentive is to increase Introduction the production or consumption of the good or service Over the years, incentives and mandates for renewable over what it otherwise would have been without the energy have been used to advance different energy incentive. Examples of financial incentives are: tax policies, such as ensuring energy security or promoting credits, production payments, trust funds, and low-cost environmentally benign energy sources. Renewable loans. Research and development is included as a energy has beneficial attributes, such as low emissions support program because its effect is to decrease cost, and replenishable energy supply, that are not fully thus enhancing the commercial viability of the good(s) reflected in the market price. Accordingly, governments provided.4 have used a variety of programs to promote renewable Regulatory mandates include both actions required by energy resources, technologies, and renewable-based legislation and regulatory agencies (Federal or State).

transportation fuels.1 This paper discusses: (1) financial Examples of regulatory mandates are: requiring utilities incentives and regulatory mandates used by Federal and to purchase power from nonutilities and requiring the State governments and Federal research and develop- incorporation of environmental impacts and other social ment (R&D),2, 3 and (2) their effectiveness in promoting costs in energy planning (full cost pricing). Another renewables. example is a requirement for a minimum percentage of A financial incentive is defined in this report as pro- generation from renewable energy sources (viz., a viding one or more of the following benefits: renewable portfolio standard, or, RPS). Regulatory mandates and financial incentives can produce similar

A transfer of economic resources by the Govern- results, but regulatory mandates generally require no ment to the buyer or seller of a good or service that expenditures or loss of revenue by the Government.

has the effect of reducing the price paid, or, increasing the price received, respectively; It is very difficult to quantify total resource expenditures resulting from even just direct financial incentives, due

Reducing the cost of production of the good or to the large number of energy incentives that have been service; or, enacted over the past quarter of a century.5 In addition,

Creating or expanding a market for producers. the resulting interactive effect of these incentives makes 1

A renewable energy source is one that is regenerative or virtually inexhaustible. It includes biomass, geothermal, hydro (water),

municipal solid waste, solar photovoltaic, solar thermal, and wind use in the electric utility, or transportation sector.

2 The term incentive is used instead of subsidy. Incentives include subsidies in addition to other Government actions where the Governments financial assistance is indirect. A subsidy is, generally, financial assistance granted by the Government to firms and individuals.

3 The incentives examined in this article refer only to resource-based incentives. Also, this report excludes discussion of local government incentives.

4 Determining the extent to which Government energy R&D is a subsidy is . . . problematic: often it takes the form of a direct payment to producers or consumers, but the payment is not tied to the production or consumption of energy in the present. If successful, Federal-applied R&D will affect future energy prices and costs, and so could be considered an indirect subsidy. Energy Information Administration, Federal Energy Subsidies: Direct and Indirect Interventions in Energy Markets, SR/EMEU/92-02 (Washington, DC, November 1992), p. 3. In addition, Government R&D substitutes for private R&D expenditures.

5 An effort to quantify expenditures in non-energy areas is shown in an Office of Management and Budget (OMB) study, Report to Congress on the Costs and Benefits of Federal Regulations (Washington, DC, September 30, 1997). The report estimates the net benefits from Federal health, safety, and environmental regulations at between $30 billion and $3.3 trillion annually, with costs to implement them falling somewhere between $170 billion and $230 billion.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 1

it extremely difficult to correlate the effect of any one in- Reduction in cost of the renewable technology/or centive on a specific energy program or on the economy. cost competitiveness in the market A 1992 Energy Information Administration (EIA) report6 estimated the annual cost for Federal energy subsidies in Cost to consumers 1990 of between $5 billion and $10 billion. EIA recently updated certain portions of this study in order to update Market sustainability of the renewable tech-cost estimates for continuing subsidies and to provide nologies.

cost estimates for new subsidies for primary energy sources only (i.e., excluding electricity).7 This report Sustainability of the renewable technology in a com-estimated the value of Federal financial interventions petitive market is an ultimate long-term goal.

and subsidies for renewable energy at $1.3 billion. Of this amount, $725 million represents the reduction in excise tax for alcohol motor fuels.8 Federal Incentives, Mandates, and Whereas these EIA subsidy reports discussed the scope of Federal energy subsidies and attempted to measure Programs for Renewable Energy the cost of all energy subsidies, this article differs from those studies in three ways. First, this article focuses on In response to energy security concerns of the mid-regulatory and legislative mandates, as well as, financial 1970s, President Carter signed into law the National incentives and Federal R&D for renewable energy, Energy Act of 1978 (NEA), a compendium of five bills including renewable transportation fuels. Federal R&D that sought to decrease the Nations dependence on is included because its cost to the government is well foreign oil and increase domestic energy conservation measured by the Federal budget process, and R&D is and efficiency. A major regulatory mandate that has integral to lowering costs and/or reducing the time it encouraged renewable energy, the Public Utility Reg-takes for renewable technologies to become com- ulatory Policies Act of 1978 (PURPA), was established as mercially viable. Second, this article does not measure a result of the NEA. Most of the remaining Federal the total cost of incentives, though it does provide some renewable energy legislation enacted since the late 1970s measures related to incentive costs. Finally, this article are financial.

provides an assessment of the aggregate impact of the various programs for promoting renewable energy. Regulatory Mandates Generally speaking, Government policies have goals, while incentives, mandates, and Government programs Public Utility Regulatory Policies Act of 1978 in support of those policies have more specifically stated objectives. One gauge of the effectiveness of these PURPA was the most significant section of the National measures can be the progress made toward meeting Energy Act in fostering the development of facilities to objectives. The following criteria are used to evaluate generate electricity from renewable energy sources.9 the incentives, mandates, and programs discussed in However, with the electric power industry challenging this article: its legality and implementation issues, the broad appli-cation of PURPA did not occur until after the legality of

Growth in electric generating capacity using PURPA was upheld in 1981. PURPA opened the door to renewable resources competition in the U.S. electricity supply market by

Growth in electricity generation by renewable requiring utilities to buy electricity from qualifying resources facilities (QFs). QFs are defined as nonutility facilities that produce electric power using cogeneration tech-

Growth in the production of ethanol fuels nology, or power plants no greater than 80 megawatts 6

Energy Information Administration, Federal Energy Subsidies: Direct and Indirect Interventions in Energy Markets, SR/EMEU/92-02 (Washington, DC, November 1992).

7 Energy Information Administration, Federal Financial Intervention and Subsidies in Energy Markets 1999: Primary Energy, SR/OIAF/99-03 (Washington, DC, September 1999).

8 Ibid., Table 5, p. 15. Includes: Renewable Energy Production Incentive, Alternative Fuel Production Credit, Alcohol Fuel Credit, Research and Development for renewable energy, and the Federal Energy Management Program.

9 For an extensive discussion of PURPA, see Energy Information Administration, Changing Structure of the Electric Power Industry: An Update, DOE/EIA-0562 (96) (Washington, DC, December 1996).

2 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

of capacity10 that use renewable energy sources. There is incentive payments. (See Tables 1 and 2 for a summary no size restriction for cogeneration plants; however, at of major Federal provisions that affect renewable energy least 5 percent of the energy output from a qualifying and renewable-based transportation fuels, respectively.)

cogeneration facility must be dedicated to useful For renewable energy, tax credits for purchases of thermal applications. renewable energy equipment were aimed at both the residential and business sectors. Accelerated deprecia-Under PURPA, utilities are required to purchase elec- tion of renewable energy equipment and production tricity from QFs at the utilities avoided cost.11 The incentives were aimed at investors. From 1978 through Federal government, in formulating regulations, often 1998, similar types of tax credits have been in existence.

delegates implementation to the States. This occurred Over time, the various laws have usually expanded the with PURPA, as the Federal Energy Regulatory Com- technologies covered, increased the credit amount, or mission (FERC) delegated the authority for the deter- extended the time period.

mination of avoided cost to the States. In several States including California, avoided cost purchase contracts Two new types of financial incentives were introduced were very favorable to non-utility generators. For as part of the Energy Policy Act of 1992 (EPACT)&a example, between 1982 and 1988, Standard Offer 4 (SO4) production tax credit (PTC) and a renewable energy contracts written in California allowed QFs to sell re-production incentive (REPI). The PTC is a 1.5 cents-per-newable energy under 15-to-30 year terms. The contract kilowatthour (kWh) payment, payable for 10 years, to guarantees fixed payment rates (based on forecasted private investors as well as to investor-owned electric short-run avoided costs) for up to 10 years if the QF has utilities for electricity from wind and closed-loop bio-signed a contract for at least 20 years. After the 10th year, mass facilities. The REPI provides a 1.5 cents-per- kWh energy prices moved to the short-run avoided cost of the incentive, subject to annual congressional appropri-purchasing utility. The 10-year provisions were tied to ations, for generation from biomass (except municipal forecasts of increases in oil and gas prices, and were the solid waste), geothermal (except dry steam), wind and basis for the fixed payments for the first ten years of the solar from tax-exempt publicly owned utilities, local and contracts. The forecasts were much higher than prices county governments, and rural cooperatives.

actually turned out to be. Therefore, a price and revenue drop occurred in the eleventh year when the fixed contract energy prices converted to variable prices For renewable transportation fuels, tax credits and tax (based on short-term avoided cost), greatly lessening the exemptions are used to promote the use of renewable economic viability of affected projects. fuels, with the goal of displacing petroleum use in the transportation sector. There are four12 Federal tax Financial Incentives subsidies for the production and use of alcohol transpor-tation fuels: (1) a 5.4-cents-per-gallon excise tax exemp-The major Federal legislation on financial incentives for tion, 13 (2) a 54-cents-per-gallon blenders tax credit,14 renewable energy and renewable transportation fuels (3) a 10-cents-per-gallon small ethanol production tax has been structured as tax credits and production credit,15 and (4) the alternative fuels production tax.

10 In 1990, the Solar, Wind, Waste, and Geothermal Incentives Act was passed (Public Law 101-575), giving a window of opportunity for generating plants using these sources to file by Dec. 31, 1994 for QF status with an exemption on the PURPA size limit, lowering the threshold to 50 MW. Construction of the project had to be completed by 1999. The Act was not extended after its effective end date (December 31, 1994), so subsequent to 1994 the 80 megawatt size limit for these energy sources was restored.

11 Avoided cost is the cost to the utility to generate or otherwise purchase electricity from another source.

12 A fifth incentive which is an income tax deduction for alcohol produced from coal and lignite is available. However, currently no alcohol is produced from these sources. Alcohol fuel producers do not qualify for this credit if the source is biomass. Also, there is an income tax deduction for alcohol-fueled vehicles. This article discusses only incentives for renewable resources, so discussion of this deduction is not included.

13 Established by the Omnibus Budget Reconciliation Act of 1990 (P.L. 101-508), which lowered the 6-cents-per-gallon credit for gasohol established in the Tax Reform Act of 1984 (P.L.99-198).

14 Originally, the excise tax exemption was part of the National Energy Act of 1978. The excise tax credits and the blenders credit are authorized in the Intermodal Surface Transportation Acts Federal Motor Fuels Excise Tax Credit Provisions. The excise tax credits apply both to pure fuel ethanol (e.g., E-85, E-95) and to low-ethanol blends of gasoline (gasoline having as little as 5.7 percent ethanol). The Tax Reform Act of 1984 (P.L.98-369) subsequently increased the blenders income tax credit to 60 cents per gallon for ethanol, before the Omnibus Budget Reconciliation Act of 1990 lowered it to 54 cents. The blenders credit is offset by any excise tax exemptions claimed on the same fuel.

15 The credit is for a maximum of 15 million gallons annually. Eligible producers are those whose annual production is less than 30 million gallons. As with the blenders credit, the small ethanol producer credit is reduced to take into account any excise tax exemption claimed on ethanol output and sales.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 3

Table 1. Timeline % Major Tax Provisions Affecting Renewable Energy 1978 Energy Tax Act of 1978 (ETA) (P.L.95-618)

Residential energy (income) tax credits for solar and wind energy equipment expenditures: 30 percent of the first

$2,000 and 20 percent of the next $8,000.

Business energy tax credit: 10 percent for investments in solar, wind, geothermal, and ocean thermal technologies; (in addition to standard 10 percent investment tax credit available on all types of equipment, except for property which also served as structural components, such as some types of solar collectors, e.g., roof panels). In sum, investors were eligible to receive income tax credits of up to 25 percent of the cost of the technology.

Percentage depletion for geothermal deposits: depletion allowance rate of 22 percent for 1978-1980 and 15 percent after 1983.

1980 Crude Oil Windfall Profits Tax Act of 1980 (WPT) (P.L.96-223)

Increased the ETA residential energy tax credits for solar, wind, and geothermal technologies from 30 percent to 40 percent of the first $10,000 in expenditures.

Increased the ETA business energy tax credit for solar, wind, geothermal, and ocean thermal technologies from 10 percent to 15 percent, and extended the credits from December 1982 to December 1985.

Expanded and liberalized the tax credit for equipment that either converted biomass into a synthetic fuel, burned the synthetic fuel, or used the biomass as a fuel.

Allowed tax-exempt interest on industrial development bonds for the development of solid waste to energy (WTE) producing facilities, for hydroelectric facilities, and for facilities for producing renewable energy.

1981 Economic Recovery Tax Act of 1981 (ERTA) (P.L.97-34)

Allowed accelerated depreciation of capital (five years for most renewable energy-related equipment), known as the Accelerated Cost Recovery System (ACRS); public utility property was not eligible.

Provided for a 25 percent tax credit against the income tax for incremental expenditures on research and development (R&D).

1982 Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA) (P.L.97-248)

Canceled further accelerations in ACRS mandated by ERTA, and provided for a basis adjustment provision which reduced the cost basis for purposes of ACRS by the full amount of any regular tax credits, energy tax credit, rehabilitation tax credit.

1982- Termination of Energy Tax Credits 1985 In December 1982, the 1978 ETA energy tax credits terminated for the following categories of non-renewable energy property: alternative energy property such as synfuels equipment and recycling equipment; equipment for producing gas from geopressurized brine; shale oil equipment; and cogeneration equipment. The remaining energy tax credits, extended by the WPT, terminated on December 31, 1985.

1986 Tax Reform Act of 1986 (P.L.99-514)

Repealed the standard 10 percent investment tax credit.

Eliminated the tax-free status of municipal solid waste (MSW) powerplants (WTE) financed with industrial development bonds, reduced accelerated depreciation, and eliminated the 10 percent tax credit (P.L.96-223).

Extended the WPT business energy tax credit for solar property through 1988 at the rates of 15 percent for 1986, 12 percent for 1987, and 10 percent for 1988; for geothermal property through 1988 at the rates of 15 percent for 1986, and 10 percent for 1987 and 1988; for ocean thermal property through 1988 at the rate of 15 percent; and for biomass property through 1987 at the rates of 15 percent for 1986, and 10 percent for 1987. (The business energy tax credit for wind systems was not extended and, consequently, expired on December 31, 1985.)

Public utility property became eligible for accelerated depreciation.

See notes at end of table.

4 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 1. Timeline % Major Tax Provisions Affecting Renewable Energy (Continued) 1992 Energy Policy Act of 1992 (EPACT) (P.L.102-486)

Established a permanent 10 percent business energy tax credit for investments in solar and geothermal equipment.

Established a 10-year, 1.5 cents per kilowatthour (kWh) production tax credit (PTC) for privately owned as well as investor-owned wind projects and biomass plants using dedicated crops (closed-loop) brought on-line between 1994 and 1993, respectively, and June 30, 1999.

Instituted the Renewable Energy Production Incentive (REPI), which provides 1.5 cents per kWh incentive, subject to annual congressional appropriations (section 1212), for generation from biomass (except municipal solid waste),

geothermal (except dry steam), wind and solar from tax exempt publicly owned utilities and rural cooperatives.

Indefinitely extended the 10 percent business energy tax credit for solar and geothermal projects.

1999 Tax Relief Extension Act of 1999 (P.L. 106-170)

Extends and modifies the production tax credit (PTC in EPACT) for electricity produced by wind and closed-loop biomass facilities. The tax credit is expanded to include poultry waste facilities, including those that are government-owned . All three types of facilities are qualified if placed in service before January 1, 2002. Poultry waste facilities must have been in service after 1999.

A nonrefundable tax credit of 20 percent is available for incremental research expenses paid or incurred in a trade or business.

Notes: The residential energy credit provided a credit (offset) against tax due for a portion of taxpayer expenditures for energy conservation and renewable energy sources. The general business credit is a limited nonrefundable credit (offset) against income tax that is claimed after all other nonrefundable credits.

Table 2. Timeline % Major Tax Provisions Affecting Renewable Transportation Fuels 1978 Energy Tax Act of 1978 (ETA) (P.L.95-618)

Excise tax exemption through 1984 for alcohol fuels (methanol and ethanol): exemption of 4 cents per gallon (the full value of the excise tax at that time) of the Federal excise tax on gasohol (gasoline or other motor fuels that were at least 10 percent alcohol (methanol and ethanol))

1980 Crude Oil Windfall Profits Tax Act of 1980 (WPT) (P.L.96-223)

Extended the gasohol excise tax exemption from October 1, 1984, to December 31, 1992.

Introduced the alternative fuels production tax credit. The credit of $3 per barrel equivalent is indexed to inflation using 1979 as the base year, and is applicable only if the real price of oil is bellow $27.50 per barrel. The credit is available for fuel produced and sold from facilities placed in service between 1979 and 1990. The fuel must be sold before 2001.

Introduced the alcohol fuel blenders tax credit; available to the blender in the case of blended fuels and to the user or retail seller in the case of straight alcohol fuels. This credit of 40 cents per gallon for alcohol of at least 190 proof and 45 cents per gallon for alcohol of at least 150 proof but less that 190 proof was available through December 31, 1992.

Extended the ETA gasohol excise tax exemption through 1992.

Tax-exempt interest on industrial development bonds for the development of alcohol fuels produced from biomass, solid waste to energy producing facilities, for hydroelectric facilities, and for facilities for producing renewable energy.

1982 Surface Transportation Assistance Act (STA) (P.L.97-424)

Raised the gasoline excise tax from 4 cents per gallon to 9 cents per gallon, and increased the ETA gasohol excise tax exemption from 4 cents per gallon to 5 cents per gallon. Provided a full excise tax exemption of 9 cents per gallon for neat alcohol fuels (fuels having an 85 percent or higher alcohol content).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 5

Table 2. Timeline % Major Tax Provisions Affecting Renewable Transportation Fuels (Continued) 1984 Deficit Reduction Act of 1984 (P.L.98-369)

The STA excise tax exemption for gasohol was raised from 5 cents per gallon to 6 cents per gallon.

Provided a new exemption of 4.5 cents per gallon for alcohol fuels derived from natural gas.

The alcohol fuels blenders credit was increased from 40 cents to 60 cents per gallon of blend for 190 proof alcohol.

The duty on alcohol imported for use as a fuel was increased from 50 cents to 60 cents per gallon 1986 Tax Reform Act of 1986 (P.L.99-514)

Reduced the tax exemption for neat alcohol fuels (at least 85 percent alcohol) from 9 cents to 6 cents per gallon.

Permitted alcohol imported from certain Caribbean countries to enter free of the 60 cents per gallon duty.

Repealed the tax-exempt financing provision for alcohol-producing facilities.

1990 Omnibus Budget Reconciliation Act of 1990 (P.L. 101-508)

Allows ethanol producers a 10 cent per gallon tax credit for up to 15 million gallons of ethanol produced annually.

Reduced the STA gasohol excise tax exemption to 5.4 cents per gallon.

1992 Energy Policy Act of 1992 (EPACT) (P.L. 102-486)

Provides: (1) a tax credit (variable by gross vehicle weight) for dedicated alcohol-fueled vehicles; (2) a limited tax credit for alcohol dual-fueled vehicles; and (3) a tax deduction for alcohol fuel dispensing equipment.

1998 Energy Conservation Reauthorization Act of 1998 (ECRA) (P.L. 105-388)

Amended EPACT to include a credit program for biodiesel use by establishing Biodiesel Fuel Use Credits. An EPACT-covered fleet can receive one credit for each 450 gallons of neat (100 percent) biodiesel purchased for use in vehicles weighing in excess of 8500 lbs (gross vehicle weight (GVW)). One credit is equivalent to one alternative fueled vehicle (AFV) acquisition. To qualify for the credit, the biodiesel must be used in biodiesel blends containing at least 20 percent biodiesel (B20) by volume. If B20 is used, 2,250 gallons must be purchased to receive one credit.

Transportation Equity Act for the 21st Century (TEA-21) (P.L. 105-178)

Maintains, through 2000, the 5.4 cent per gallon (of gasoline) excise tax exemption for fuel ethanol set by the Omnibus Budget Reconciliation Act of 1990 (P.L. 101-508). Extends the benefits through September 30, 2007, and December 31, 2007, but cuts the ethanol excise tax exemption to 5.3, 5.2, and 5.1 cents for 2001-2002, 2003-2004, and 2005-2007, respectively, and the income tax credits by equivalent amounts. The exemption is eliminated entirely in 2008.

However, only the partial exemption from motor fuels because, when successful, it reduces the capital and/or excise tax is used to any extent. It is important to note operating costs of new products or processes. Research that there are important financial incentive issues in the and development comprises three components: basic form of tax equity regarding all of the alternate research (original investigation in some area but with no transportation fuels. However, only the alcohol fuels specific commercial objective), applied research (investi-are renewable, so this paper is confined to those. The gation with a specific commercial objective in mind),

primary incentive is the ethanol excise tax exemption. and development (translating scientific discovery into commercial products or processes).16 Research and Development The Department of Energy (DOE) applied research pro-Government research and development (R&D), espe- gram for renewable energy is accomplished through the cially applied research, is considered a support program use of partnership programs. These programs, in which 16 An alternative formulation is provided in Solar Energy Research Institute, The Potential of Renewable Energy: An Interlaboratory White Paper (SERI/TP-260-3674, March 1990), p. 29.

6 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

the Department acts primarily as a facilitator, have been Figure 1. R&D Funding for Selected Renewable a prominent part of renewables R&D funding since the Energy Technologies mid-1980s. There are two funding components to this (1999 Dollars) type of program: cost-sharing and in-kind contributions. 800 Wind Biofuels Geothermal Cost sharing refers to project funding contributions by 700 Hydropower Solar all parties involved in the project. In-kind contributions R&D Funding (Millions) 600 refer primarily to, on the company side, the payment of 500 salaries and the use of equipment and resources during the course of work on the project, and on the gov- 400 ernment side, the use of capital equipment, such as 300

scientific and engineering equipment and facilities at 200

DOEs national laboratories. (In the past, such programs

100

have included a payback feature where the contractor

repaid the government its original investment once the 0 FY78 FY83 FY88 FY93 FY98 project became commercial and profitable.) In partnering programs, the Department also works with Source: Data obtained from U.S. Department of Energy, Office of the ultimate product consumer to determine desired Budget, April 1998. Current (Then-Year) Dollars normalized to 1999 product characteristics and feeds this information back dollars. See website at http://www.eia.doe.gov/cneaf/solar.renew-ables/rea_issues/rea_issues_sum.html.

to its partner(s). For R&D projects, the private sector Note: Figure excludes the following items: Renewable Energy cost share is 20 percent. By comparison, demonstration Production Incentive Program, Ocean Energy Systems, National projects require at least a 50 percent cost share by Renewable Energy Laboratory Program Support and Resource private firms. Figure 1 shows renewable energy R&D Assessment, Alcohol Fuels, Hydrogen Research, Electric Energy Systems, Energy Storage Systems, Policy and Management, and funding over time in 1999 dollars. Renewable Indian Energy Resources.

The DOE has consistently supported solar (including solar thermal, passive solar, and photovoltaic) R&D modest studies on microwave energy from solar panels efforts at a higher level than other renewables. However, which would orbit the earth. The Department of major new Presidential biofuels energy initiatives during Agriculture (USDA) has the Alternative Agricultural the past 2 years have increased 1999 DOE R&D Research and Commercialization Corporation, a venture spending for biomass energy systems (including both capital firm for alternate energy sources. USDA also electric and transportation applications) by 64 percent joins effort with the Environmental Protection Agency to over its 1997 level. In 1999, more than 35 percent of capture methane from lagoons to supply heat and biomass energy system R&D was directed toward power.

ethanol.17 Major areas being investigated are: advanced fermentation organisms, advanced cellulase (enzyme) development, integrating the various stages of cellulose to ethanol production, and support for cellulose to State Incentives, Mandates, and ethanol demonstration production facilities.18 The prin- Programs for Renewable Energy cipal method for achieving production increases is via leveraged partnerships with private ethanol producers. Electric industry restructuring is the major issue affecting renewable energy at the State levels. In a few Other Federal agencies have also contributed to renew- States, electric industry restructuring legislation sup-able energy R&D efforts. The National Aeronautics and ports renewable energy with financial incentives Space Administration (NASA) works on fuel cell through funds from surcharges on electricity sales or research (in conjunction with DOE), solar energy renewable portfolio standards.19 Most States provide for applications in underdeveloped countries, and conducts net metering.20 Even prior to electric restructuring 17 Information on ethanol R&D expenditures is from unpublished budget documents of the U.S. Department of Energys Office of Energy Efficiency and Renewable Energy, Office of Transportation Technologies, Office of Fuels Development.

18 Cellulosic feedstocks include agricultural residues from harvesting operations (corn, wheat, rice, etc.), forest wastes/residues (excess growth, dead trees, etc.), and energy crops, i.e., trees and grasses grown specifically for use as energy feedstocks.

19 A renewable portfolio standard (RPS) is a mandate requiring that renewable energy provide a certain percentage of total energy generation or consumption.

20 Net metering refers to an arrangement that permits a facility (using a meter that reads inflows and outflows of electricity) to sell any excess power it generates over its load requirement back to the electrical grid to offset consumption.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 7

legislation, many States had financial incentives for California grew from 176 MW in 1982 to 1,015 MW in renewable energy. (A DOE-sponsored North Carolina 1985. California also strongly supported renewables State University website provides summary information, beginning in 1982 via pricing terms of the Standard updated periodically, on State-level financial incentives, Offer 4 contract mentioned earlier, which utilities were and regulatory programs and policies for renewable required to sign with qualifying facilities.

energy.)21 With the move toward deregulation and restructuring of State financial incentives include personal income tax the electric power industry, the California General credits and deductions for the purchase of various Assembly passed a law in 1996, which on March 31, renewable-based technologies or alternative fuel 1998, opened electricity markets to retail competition.

vehicles; corporate income tax credits, exemptions, and Although California had previously been aggressive in deductions for investments in renewable technologies; promoting renewable energy, Assembly Bill (AB) 1890 sales tax exemptions on renewable equipment pur- enacted an entirely different approach. It established a chases; variable property tax exemptions on the value new statewide renewables policy by providing $540 added by the renewable energy system; renewable tech- million collected from the States three largest investor-nology and demonstration project grants; and special owned utilities over 4 years starting in 1998 to support loan programs for renewable energy investments. existing, new, and emerging renewable technologies to make the transition to a competitive market. The bill also Some State incentives for renewable energy technologies allocates an additional $62.5 million for energy projects overlap the Energy Policy Act of 1992 (EPACT) Pro- deemed to be in the public interest.

duction Tax Credit (PTC). When State and Federal incentives overlap, the PTC may or may not be reduced, After the California Energy Commission submitted its depending on Internal Revenue Service rulings. In recommendations to the Legislature for allocating and California, for example, wind projects can get renewable distributing these funds ($540 million) in March 1997, resource funds without jeopardizing eligibility for the the General Assembly enacted Senate Bill 90, which PTC. In other cases, the PTC is reduced by the amount created a Renewable Resource Trust Fund containing of the State incentive.22 four accounts: Existing Renewable Resources Account

($243 million), New Renewable Resources Account ($162 million), Emerging Renewable Resources Account ($54 While some ethanol-producing States do not subsidize million), and Customer-side Renewable Resources ethanol, others offer tax incentives for gasoline blended Account ($81 million).

with ethanol and for ethanol production, which vary from $0.10 to $0.40 per gallon.

The program has a competitive bidding mechanism to reward the most cost-effective projects with a produc-California tion incentive for existing and new technologies.24 The funds are distributed by program type as follows:

Because of its long history of promoting renewable energy and the dominant position which the State holds Existing technologies: funds are distributed differ-in renewable energy production,23 this report examines entially among three technology tiers (groupings) renewable energy incentives promulgated by California. through a cents per kilowatthour production incen-From about 1980 through 1983, California had a 25- tive, with a cap of 1.5 cents per kWh. Funds for percent tax credit for wind energy systems. Combined existing technologies may decrease annually from with Federal tax credits, the effective tax credit for wind January 1, 1998, to January 1, 2002, to increase plants during that time was nearly 50 percent. It is there- funds for the development of new renewable fore hardly surprising that wind energy capacity in technologies.

21 See http://www-solar.mck.ncsu.edu/dsire.htm, June 27, 2000, and Interstate Renewable Energy Council, North Carolina Solar Center National Summary Report on State Programs and Regulatory Policies for Renewable Energy (Raleigh, NC, January 1998).

22 See, for instance, Lawrence Berkeley National Laboratory, Evaluating the Impacts of State Renewables Policies on Federal Tax Credit Programs (Berkeley, California, December 1996).

23 California has more non-hydroelectric renewable generating capability than any other State; see Energy Information Administration, Renewable Energy Annual 1999, DOE/EIA-0603(99) (Washington, DC, March 2000), Table C54.

24 Production incentives do not apply to emerging technologies.

8 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

New technologies: funds are distributed through owned utility ratepayers, be used for public interest a production incentive based on a competitive energy research development and demonstration (RD&D) solicitation process, with a cap of 1.5 cents per efforts that would not be provided adequately by either kWh, to be paid over a 5-year period after a project a competitive or regulated market. Senate Bill 90 begins generating electricity. The funds may in- required that the PIER portfolio include the following crease annually from January 1, 1998, to January 1, areas: renewable energy technologies; environmentally 2002. preferred advanced generation; energy-related environ-mental enhancements; end-use energy efficiency; and

Emerging technologies: funds are provided strategic energy research.

through rebates, buy-downs, or equivalent incen-tives to purchasers, lessees, lessors, or sellers of eligible electricity generation systems.

Effectiveness of Incentives,

Customer-side account: funds determined by Mandates, and Government dividing available funds by eligible renewable gen-eration with a 1.5-cents-per-kWh cap, and for Programs industrial customers a limit of $1,000 in rebates.

The size of this account is fixed, so that as How effective have renewable energy incentives, man-customer demand increases, the payment de- dates, and Federal and State programs been? It is creases; it is presently 1.0 cent per kWh. virtually impossible to quantify the effect of any single action, because of the interdependence of many of the By early July 1998, the new technologies auction renewable energy programs in effect at any one time.

received 56 bids representing nearly 600 megawatts of Even the effects of straightforward incentives such as the new renewable energy resources. All of the bids Renewable Energy Production Incentives (REPI) are received amounted to a total of $182 million in incentive difficult to determine, because it is not known how much payments, $20 million more than the $162 million renewable generation would have been produced in the allocated in the renewable energy program for new gen- absence of REPI. Further, REPI itself may not have been eration. Bids were used to ensure a competitive, market- sufficient to induce the renewable generation eligible for based, environment using a performance-based cri- REPI payments, but rather a combination of REPI and terion. They were submitted on a cents per kWh basis other Federal and State incentives. Following is a for electricity production, not to exceed 1.5 cents. The discussion of the effectiveness of four Federal renewable renewable resource technologies determined eligible to energy support programs&PURPA, REPI, the Federal receive funding at an average incentive of 1.2 cents per ethanol incentive program, and R&D funding. The kWh include: wind, approximately 300 megawatts (also characteristics of these programs and an assessment of eligible for the PTC); geothermal, 157 megawatts; land- whether they have proven effective in achieving the fill gas, 70 megawatts; biomass, 12 megawatts; digester desired results are discussed.

gas, 1 megawatt; and small hydro, 1 megawatt. The combined impact of all incentives (State and Federal) PURPA has assisted in bringing 290 MW of new or repowered wind capacity online in 1999.25 Thus, the incentives used This assessment of the effectiveness of PURPA is actually in California have been successful in meeting the an assessment of PURPA in combination with various tax objective of increasing the number of renewable projects incentives in place between 1978 and 1998. PURPA in the State. established a new class of generator, qualifying facilities (QF), that afforded cogenerators and certain renewable A major characteristic responsible for this success is the generators the opportunity to sell electricity to electric incentive programs competitive bidding mechanism to utilities at the utilitys avoided cost rates. These facilities reward the most cost-effective projects, using a pro- were also granted tax benefits described in Table 1, which duction incentive rather than an investment tax credit. lowered their overall costs.

Public Interest Energy Research Program (PIER) % PURPAs QF status applied to existing as well as new Assembly Bill 1890 also requires that a minimum of projects. Together, by year-end 1998, existing and new

$62.5 million in funds, collected annually from investor- projects totaled 12,658 megawatts of QF renewable 25 American Wind Energy Association, http://www.awea.org/projects/california.html, September 15, 2000.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 9

capacity (Table 3). Of this, two-thirds (8,219 megawatts) capacity increased by 11.9 percent. At the national level, of QF capacity was biomass. Some of these biomass QFs, non-hydroelectric renewable generating capacity rose by however, were not new facilities, but rather had gone 4,426 MW; the increase in hydroelectric capacity was into commercial operation prior to PURPA.26 PURPA 5,703 MW. Renewable generation rose by 22 percent enabled these facilities to connect to the grid, if they (Table 5). Most of the increase in electricity generation chose to become QFs, and sell any generation beyond from renewable energy is in the utility hydropower their own use at avoided cost rates. sector, including net imports. Nearly all of the increase in biomass, geothermal, solar, and wind generation As stated in the Introduction, two of the criteria for occurred between 1989 and 1993. Non-hydro renewable evaluating the effectiveness of incentives and mandates generation, excluding imports, actually declined by such as PURPA are renewable capacity and generation more than 5 percent between 1993 and 1998, due pri-growth. The EIA began collecting data from nonutility marily to California replacing Standard Offer 4 contract companies in 1989 (Table 4), 11 years after the passage of avoided cost provisions with competitive bidding PURPA. However, between 1989 and 1998, renewable mechanisms, and declining production at The Geysers Table 3. Nonutility Qualifying Facilities Using Renewable Resources as of December 31, 1998 Nameplate Capacity Gross Generation Fuel Source (megawatts) (thousand megawatthours)

Biomass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8,219 45,032 Geothermal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,449 9,882 Hydroelectrica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,263 5,756 Wind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,373 2,568 Solar Thermal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340 876 Photovoltaic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 11 Total Renewable QF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12,658 64,126 Total QF, All Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60,384 327,977 Total Nonutility, All Sources . . . . . . . . . . . . . . . . . . . . . . . . . 98,085 421,364 a

Conventional; excludes pumped storage.

Notes: Totals may not equal sum of components due to independent rounding.

Source: Form EIA-860B, Annual Electric Generator Report - Nonutility.

Table 4. U.S. Electric Power Sector Net Summer Capability, 1989-1998 (Megawatts)

Source 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 a

Hydroelectric ....... 74,587 73,964 76,179 74,773 77,405 78,042 78,563 76,437 79,788 79,573 Geothermal . . . . . . . . . . 2,603 2,669 2,632 2,910 2,978 3,006 2,968 2,893 2,853 2,917 Biomass . . . . . . . . . . . . 7,840 8,796 9,627 9,701 10,045 10,465 10,280 10,557 10,535 10,266 Solar/PV . . . . . . . . . . . . 264 339 323 339 340 333 333 333 334 365 Wind . . . . . . . . . . . . . . . 1,697 1,911 1,975 1,823 1,813 1,745 1,731 1,678 1,579 1,698 Total Renewables . . 86,990 87,679 90,736 89,547 92,582 93,591 93,874 91,897 95,090 94,819 Non Renewables . . . . . 637,275 647,241 649,741 657,016 662,373 670,423 675,643 683,975 683,412 681,065 Total . . . . . . . . . . . . . 724,265 734,920 740,477 746,563 754,955 764,014 769,517 775,872 778,502 775,884 a

Conventional; excludes pumped storage.

Notes: Biomass capability does not include capability of plants where the Btu of the biomass consumed represents less than 50 percent of the Btu consumed from all energy sources. Totals may not equal sum of components due to independent rounding.

Sources: Energy Information Administration, Form EIA-860A, Annual Electric Generator Report -- Utility and predecessor forms, and estimated data using Form EIA-860B, Annual Electric Generator Report -- Nonutility, and predecessor form.

26 Sources: See Table 6 of this report, as well as the Renewable Electric Plant Information System (REPiS Database), developed by the National Renewable Energy Laboratory. See http://www.eren.doe.gov/repis, February 15, 2000. These data include facilities which have retired since 1996.

10 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 5. Electricity Generation From Renewable Energy by Energy Source, 1989-1998 (Thousand Kilowatthours)

Source 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 Nonutility Sector (Gross Generation)

Biomass . . . . . . . . . . . . . . . . . . . . . . . . 36,350,275 42,499,581 48,259,818 53,606,891 55,745,781 57,391,594 57,513,666 57,937,058 55,144,102 53,744,724 Energy Information Administration/ Renewable Energy 2000: Issues and Trends Geothermal . . . . . . . . . . . . . . . . . . . . . 5,416,495 7,235,113 8,013,969 8,577,891 9,748,634 10,122,228 9,911,659 10,197,514 9,382,646 9,881,958 Hydroelectric . . . . . . . . . . . . . . . . . . . . 7,124,418 8,152,891 8,180,198 9,446,439 11,510,786 13,226,934 14,773,801 16,555,389 17,902,653 14,632,521 Solar . . . . . . . . . . . . . . . . . . . . . . . . . . 488,527 663,387 779,206 746,277 896,796 823,973 824,193 902,830 892,892 886,553 Wind . . . . . . . . . . . . . . . . . . . . . . . . . . 1,832,537 2,250,846 2,605,505 2,916,379 3,052,416 3,481,616 3,185,006 3,399,642 3.248,140 3,015,497 Total . . . . . . . . . . . . . . . . . . . . . . . . . 51,212,252 60,801,818 67,838,696 75,293,877 80,954,413 85,046,345 86,208,325 88,992,433 86,569,433 82,161,253 Electric Utility Sector (Net Generation)

Biomass . . . . . . . . . . . . . . . . . . . . . . . . 1,959,864 2,064,331 2,038,229 2,088,109 1,986,535 1,985,463 1,647,247 1,912,472 1,983,532 2,024,377 Geothermal . . . . . . . . . . . . . . . . . . . . . 9,341,677 8,581,228 8,087,055 8,103,809 7,570,999 6,940,637 4,744,804 5,233,927 5,469,110 5,176,280 Hydroelectric . . . . . . . . . . . . . . . . . . . . 265,063,067 283,433,659 280,060,621 243,736,029 269,098,329 247,070,938 296,377,840 331,058,055 341,273,443 308,843,770 Solar . . . . . . . . . . . . . . . . . . . . . . . . . . 2,567 2,448 3,338 3,169 3,802 3,472 3,909 3,169 3,481 2,518 Wind . . . . . . . . . . . . . . . . . . . . . . . . . . 479 398 285 308 243 309 11,097 10,123 5,977 2,957 Total . . . . . . . . . . . . . . . . . . . . . . . . . 276,367,654 294,082,064 290,189,528 253,931,424 278,659,908 256,000,819 302,784,897 338,217,746 348,734,543 316,050,902 Imports and Exports Geothermal (Imports) . . . . . . . . . . . . . 533,261 538,313 736,980 889,864 877,058 1,172,117 884,950 649,514 16,493 45,145 Conventional Hydroelectric (Imports) . 19,148,542 16,302,116 22,318,562 26,948,408 28,558,134 30,478,863 28,823,244 33,359,983 27,990,905 26,031,784 Conventional Hydroelectric (Exports) . 5,464,824 7,543,487 3,138,562 3,254,289 3,938,973 2,806,712 3,059,261 2,336,340 6,790,778 6,158,582 Total Net Imports . . . . . . . . . . . . . . . 14,216,980 9,296,942 19,916,921 24,583,983 25,496,219 28,844,268 26,648,933 31,673,157 21,216,620 19,918,347 Total Renewable Electricity Generation . . . . . . . . . . . . . . . . . . . . . . 341,796,886 364,180,824 377,945,145 353,809,284 385,110,540 369,891,432 415,642,155 458,883,336 456,520,167 418,129,367 Note: Totals may not equal sum of components due to independent rounding.

Sources: Nonutility Sector -- 1989-1997: Energy Information Administration, Form EIA-867, Annual Nonutility Power Producer Report. Nonutility Sector -- 1998: Energy Information Administration, Form EIA-860B, Annual Electric Generator Report -- Nonutility. Electric Utility Sector -- 1989-1997: Energy Information Administration, Form EIA-860, Annual Electric Generator Report. Electric Utility Sector -- 1998: Form EIA-860A Annual Electric Generator Report -- Utility. Imports and Exports: Energy Information Administration, Renewable Energy Annual, DOE/EIA-0603(95-99) (Washington, DC).

11

geothermal plant. Also, in 1992, New York amended its increased just 3 percent to 3,878 megawatts, and be-Six-Cent Rule, which established a 6-cents-per-kilowatt- tween 1993 and 1995, capacity actually dropped to 3,553 hour0.0064 days 0.154 hours 9.143518e-4 weeks 2.104165e-4 months floor on avoided costs for projects less than 80 MW megawatts; generation followed a similar pattern. The in size, such that it was not applicable to any future principal reasons for this decline were the lower PURPA power purchase agreements.27 avoided costs when the long-term energy payment provisions of the contracts (usually 10-years), mostly Data on renewable capacity in California were available signed in the early 1980s, expired. Natural gas prices in for years prior to 1989. These data, for 1980 through 1996 nominal dollars paid by electric utilities in California (Table 6), more clearly show the growth in renewable declines from a high of $6.77 per million Btu in 1982 to capacity owned by nonutilities since the passage of between $2.50 to $3.00 in 1986 through 1993. By 1995, the PURPA. Renewable-based nonutility capacity (excluding price declined further to $2.22.28 This, along with the cogeneration) rose from 187 megawatts in 1980 to 3,777 repeal of the standard investment tax credits in 1986, megawatts (excluding small hydropower and cogenera- caused some wind, biomass, and solar facilities to tion plants) in 1996. reduce output or cease operation.29 Also, there was a substantial slowdown in the construction of new capa-Most of the growth had occurred by 1990. Between 1990 city. This slowdown transpired despite substantial and 1993, California nonutility renewable capacity (ex- decreases in short-run average costs of renewables cluding small hydropower and cogeneration plants) because the operation costs were not reduced enough to Table 6. California Nonutility Power Plants Installed Capacity, 1980-1996 (Megawatts)

Year Cogenerationa Waste-to-Energyb Geothermal Small Hydro Solar Wind Total 1980 . . . . . . . . . 227 14 0 0 0 173 414 1981 . . . . . . . . . 261 14 0 0 0 176 451 1982 . . . . . . . . . 412 32 0 48 1 176 669 1983 . . . . . . . . . 658 46 9 59 8 227 1,007 1984 . . . . . . . . . 893 79 96 67 27 496 1,658 1985 . . . . . . . . . 1,444 140 178 107 57 1,015 2,941 1986 . . . . . . . . . 1,788 275 188 144 122 1,235 3,752 1987 . . . . . . . . . 3,063 396 319 176 155 1,366 5,475 1988 . . . . . . . . . 3,662 513 587 229 221 1,378 6,590 1989 . . . . . . . . . 4,942 783 806 298 301 1,382 8,512 1990 . . . . . . . . . 5,315 878 870 321 381 1,647 9,412 1991 . . . . . . . . . 5,838 883 813 330 374 1,698 9,936 1992 . . . . . . . . . 5,684 804 831 371 408 1,729 9,827 1993 . . . . . . . . . 5,778 845 863 370 373 1,797 10,026 1994 . . . . . . . . . 5,857 795 863 410 373 1,629 9,927 1995 . . . . . . . . . 6,280 709 846 349 368 1,630 10,182 1996 . . . . . . . . . 6,177 823 885 362 360 1,709 10,316 a

Includes gas-fired facilities and biomass co-firing and cogeneration.

b Waste-to-Energy includes wood and wood waste, municipal solid waste, landfill gas, and other biomass. However, biomass co-firing and cogeneration capacity is included under Cogeneration.

Source: California Energy Commission, Draft Final Report, California Historical Energy Statistics, January 1998, Publication Number: P300-98-001.

Notes: Data exlude facilities rated less than 5 megawatts. Some data in this table are inconsistent with national data in Table 4 due to different sources, categories, and coverage. Also, these data represent installed capacity, while the data in Table 4 represent net summer capability.

27 In 1981, New York State enacted legislation which established a minimum price of 6 cents per kilowatthour for utility purchases from QFs. As a result, nearly one-third of New Yorks generation comes from QFs. (See Edison Electric Institute, 1996 Capacity and Generation of Non-Utility Sources of Energy, 30 (1997).)

28 Energy Information Administration, State Energy Price and Expenditures Report 1995, DOE/EIA-0376(95) (Washington, DC, August 1998), p. 50.

29 Science Applications International Corporation, Assessment of Incentives for Renewable and Alternative Fuels, prepared for the Energy Information Administration (McLean, VA, September 1998).

12 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

be competitive in the market conditions of the mid-to- lowest cost of all electricity resources.37 In analyzing late 1990s.30 these data, the reader should bear in mind that by 1995, many of the original PURPA power purchase contracts Another criterion in evaluating the effectiveness of between utilities and nonutilities had expired. Therefore, PURPA, in addition to expansion of renewable energy the data reflect a mixture of the original avoided cost capacity and generation, is the cost competitiveness of contracts and newer contracts.38 the renewable facilities in the market. Utility wholesale power purchases from other utilities, which are more Renewable-based generation costs would obviously often made on a mutually agreeable economic basis have compared much more favorably with other genera-between utilities and may be regarded as reflecting tion costs during 2000, when California experienced wholesale prices, averaged 3.53 cents per kWh nation- severe electricity and natural gas shortages. Natural gas wide in 1995.31 Although EIA has not attempted to prices&the primary basis for determining alternative estimate the cost of PURPA directly,32 it has examined generation cost&rose sharply during 2000. Through the prices that utilities paid in 1995 to purchase power September, the average cost of gas delivered to electric from nonutilities and, in particular, PURPA QF non- utilities in California increased to $4.32 per million Btu utilities using renewable resources.33 The average price as compared to $2.68 for deliveries through September utilities paid all nonutilities was 6.31 cents per kWh 1999.39 nationwide, considerably higher than the average whole-sale price. Higher still was the price utilities paid non- Renewable Energy Production Incentive (REPI) utilities for renewable-based electricity. Utilities paid an average of 8.78 cents per kWh for power generated from Initial payments under the Energy Policy Act of 1992 renewable sources, compared with 5.49 cents per kWh (EPACT) Renewable Energy Production Incentive (REPI, for power from non-renewable sources.34 Utilities paid summarized in Table 1), for Fiscal Year (FY) 1994 totaled an average of 9.05 cents per kWh for nearly 42,800 $693,120 and were distributed among four State-owned million kWh of power from renewable QFs in 1995, and three city-owned facilities which generated 42 mil-compared with just 5.17 cents per kWh for 3,300 million lion kWh of electricity from seven facilities (Table 7). One kWh of power from non-QF renewables. This difference used wind, two used solar photovoltaics (PV), and four was even more extreme in California, where the renew- used methane from landfills.40 By FY 1998, net gener-able QF/non-QF purchased power costs were 12.79 and ation eligible for REPI payment had reached 529 million 3.33 cents per kWh, respectively.35 All non-QF purchases kWh from 19 facilities. Interesting points to note about of renewable energy, however, were from hydropower the REPI program are: (1) The number of facilities has facilities,36 the lowest cost renewable resource%and the remained relatively stable since FY 1996; (2) The number 30 In fact, the result of PURPA and California/Federal financial energy incentive programs of the late 1970s and early 1980s was that the proportion of natural gas-fired nonutility capacity (cogeneration) actually increased between 1980 and 1993, from 55 to 57 percent.

31 Energy Information Administration, Renewable Electricity Purchases: History and Recent Developments, from Renewable Energy 1998: Issues and Trends, DOE/EIA-0628(98) (Washington, DC, March 1999), Figure 1, p. 2.

32 For a private analysis of PURPA costs, see, Utility Data Institute, Measuring the Competition: Operating Cost Profiles for U.S. Investor-Owned Utilities 1995, 1(1996).

33 Energy Information Administration, Electric Power Monthly, DOE/EIA-0226 (2001/01) (Washington, DC, January 2001), Table 42.

34 Ibid, Figure 2.

35 Refer to Federal Energy Regulatory Commission, FERC Form 1, Annual Report of Major Electric Utilities, Licensees and Others, Energy Information Administration, Form EIA-412, Annual Report of Public Electric Utilities, and Rural Utilities Service, RUS Form 7, Financial and Statistical Report, RUS Form 12a through 12i, Electric Power Supply Borrowers, and RUS Form 12c through 12g, Electric Distribution Borrowers with Generating Facilities.

36 The reverse is not true, however. Fifty-five percent (4,474 MWh) of total hydropower purchases in 1995 were from QFs. However, these purchases represented only 10 percent of total 1995 utility power purchases from QFs, so a QF/non-QF comparison is still largely a non-hydro/hydro comparison.

37 California, which accounted for almost 40 percent of U.S. renewable power purchases in 1995, did not use market transaction costs for the first round of PURPA contracts. However, since avoided costs are defined by the States, some States may have done so.

38 The California Energy Commission and the California Public Utilities Commission estimated in 1988 above-market costs of electricity due to Standard Offer 4 (SO4) contracts. While their approach only looked at nonutility facilities with SO4 contracts having prices based on 1983 forecasts of natural gas prices, the study unfortunately does not break out costs associated with renewables. See California Energy Commission/California Public Utilities Commission, Final Report to the Legislature on: Joint CEC/CPUC Hearings on Excess Electrical Generating Capacity, P150-87-002 (Sacramento, CA, June 1988).

39 Energy Information Administration, Electric Power Monthly, DOE/EIA-0226 (2001/01) (Washington, DC, January 2001), Table 42.

40 For a complete discussion of REPI payments, see website http://www.eren.doe.gov/power/repi.html, December 17, 1999.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 13

Table 7. Renewable Energy Production Incentive (REPI) Disbursements Net Generation Nominal Payments Fiscal Year Facilities Energy Source (million kWh) (thousand dollars) 1994 . . . . . . . . . . . . . . . . . . . . . . .

2 Solar PV 8 1 Wind 93 4 Landfill Methane 592 Total . . . . . . . . . . . . . . . . . . . . . 7 42 693 1995 . . . . . . . . . . . . . . . . . . . . . . .

4 Solar PV 15 2 Wind 205 5 Landfill Methane 2,178 Total . . . . . . . . . . . . . . . . . . . . . 11 153 2,398 1996 . . . . . . . . . . . . . . . . . . . . . . .

9 Solar PV 28 3 Wind 205 5 Landfill Methane 1,879 1 Biomass Digester Gas 417 Total . . . . . . . . . . . . . . . . . . . . . 18 177 2,529 1997 . . . . . . . . . . . . . . . . . . . . 2 Solar PV 31 3 Wind 123 8 Landfill Methane 1,212 1 Biomass Digester Gas 265 1 Wood Waste 1,222 Total . . . . . . . . . . . . . . . . . . . . . 15 458 2,853 1998 . . . . . . . . . . . . . . . . . . . . . . .

3 Solar PV 91 5 Wind 31 9 Landfill Methane 1,716 1 Biomass Digester Gas 359 1 Wood Waste 1,803 Total . . . . . . . . . . . . . . . . . . . . . 19 529 4,000 Source: http://www.eren.doe.gov/power/repi.html (October 22, 1999).

of solar/PV facilities has been quite modest, except for be a factor in maintaining production from economically a one-time increase in FY 1996 which did not result in a marginal wind farms, or, more likely, because of the sizable increase in REPI-eligible generation; and (3) The uncertainty associated with year-to-year congressional greatest increase in both eligible facilities and generation appropriations, or both. For existing biomass generators, occurred in two areas, landfill methane and wood waste, whose variable costs per kWh are generally higher than which are often excluded (along with municipal solid those for wind generators, the 1.5-cents-per-kWh credit waste) from actual and proposed renewable energy is much less likely to support continued operation of incentives; and (4) only tax-exempt facilities are eligible. marginal plants.

It is important to note that while the generation eligible Federal Ethanol Incentive Program for REPI payments increased more than twelvefold, the number of facilities receiving REPI support increased Prior to the Federal ethanol subsidy program, begun in only threefold, and that increase occurred during the 1979,41 the United States produced virtually no fuel first 3 years of the program. This could have occurred ethanol. In the first year of the subsidy program, the because the 1.5 cents per kWh has not been sufficient to United States produced 10 million gallons. Production encourage much additional construction, though it may increased rapidly, to 175 million gallons in 1981, 870 41 The ethanol subsidy program began with a provision of the Energy Tax Act of 1978. This provision suspended the Federal excise tax on gasoline blended with alcohol derived from biomass (e.g., corn).

14 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

million gallons in 1990, 1.4 billion gallons in 1998, and 1999. Clearly, increasing the ethanol content of gasoline 1.5 billion gallons in 1999.42 Virtually all production is in in the near term increases its cost vis-a-vis MTBE-based the Midwest, and fuel ethanol stocks are sizable only in gasoline.

the Midwest and Gulf Coast regions.

It is also important to note that ethanols one-third share To determine what production of ethanol would be of the oxygenate market is concentrated in the Midwest without the subsidies, it is necessary to analyze ethanols where most of the corn is grown. Many States in the three distinct purposes as an additive to gasoline. Midwest have sizable ethanol support programs.46 Originally, it was used to extend gasoline supplies as gasohol, a mixture of 10 percent ethanol and 90 per- The use of MTBE in some parts of the country may have cent gasoline. As such, it was necessary for ethanol to less to do with economics than with the cost of trans-compete economically with gasoline, necessitating the porting ethanol far from where it is produced. Ethanol 54-cent-per gallon subsidy of corn-based ethanol. Etha- is splash blended at gasoline distribution tank farms nol also is used to raise the octane level of gasoline&its because it cannot be transported via pipeline.

octane rating is 133. Beginning in the late 1970s, the use of lead, the only major octane enhancer used until then, Assessments of repealing the Federal ethanol subsidies was phased down. Both MTBE43 and ethanol were used. differ widely, from no industry47 to the continuance of the market (about one-third of the current market for For octane-enhancing purposes, MTBE has a clear eco- ethanol) for the use of ethanol as an oxygenate. Clearly, nomic advantage over ethanol. More recently, ethanol the continuance of State support for ethanol is a critical and MTBE have been added to gasoline as an oxygenate issue if the Federal subsidies were to repealed.

to reduce harmful emissions. The incremental cost per gallon of MTBE-based gasoline (which receives no Returns to Research and Development subsidy) is 2 to 3 cents per gallon. Using a 7.7 percent blend of ethanol, the value of the ethanol subsidy alone Returns to renewable energy R&D are difficult to cal-in a gallon of gasoline would be 4.1 cents. The total culate, especially, given the diffuse nature of R&D incremental cost per gallon of ethanol-based gasoline is activity. Research and development is conducted in a 4.4 cents.44 While MTBE has an economic advantage per number of countries world wide, and the learning effects gallon of additive, ethanol has a higher oxygen content cross borders and cannot always be attributed to a than MTBE. Thus, only about half the volume of ethanol specific R&D activity.

is required to produce the same oxygen level in gasoline as if MTBE is used. This allows ethanol, typically more If the goal of R&D is to lower costs, then one measure of expensive than MTBE per unit of product, to compete effectiveness is to examine the cost of renewable tech-favorably with MTBE for the wintertime oxygenate nologies over time. For the Sacramento Municipal Utility market.45 However, recent EPA Tier 2 requirements District (SMUD), which has the largest distributed utility for summer time reformulated gasoline made it PV system in the world, the PV system average cost necessary to increase the ethanol content to 13 percent in (1996 dollars) per watt has fallen from $79 in 1975 to 42 Source: 1980-1992, Renewable Fuels Association (see website http://www.ethanolrfa.org/outlook99/99industryoutlook.html); 1993-1999, Energy Information Administration, EIA-819M Monthly Oxygenate Telephone Report (January 2000 and prior issues).

43 Methyl Tertiary Butyl Ether is a fuel oxygenate produced by reacting methanol with isobutylene.

44 This calculation is based on the average prices of gasoline and ethanol between July 1998 and June 1999 and the ethanol subsidy in effect then of 54 cents per gallon of ethanol. See http://www.cnie.org/nle/eng-59.html#_1_13, Table 5.

45 The continued need for octane levels in gasoline initially left the refiner with few choices: increase the aromatic and olefin contents of the fuel, or seek alternative products with favorable blending and performance properties. The increased use of aromatics and olefins meant more severe refinery processes needed to be used, having lower yields per barrel and higher costs for the final gasoline product.

Additionally, potential health concerns about these components&from both the direct exposure due to evaporation from the gasoline and the reaction of combustion products contributing to ozone formation&limited the levels at which it was desirable to blend them into fuel.

Methanols use ceased when the Environmental Protection Agency approved MTBE in 1979.

46 Many corn-producing States mandate the use of methanol. In Minnesota, for example, the Omnibus Environment, Natural Resources and Agriculture Appropriations bill (SF 3353) mandated that ethanol plants in the State attain a total annual production level of 240 million gallons per year, enough ethanol to completely satisfy in-State demand. Minnesota will now allocate up to $36.4 million per year for payments to the States ethanol producers.

47 See GAO Congressional testimony, http://frwebgate.access.gpo.gov/cgi-bin/useftp.cgi?IPaddress=162.140.64.21&filename=

gg97041.txt&directory=/diskb/wais/data/gao, August 4, 2000.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 15

$11.88 in 1990, to $4.90 in 1998 and to $3.65 in 2000.48, 49 Table 8 Patents Issued to DOE and NREL Also, the cost of wind power has declined markedly. Fiscal Year Number of Patents The average cost of electricity from wind energy has 1981 ..................... 1 dropped from 50 cents per kilowatthour in 1980 to a pro-1982 ..................... 0 jected 6 cents per kilowatthour in 2000 in favorable wind 1983 ..................... 1 regimes.50 Despite these successes in reducing costs, 1984 ..................... 3 these technologies are still not generally commercially 1985 ..................... 14 viable. 1986 ..................... 7 1987 ..................... 13 Another performance measure of applied R&D success 1988 ..................... 2 is inventions patented. In order to protect the rights to 1989 ..................... 4 an invention, a patent is usually applied for.51 A patent 1990 ..................... 6 1991 ..................... 8 has to be obtained within 1 year of publishing the results 1992 ..................... 7 of the relevant research in order to gain protection in the 1993 ..................... 18 United States, and immediately upon publication to 1994 ..................... 17 obtain protection abroad. This is generally insufficient 1995 ..................... 41 time for market studies, so that more patents are applied 1996 ..................... 17 for than are commercially successful. In general, fewer 1997 ..................... 16 than 10 percent of patents are licensed and, therefore, 1998 ..................... 25 commercialized. The number of patents resulting from Source: National Renewable Energy Laboratory.

renewable energy R&D is therefore considered as a proxy for returns to R&D (Table 8). For the reasons stated above, however, it is a very crude measure of success of R&D expenditures. In addition, the market Summary success of any one product (resulting from one patent) can dwarf the successes of numerous other products, yet The effectiveness of tax credits and production incen-be sufficient to spawn a new industry. This thereby tives has varied considerably, depending on the results in large returns to R&D. Finally, there is a widely amounts and certainty of the incentive. The long-term varying, unknown time lag between R&D efforts and nature and financial support levels of the PURPA successes. Given these conditions, annual patent Standard Offer 4 contracts in California, in addition to counts are, at best, only a very general indicator of R&D the Federal and State tax credits, provided reasonable success. It should be noted that the counts include only assurance that investors in power plants using renew-patents issued to DOE and the National Renewable able resources would make a profit.52 In contrast, the Energy Laboratory (NREL) on inventions reported Renewable Energy Production Incentive of EPACT relies during each listed fiscal year for contracts with NREL upon year-to-year congressional funding, raising the and its predecessor, the Midwest Research Institute. It level of uncertainty investors face. It has resulted in only does not include patents retained by DOE contractors. a small amount of additional renewable generating 48 Sources: Sacramento Municipal Utility District, Sacramento, CA, 1975-1990: Photovoltaic Validation Study; 1998 and 2000: American Solar Energy Society, Advances in Solar Energy XIV, 2000, Sustained Orderly Development and Commercialization of Grid-Connected Photovoltaics: SMUD as a Case Example, Donald E. Osborn, Sacramento Municipal Utility District, February 24, 2000.

49 Because of SMUDs long experience with PV technology and the high volume of their PV purchases and installations, it is likely that their costs are lower than for others.

50 Energy Information Administration, Annual Energy Outlook 2000, DOE/EIA-0383(2000) National Energy Modeling System run AEO2k.d100199A.

51 A patent is a grant by the United States Patent and Trademark Office to the inventor, of the right to exclude others for a period of 17 years from making, using, or selling the invention throughout the country. Thus, the primary reason to apply for a patent is to provide exclusive commercial rights for viable inventions.

52 Energy Information Administration, Renewable Energy 1998: Issues and Trends, DOE/EIA-0628(98) (Washington, DC, March 1999),

p. 65. See also, Lawrence Berkeley Laboratory, R. Wiser and E. Kahn, Alternative Windpower Ownership Structures: Financing Terms and Project Costs, May 1996, LBNL-38921. According to this study, the most important variable in comparing wind and natural gas project costs is the relatively low return on equity (12 percent) that is required by investors in gas projects compared to 18 percent for wind projects.

16 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

facilities. Other tax credits (e.g., the residential solar/ renewable energy cost reduction, the data suggest that wind tax credit) have generally had much less impact, such benefits have occurred.

simply because the gap between competitive energy prices and energy production costs is greater than the Together, the Federal and State incentives, mandates, benefit investors perceive such tax credits are worth. and support programs, including R&D, have been effective when measured by growth in electric gen-In the case of alcohol fuels, the impact of the Federal 54 erating capacity and electricity generation, or, in the cents per gallon incentive was substantial and immed- transportation sector with growth in ethanol production.

iate. Production of fuel ethanol would no doubt drop However, they failed to ensure the future self-sustaina-sharply if the Federal 54 cents per gallon (of ethanol) bility of renewable facilities that would substantially incentive were removed and States provided no contribute to the overall energy security policy of the era supports for, or, mandates to use, ethanol. in which the incentives were created. One reason for this is that although there have been some reductions in the The cost of photovoltaic and wind electricity generation cost of renewable electric generating technologies, these has declined consistently over the past 20 to 25 years. cost reductions have not kept pace with the general Federal renewable energy R&D, though inconsistently declines in cost seen in natural gas-fired generation.

funded, has been undertaken continuously during this These cost reductions, however, have put renewables in time. Although available data are insufficient to establish a better competitive position, especially given the sharp a quantifiable relationship between R&D funding and increases in natural gas prices in 2000.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 17

Technology, Manufacturing, and Market Trends in the U.S. and International Photovoltaics Industry by Peter Holihan prove to be cost effective when compared to the Introduction common distributed generation alternative, diesel gener-ators, which may be high priced because of the cost of In 1954, Bell Laboratories researchers announced the transporting fuel to remote regions.

development of a silicon solar cell with a 4.5-percent energy efficiency,1 sparking photovoltaic (PV) cell More recently, photovoltaic cell and module shipments development that has progressed from space applica-have grown on an international scale. Data for 1999 tions in the late 1950s to terrestrial applications today.

show 201 peak megawatts (MWp) of worldwide Over this period, research and development have shipments (Figure 2). Shipments from manufacturing resulted in lower prices for solar cells and modules capacity in the United States and Japan dominate the (Figure 1) and higher efficiency. U.S.-based photovoltaic market, with about 30 percent of shipments from the manufacturers development efforts have benefitted United States and about 40 percent of shipments from from a partnership with the Federal government. Similar Japan (Figure 3). This represents a marked change from partnerships at the State level have also been beneficial.

1995, when U.S.-based manufacturing capacity ac-Additionally, rising electricity prices and an increase in counted for 45 percent of world shipments, with Japan the cost of building new generation, transmission, and at 26 percent. The increase in Japanese market share is distribution capacity have had a positive impact on photovoltaic system economics and sales. Also during Figure 2. World Photovoltaic Shipments, this period, photovoltaic system sales have expanded as 1992-1999 a solution to remote distributed generation require-ments. In such markets, photovoltaic systems often 250 200 Figure 1. Decline in Photovoltaic Module Prices, Shipments (MWp)

Total 1975-1998 150 35 Module Prices (Dollars per Watt) 100 30 Japan Europe 25 50 United States 20 ROW 0

15 1992 1993 1994 1995 1996 1997 1998 1999 10 ROW = Rest of World.

5 MWp = Peak megawatts.

Note: The number of U.S. total PV shipments is a third quarter estimate 0 given by the companies, while in Figure 4 the number of U.S. total PV 197519771979198119831985198719891991199319951997 shipments is an end-of-year actual accounting.

Sources: 1993 through 1999 revised data from: Paul Maycock, PV News, Vol. 19, No. 3 (Warrenton, VA: PV Energy Systems, Inc., March Source: P. Maycock, The World Photovoltaic Market 1975-1998 2000). 1992 data from: P. Maycock, PV News, Vol. 18, No. 2 (Warrenton, VA: PV Energy Systems, Inc., August 1999), p. A-3. (Warrenton, VA: PV Energy Systems, Inc., February 1999).

1 M. Fitzgerald, The History of PV (Highlands Ranch, Colorado: Science Communications, Inc.). See website http://www.pvpower.com/pvhistory.html (December 1999).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 19

Figure 3. Photovoltaic Shipments Market Share, Solec International and United Solar Systems Cor-1992-1999 poration (USSC), which are joint ventures between 60 U.S. and Japanese corporations.3 50 (2) U.S. Shipments Dominated by Exports: Most PV United States 40 cell/module shipments from U.S. manufacturing facilities are exported (Figure 4). In 1998, U.S. manu-Percent Europe 30 facturing facilities exported 35 megawatts (MW) of 20 PV cells and modules, or 70 percent of total U.S.

Japan shipments,4 continuing a trend. Exports of PV 10 cells/modules manufactured in the United States Rest of World have exceeded 55 percent of total U.S. cell/module 0

1992 1993 1994 1995 1996 1997 1998 1999 shipments every year since 1987.

Sources: 1993 through 1999 revised data from: P. Maycock, PV Figure 4. U.S. Photovoltaic Cell and Module News, Vol. 19, No. 3 (Warrenton, VA: PV Energy Systems, Inc.,

Shipments, 1983-1999 March 2000). 1992 data from: P. Maycock, PV News, Vol. 18, No. 2 (Warrenton, VA: PV Energy Systems, Inc., February 1999). 80,000 Total 70,000 U.S.

U.S. Shipments (kWp)

Shipments 60,000 due to growth of the building-integrated photovoltaic 50,000 Exports (BIPV) applications market in Japan, which benefits 40,000 from Ministry of International Trade and Industry 30,000 Cells and 20,000 Modules (MITI) programs, subsidies, and net metering regu- Sold in U.S.

10,000 lations. 0 1983 1985 1987 1989 1991 1993 1995 1997 1999 The following analysis discusses the dynamics of the international photovoltaic (PV) market, addressing the kWp = Peak kilowatts.

activities of PV manufacturers and consumers that have Note: The number of U.S. total PV shipments is an end-of-year actual shaped the international market and their impact on the accounting while in Figure 2, the number of U.S. total PV shipments is a U.S. domestic PV industry. It will explain three major third quarter estimate given by the companies.

Source: Energy Information Administration, Form EIA-63B, Annual features of recent PV manufacturing and shipment Photovoltaic Module/Cell Manufacturers Survey.

history.

Three Major Features (3) Market Growth in Either Subsidized or High Value Markets: Countries experiencing growth in (1) Industry Consolidation: In the early 1990s, owner- photovoltaic shipments either have programs that ship of PV manufacturing capacity consolidated as heavily subsidize photovoltaic system purchases or Siemens purchased Arco Solar in March 1990 and market characteristics that lend value to photovoltaic ASE purchased Mobil Solar in July 1994. By 1997, electricity. Several subsidy programs exist to pro-about 80 percent of PV shipments from the United mote installation of distributed photovoltaic systems, States were attributable to manufacturing capacity including building-integrated photovoltaic systems.

owned by Siemens Solar and ASE Americas, both Value characteristics that enable photovoltaic German firms, and BP Solarex, a British firm.2 At the systems to compete include high electricity prices heart of these corporate entities are firms that were (e.g., high cost of generating fuel), or no electricity at originally founded as U.S. corporations: Arco Solar, all, and environmental concerns that entice con-Mobil Solar, and Solarex, respectively. About 11 sumers to pay a premium for electricity from photo-percent of PV shipments from the United States in voltaic or other renewable sources (i.e., through 1997 were attributable to manufacturing capacity at green pricing/marketing programs).

2 P. Maycock, Photovoltaic Technology, Performance, Cost and Market, V. 7 (Warrenton, VA: PV Energy Systems, August 1998), pp. 15-18.

3 Ibid.

4 Energy Information Administration, Form EIA-63B, Annual Photovoltaic Module/Cell Manufacturers Survey.

20 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

History Globalization of the Market The market for photovoltaic systems has developed in The U.S. photovoltaic industry is now in the third three stages, distinguished by the type of application market development stage, which began with increased and by the focus of State, Federal, and international sales to the international terrestrial electric power market development initiatives. market in the late 1980s. U.S. Energy Information Administration (EIA) data show that in 1985, the year in which Federal tax credits expired, U.S. exports of Space Program photovoltaic cells/modules represented approximately During the first stage (1950s through 1960s), PV 29 percent of total U.S. photovoltaic shipments. This development was motivated primarily by a need for percentage jumped to about 49 percent in 1986 and has electricity generation technology that would be suited remained at or above 55 percent since 1987, as photo-for the space program. In 1958, Vanguard I became the voltaic cells and modules manufactured in the United first PV-powered satellite. The 0.1 watt (W), approxi- States have been shipped internationally to serve mately 100 cm2 (square centimeters), silicon cell system terrestrial markets for PV in areas remote from a central powered a 5 milliwatt backup transmitter for 8 years.5 It station power grid (Table 1). Such areas face the high offered a relatively lightweight solution to power supply cost of diesel power generation, which make PV cost-for satellites and spacecraft. The single-crystal silicon effective. The 1990s have witnessed continued growth of photovoltaic cells deployed in space in the late 1950s had these markets aided, for example, by initiatives of donor cell efficiencies that ranged from 8 to 10 percent.6 By agencies (e.g., World Bank, United Nations Develop-1998, efficiencies of modules made from such cells had ment Programme, U.S. Agency for International increased to between 14 percent and 16 percent.7 Table 1. U.S. Photovoltaic Cell and Module Shipments, 1983-1998 Oil Price Pressures Total Shipments Exports Exports The second stage (1970s through mid-1980s) commenced Year (kWp) (kWp) (percent) with the Arab OPEC oil embargo of 1973, which resulted 1983 . . . . . . . . . . . 12,620 1,903 15.1 in a significant increase in oil prices. One response in the 1984 . . . . . . . . . . . 9,912 2,153 21.7 United States and other countries was to fund develop-1985 . . . . . . . . . . . 5,769 1,670 28.9 ment of renewable and energy-efficient technologies that 1986 . . . . . . . . . . . 6,333 3,109 49.1 would relieve dependence on fossil fuels. Federal and 1987 . . . . . . . . . . . 6,850 3,821 55.8 State tax credits for both residential and commercial 1988 . . . . . . . . . . . 9,676 5,358 55.4 customers subsidized expansion of terrestrial applica-1989 . . . . . . . . . . . 12,825 7,363 57.4 tions markets during this period. In addition, in 1978, 1990 . . . . . . . . . . . 13,837 7,544 54.5 the Public Utilities Regulatory Policy Act (PURPA) 1991 . . . . . . . . . . . 14,939 8,905 59.6 provided another market development support by 1992 . . . . . . . . . . . 15,583 9,823 63.0 guaranteeing qualifying facilities access to the elec-1993 . . . . . . . . . . . 20,951 14,814 70.7 tricity utility grid and requiring utilities to purchase the 1994 . . . . . . . . . . . 26,077 17,714 67.9 electricity. In California, the Standard Offer Number 4 1995 . . . . . . . . . . . 31,059 19,871 64.0 electricity purchase contract offered renewable electric 1996 . . . . . . . . . . . 35,464 22,448 63.3 qualifying facilities a very attractive purchase price, 1997 . . . . . . . . . . . 46,354 33,793 72.9 which was guaranteed for a period of 10 years.

1998 . . . . . . . . . . . 50,562 35,493 70.2 Qualifying facilities included renewable electric gen-erators, such as photovoltaic systems. By the late 1980s, kWp = Peak kilowatts.

Source: 1983-1997 data from Energy Information Administration, Federal tax credits had expired and other market Annual Energy Review 1998, DOE/EIA-0384(98) (Washington, DC, July mechanisms for new applicants were terminated. The 1999), Table 10.6; 1998 data from Energy Information Administration, result was a significant drop in the addition of new Form EIA-63B, Annual Photovoltaic Module/Cell Manufacturers photovoltaic electric generation capacity. Survey.

5 M. Fitzgerald, The History of PV (Highlands Ranch, Colorado: Science Communications, Inc.). See website http://www.pvpower.com/pvhistory.html (December 1999).

6 U.S. Department of Energy, History: PV Timeline, About Photovoltaics. See website http://www.eren.doe.gov/pv/history.html (May 2000).

7 P. Maycock, Photovoltaic Technology, Performance, and Cost 1995-2010 (Warrenton, VA: PV Energy Systems, Inc., January 2000), p. x.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 21

Development) and regional development banks. Addi- Figure 5. U.S. Shipments by Cell/Module Type, tionally, the 1990s have witnessed a growing interest in 1993-1998 renewables as a means to address environmental prob- 35,000 lems such as global warming. This interest is driving Single Crystal Silicon 30,000 programs such as the Million Solar Roofs Initiative and Shipments (kWp) 25,000 State initiatives to promote renewables in a deregulated electricity generation market. In addition, the govern- 20,000 Cast and Ribbon Silicon ments of Japan and Germany strongly support PV 15,000 programs. 10,000 5,000 Thin - Film Silicon Japan has a subsidy program goal of increasing PV 0 Concentrator Silicon demand by 400 MW per year through 2010 and 1993 1994 1995 1996 1997 1998 & Other Germany has a goal of 100 MW per year through 2005.

This increased demand is being met by domestic cell and kWp = Peak kilowatts.

module production and imports from the United States. Source: Energy Information Administration, Form EIA-63B, Annual Photovoltaic Module/Cell Manufacturers Survey.

Domestic and International Supply Figure 6. World Shipments by Module Type, 1998 U.S.-based manufacturers had an early market lead based on inventing and patenting PV technology. This 35,000 4 Rest of World (1) lead is being challenged by competition from countries 30,000 Europe (2) such as Japan and Germany. This international compe-Shipments (kWp) 25,000 Japan (3) tition, along with years of manufacturing experience and United States (4) 20,000 government research and development funding, has 15,000 1 produced gains in photovoltaic module energy efficiency 10,000 23 and cost reductions. New photovoltaic technologies that 5,000 show promise for further energy efficiency gains and cost reductions are starting to emerge. However, single 0 Single Crystal Amorphous Silicon Crystal Silicon Silicon on crystal silicon technology continues to dominate both Flat Plate Concentrators Low-Cost Substrate Polycrystal Silicon Cadmium Telluride Ribbon Silicon U.S. and some international cell and module shipments (Figures 5 and 6). U.S. photovoltaic cell and module kWp = Peak kilowatts.

shipments are shown in Figure 7. The following section Source: P. Maycock, The World Photovoltaic Market 1975-1998 reviews manufacturing and research trends. It also (Warrenton, VA: PV Energy Systems, Inc. , August 1999), p. 13.

discusses the impact that factors such as an educated labor force, Federal and State support of research and (manufacturing capacity in the United Kingdom (10 development (R&D), and availability of venture capital percent); France (5 percent); India (4 percent); Italy (3 have on growth of manufacturing capacity in a country. percent); and other countries (8 percent), including Spain, Taiwan, The Netherlands, and the Peoples U.S. and International Shipment and Republic of China.8 By 1999, Japanese manufacturers Capacity Trends (Kyocera, Sharp, and Sanyo) grew to lead world ship-ments, supported by government programs in Japan to From 1994 to 1999, annual worldwide shipments of use PV in building applications (Table 3). In 1999, the photovoltaic cells and modules almost tripled, growing combined market share of Kyocera, Sharp, and Sanyo from about 69 MW in 1994 to about 201 MW in 1999. rose to 37 percent, up from about 19 percent in 1994.

During this period, the combined market share of 10 companies grew from about 70 percent to 85 percent To meet growing demand, an estimated 250 MW of new Table 2). These companies have a global presence for manufacturing capacity for producing PV systems are manufacturing cells and modules (Table 3). During the currently planned for post-1998 installation (Table 4).9 1990s, photovoltaic manufacturing capacity expanded Most of the new capacity will be constructed in the beyond the United States, Japan, and Germany. In 1997, United States, Japan, and Germany. This new capacity worldwide cell and module shipments came from will include new thin film materials, such as copper 8

P. Maycock, Photovoltaic Technology, Performance, Cost and Market, V. 7 (Warrenton, VA: PV Energy Systems, August 1998), pp. 15-18.

9 P. Maycock, Photovoltaic Technology, Performance, and Cost 1995-2010 (Warrenton, VA: PV Energy Systems, Inc., January 2000), p. vii.

22 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Figure 7. U.S. Photovoltaic Cell and Module Locating Near End-Use Markets. Manufacturers benefit Shipments by End Use, 1994-1998 from the end-user and system installer feedback they 16,000 gain on product design and performance when selling photovoltaic systems locally. This can be integrated into improved system design, including balance of system Shipments (kWp) 12,000 2

improvements, which may result in cost reductions.

8,000 Manufacturers hope this will support increased sales by 3 providing end-users with desired features. Increased 4,000 6 sales help reduce the cost per kW price of a PV module 1 4 5

7 by spreading development and overhead costs over a 0

89 higher kW sales volume.

1994 1995 1996 1997 1998 Grid Interactive Power (1) Remote Power (2) Communications (3) The Spire Corporation/BP Solarex venture in Chicago is Transportation (4) Water Pumping (5) Consumer Goods (6)

Cells/Mods to OEM (7) Health (8) Other (9) an example of the trend toward locating manufacturing capacity close to end-users. PV modules will be manu-kWp: Peak kilowatts. factured in Chicago and the modules, incorporated into Note: Numbers above bars correspond to end use category. solar systems, will be marketed to residential and Source: Energy Information Administration, Form EIA-63B, Annual commercial customers in the Midwest. The Spire agree-Photovoltaic Module/Cell Manufacturers Survey.

ment with the City of Chicago and Commonwealth Edison (ComEd), the local utility, will provide $8 million indium diselenide, which Siemens Solar is producing of PV systems. Funding from ComEd shareholders currently at a market introduction level. Generally, it accounts for $6 million.10 The remaining $2 million will takes about 1 year to construct a 5 to 10 megawatt be funded from the City of Chicagos budget. Installing manufacturing plant to produce single, polycrystalline, PV systems on schools is a priority. ComEd has first and amorphous photovoltaic cells using existing manu- right of refusal on an additional $6 million of PV facturing technology. It takes up to an additional 6 systems. Manufacturing plants built to service such months to bring the new manufacturing facility up to markets are generally small, modular plants.

normal operation. Longer periods are expected initially for the new thin film photovoltaic technologies. If proximity to the end-use market is beneficial, then U.S.-based manufacturers, who export most of their Manufacturing Strategies product, may be at a disadvantage when it comes to (1) designing and manufacturing photovoltaic products to Photovoltaic manufacturers have developed the fol- meet most of their end-users needs and (2) benefitting lowing diverse strategies for competing in global from the lower system costs per kW that may result markets: from advances in product design and from increased Table 2. Global Corporate Market Share, 1994-1999 (Percent)

Supply Company 1994 1995 1996 1997 1998 1999 Siemens . . . . . . . . . . . . . . . . . . . . . . 19.4 22.2 19.2 17.5 12.9 12.0 Solarex . . . . . . . . . . . . . . . . . . . . . . . 10.8 12.2 12.2 11.8 10.3 8.9 BP Solar . . . . . . . . . . . . . . . . . . . . . . 8.8 9.3 9.5 9.0 8.7 7.2 Kyocera . . . . . . . . . . . . . . . . . . . . . . 7.9 7.9 10.3 12.2 15.8 15.1 Sanyo . . . . . . . . . . . . . . . . . . . . . . . . 7.9 6.6 5.2 3.7 4.1 6.5 ASE . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 4.8 3.4 4.8 4.5 5.5 Sharp . . . . . . . . . . . . . . . . . . . . . . . . 2.9 5.2 5.6 8.4 9.0 14.9 Photowatt . . . . . . . . . . . . . . . . . . . . . 2.6 2.6 2.8 4.5 7.7 5.0 Astropower . . . . . . . . . . . . . . . . . . . . 2.4 3.2 3.2 3.4 4.5 6.0 Isophoton . . . . . . . . . . . . . . . . . . . . . 2.2 1.9 1.7 2.1 2.7 4.0 Other Companies . . . . . . . . . . . . . . . 30.7 24.2 26.8 22.5 19.8 15.0 Source: Based on data in P. Maycock, PV News, Vol. 19, No. 3 (March 2000) and Paul Maycock, PV News, Vol. 19, No. 2 (February 2000).

10 Personal communication between Kent Whitfield (Spire Solar, Chicago) and William R. King (SAIC), March 8, 2000.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 23

Table 3. Module and Cell Shipments by Company, 1994-1999 (Megawatts)

Company (Manufacturing Location) 1994 1995 1996 1997 1998 1999 ASE (Germany) . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 1.7 -- 2.0 3.0 7.0 ASE (US) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.6 2.0 3.0 4.0 4.0 4.0 Astropower (US) . . . . . . . . . . . . . . . . . . . . . . . . . 1.7 2.5 2.85 4.3 7.0 12.0 BP Solar (Australia) . . . . . . . . . . . . . . . . . . . . . . -- -- -- -- 5.1 5.5 BP Solar (India) . . . . . . . . . . . . . . . . . . . . . . . . . -- -- -- -- 3.8 4.0 BP Solar (UK) . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 7.2 8.45 11.3 4.5 5.0 Isophoton (Spain) . . . . . . . . . . . . . . . . . . . . . . . . 1.5 1.5 1.5 2.7 4.2 8.1 Kyocera (Japan) . . . . . . . . . . . . . . . . . . . . . . . . . 5.5 6.1 9.1 15.4 24.5 30.3 Photowatt (France) . . . . . . . . . . . . . . . . . . . . . . . 1.8 2.05 2.5 5.7 12.0 10.0 Sanyo (Japan) . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5 5.1 4.6 4.7 6.3 13.0 Sharp (Japan) . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.0 4.0 5.0 10.6 14.0 30.0 Siemens (Germany) . . . . . . . . . . . . . . . . . . . . . . 0.5 0.2 0.05 0 0 2.0 Siemens (US) . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.0 17.0 17.0 22.0 20.0 22.2 Solarex (US) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5 9.5 10.8 14.8 15.9 18.0 Other Companies . . . . . . . . . . . . . . . . . . . . . . . . 21.3 18.8 23.8 28.3 30.6 30.2 World Total . . . . . . . . . . . . . . . . . . . . . . . . . . . 69.4 77.6 88.6 125.8 154.9 201.3 Sources: P. Maycock, PV News, Vol. 19, No. 3. (March 2000) for companies with Manufacturing Location listed as France, Germany, Spain, United Kingdom, United States or World Total. P. Maycock, PV News, Vol. 19, No. 2 (February 2000) for companies with Manufacturing Location listed as Australia, India, or Japan.

Table 4. Examples of Post-1998 New Manufacturing Capacity Systems for PV Manufacturing On-Line Country Company Technology Capacity Date (megawatts)

United States . . . . . . . . . . . . . . . . . . . . Siemens Solar Single crystal silicon 30 to 32 2000 United States . . . . . . . . . . . . . . . . . . . . Solarex Amorphous silicon 10 2000 United States . . . . . . . . . . . . . . . . . . . . ASE Americas Octagon EFG ribbon 20 2000 United States . . . . . . . . . . . . . . . . . . . . United Solar Systems Triple stack amorphous silicon 5 2000 United States . . . . . . . . . . . . . . . . . . . . Solar Cells Inc. Cadmium telluride 50 NA United States (California, Sacramento Municipal Utility District) . . . . . . . . . . . . . . . . . . . Energy Photovoltaics Amorphous silicon 5 2000 Germany (Saxony) . . . . . . . . . . . . . . . . Energy Photovoltaics Copper indium diselenide 5 2000 Germany (Gelsenkirchen) . . . . . . . . . . Shell Renewables Cast ingot polycrystalline silicon 25 2000 Japan . . . . . . . . . . . . . . . . . . . . . . . . . . Sanyo Amorphous Silicon on crystal silicon 10 2000 Japan . . . . . . . . . . . . . . . . . . . . . . . . . . Kyocera Cast ingot polycrystalline silicon 25 2000 Japan . . . . . . . . . . . . . . . . . . . . . . . . . . Sharp Crystalline silicon 30 NA Australia . . . . . . . . . . . . . . . . . . . . . . . . Solarex Cast ingot polycrystalline silicon 20 1999 Hungary . . . . . . . . . . . . . . . . . . . . . . . . Energy Photovoltaics Amorphous silicon 2.5 1998-99 Other (various countries, companies, and technologies) . . . . . . . 12 Total . . . . . . . . . . . . . . . . . . . . . . . . . 250 NA = Not available.

Source: P. Maycock, Photovoltaic Technology, Performance, and Cost 1995-2010 (Warrenton, VA: PV Energy Systems, Inc., January 2000),

pp. viii-x.

24 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

sales of systems that meet end-user design requirements. module manufacturing facility in the Sacramento area.

U.S.-based manufacturers compensate for their distance Volume purchase contracts provide a near-term way to from many end-use markets with a willingness to place attain lower photovoltaic module wholesale prices technically trained marketing representatives on site (Table 5).

around the world. They also engineer cells and modules for long-term trouble-free operation, covering them with Other manufacturers are taking the opposite approach, warranties of 20 to 25 years. increasing plant size substantially. Large plants (e.g.,

over 20 MW) would be built to achieve economies of Production in Japan and Germany is growing, despite scale that will reduce the production cost of photovoltaic high labor costs in both countries compared with the modules. For instance, as SMUDs residential grid-United States. High labor costs are offset, however, by connected demand grows enough to support large strong domestic markets, which enable emerging photo- capacity factories (40 MW and up), the wholesale price voltaic technology product development and cost reduc- for a thin film module is expected to fall to $1/W from tion efforts to benefit from end-user feedback. Strong current costs of $4.50/W.

domestic markets also enable Japan and Germany to export lower cost systems. Price decreases are expected to occur in steps. When a higher capacity factory starts to produce modules, Changing Plant Capacity. As mentioned above, there is module prices will remain high until demand increases a trend toward building smaller PV cell and module enough to take advantage of the economies of scale of plants closer to end-user markets. These plants can be the larger manufacturing plant. Breaking the $2/W expanded as demand increases. This strategy is moti- manufacturing cost barrier for photovoltaic modules vated by several factors. within the next 5 to 10 years will depend on high efficiency thin films (e.g., copper indium diselenide First, current PV manufacturing facilities have capacities (CIS), cadmium telluride (CdTe)) and next generation of 5 MW to 20 MW per year output, designed to support production volume manufacturing facilities.11 In Ger-local or regional demand, including utility-sponsored PV many, Shell Renewables is following a strategy to build programs. Second, transportation costs are reduced for large facilities. They opened a 25-MW facility to manufacturing plants situated locally relative to the end- manufacture cells in Gelsenkirchen, Germany in January user market. Third, the proximity of the plant to end 2000.12 users enables feedback from end users that is valuable in refining product design to meet end-user requirements Separation of Cell Manufacturing and Module Fabri-and in addressing any performance problems. cation Operations. Photovoltaic cell manufacturing processes require technically qualified labor to produce For example, Energy Photovoltaics, Inc. (EPV) in Prince- quality cells. Thus, cell manufacturing operations are ton, New Jersey, has a 5-year, 10 MW purchase contract located in countries where such labor is available (e.g.,

with the Sacramento Municipal Utility District (SMUD) United States, Japan, Germany). Assembly of cells into under which EPV will locate a 5 MW amorphous silicon modules does not require the same level of technical Table 5. Photovoltaic Module Costs (Wholesale)

Capacity of Module Resulting Wholesale Manufacturing Facility Module Price Year In Which Price Type of Sales Transaction (megawatts) (dollars per watt) Will Be Attainable High-volume purchase: 5-year contract to 5-20 1.50-2.50 Current (2000) purchase 10 megawatts of amorphous thin film modules Low-volume purchase: block purchases of PV 5-20 3-4 Current (2000) modules where the total purchase is in the hundreds of kilowatts range.

Thin film module 40-100 1 2005 Source: Personal communication between Don Osborn (SMUD) and William R. King (SAIC), March 3, 2000.

11 Personal communication between Tom Surek (NREL) and William R. King (SAIC), July 3, 2000.

12 R. Curry, Photovoltaic Insiders Report, Vol. XIX, No. 2 (February 2000), p. 6.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 25

expertise; therefore, manufacturers often ship cells to modules, and assembly of modules close to the instal-countries with end-use markets for assembly into lation site benefits from low labor rates at many inter-modules. The practice helps keep photovoltaic module national sites.

costs as low as possible because many countries where photovoltaic modules are deployed also have large pools In contrast to the United States, which in recent years of low-cost labor qualified for module assembly and exported up to 70 percent of domestically manufactured because cells are less expensive to ship than modules. cells and modules, Japan is more focused on proximity For example, in South Africa the strategy is to provide to the end-use customers. Japan exported only 35 low-cost module assembly to meet demand generated percent of domestic production in 1996 and 31 percent by the South African program to promote photovoltaics in 1997 (Table 7). Japan tends to export multicrystalline for rural electric applications. South Africa has two and amorphous silicon cells produced domestically and module assembly plants, several wholesalers, and about to import single crystal silicon cells.

40 distributor/systems integration companies.13 In India, the strategy is to use a technically adept and low-cost workforce to manufacture cells. BP Solar In-Country Corporate Presence. Photovoltaic manu-manufactures cells in India to take advantage of such facturers may establish a cell or module manufacturing labor rates and exports the cells to end-use markets.

presence in a country to obtain preferential treatment.

Indian manufacturers are also developing capacity. In For instance, a country may exempt the manufacturer Pune, India, Eco Solar Systems India is using a USAID with domestic operations from certain tariffs. Addi-conditional grant (3.5 million Rupees (Rs) or about tionally, countries such as Germany provide investment

$80,000) and a commercial loan (Rs 12.2 million, or about incentives for manufacturers to build plants. The com-

$280,000) to upgrade and modify a prototype photo-panies have employed these strategies in various ways.

voltaic cell manufacturing line.14, 15 This funding comes In the United States, photovoltaic manufacturing firms from USAID/India project reflows16 of Rs 261 million have formed alliances with utilities, as well as located (about $6 million), $4 million (from USAIDs technology the manufacturing plant near the end users. Examples development program of the mid-1980s), and Rs 660 include Tucson Electric/Global Solar (Arizona) and million (about $15 million) from Public Law 480 Title III GPU, Incorporated (New Jersey, Pennsylvania), a sub-funds for private sector projects.

sidiary of GPU International, Incorporated, a worldwide developer of independent powerplants, which operates GPU Solar as a joint venture with AstroPower, Inc., a Photovoltaic Technology Development photovoltaic module manufacturer. Programs Both government and corporate photovoltaic technology Export Strategies development programs are directing funding toward photovoltaic technology that can be produced more cost-U.S. companies have also used different export effectively. There are four or five independent tech-strategies. Photovoltaic cells and modules are shipped nology paths to low-cost PV, ranging from continuation worldwide from manufacturing facilities in the United of crystalline silicon technology to thin film alternatives.

States. From 1993 to 1998, Japan and Germany were among the top three recipients of these shipments (Table Lower Cost of Single Crystal Silicon 6). Often, cells are shipped to module assembly plants.

U.S. manufacturers prefer to produce cells in the United One approach is to continue trying to push the cost of States because of the availability of technically qualified single crystal silicon lower. However, cost reductions labor needed to produce quality photovoltaic cells. are hindered because feedstock for single crystal silicon Additionally, they benefit from the availability of quality cells is the waste silicon from the electronics industry.

materials from U.S. vendors, such as polymers, for Increasing demand for waste silicon is leading to manufacturing cells. Cells are less expensive to ship than shortages.

13 R. Karottki and D. Banks, PV Power and Profit? Electrifying Rural South Africa, Renewable Energy World, Vol. 3/No. 1 (January 2000),

p. 51.

14 U.S. Agency for International Development, USAID Activities in Indias Western States: Maharastra, Gujarat, and Madhya Pradesh. See website http://www.info.usaid.gov/india/ (March 2000), p. 8.

15 Indian rupees (Rs) are converted to equivalent U.S. dollars at a 1999 annual U.S. Federal Reserve rate of 43.13 Rs/US dollar, per Federal Reserve Statistical Release G.5A, January 3, 2000.

16 Reflows are revenues from projects that are paid back to the group that originally provided project funding. Then, the group can use the funds for other projects.

26 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 6. U.S. Exports by Country of Destination, 1993-1998 Cell and Module Shipments 1993 1994 1995 1996 1997 1998 Country Peak Percent Peak Percent Peak Percent Peak Percent Peak Percent Peak Percent Kilo- of Total Kilo- of Total Kilo- of Total Kilo- of Total Kilo- of Total Kilo- of Total watts Exports watts Exports watts Exports watts Exports watts Exports watts Exports Africa South Africa . . . . . . 399 2.7 791 4.5 1,294 6.5 541 2.4 939 2.8 2,608 7.3 Asia and the Middle East Japan . . . . . . . . . . . 1,440 9.7 2,857 16.1 3,616 18.2 2,889 12.9 8,056 23.8 9,586 27.0 Hong Kong . . . . . . . 1,567 10.6 1,175 6.6 1,125 5.7 701 3.1 1,423 4.2 1,323 3.7 India . . . . . . . . . . . . 94 0.6 806 4.6 2,398 12.1 755 3.4 285 0.8 435 1.2 Singapore . . . . . . . . 639 4.3 1,072 6.1 1,352 6.8 1,168 5.2 1,106 3.3 611 1.7 Australia 92 0.6 7 -- 16 0.1 387 1.7 61 0.2 119 0.3 Europe Germany . . . . . . . . . 4,972 33.6 4,641 26.2 3,755 18.9 8,150 36.3 11,162 33.0 9,727 27.4 Spain . . . . . . . . . . . -- -- 80 0.5 664 3.3 481 2.1 651 1.9 1,442 4.1 Switzerland . . . . . . . 4 0.0 138 0.8 799 4.0 177 0.8 31 0.1 1,220 3.4 North America Canada . . . . . . . . . . 819 5.5 1,043 5.9 503 2.5 793 3.5 775 2.3 633 1.8 Mexico . . . . . . . . . . 761 5.1 2,058 11.6 493 2.5 780 3.5 1,319 3.9 1,405 4.0 South America Brazil . . . . . . . . . . . 401 2.7 61 0.3 260 1.3 269 1.2 1,259 3.7 1,012 2.9 Total U.S. Exports 14,814 75.5 17,714 83.1 19,871 81.9 22,448 76.1 33,793 80.1 35,493 84.9 Notes: Total U.S. exports do not equal 100 percent because only those countries with the largest import markets are shown. U.S. totals include exports to other countries with non-sustainable export shipments.

Sources: Energy Information Administration, Renewable Energy Annual 1999, DOE/EIA-0603(99) (Washington, DC, March 2000), for years 1994 through 1998, and Solar Collector Manufacturing Activity 1993, DOE/EIA-0174(93) (Washington, DC, August 1994), for 1993.

Table 7. Japanese Photovoltaic Cell Exports and Imports, 1996 and 1997 (Kilowatts)

Fiscal Year 1996 Fiscal Year 1997 Cell Type Domestic Domestic Production Imports Exports Production Imports Exports Single Crystal Silicon . . . . . . 5,379.0 2,118.0 850.0 9,813.1 3,351.6 601.5 Multicrystalline Silicon . . . . . 9,535.0 680.0 4,005.0 17,525.0 1,964.0 5,111.0 Amorphous Silicon . . . . . . . . 5,574.0 14.0 1,725.0 5,936.3 7.6 3,817.0 Other . . . . . . . . . . . . . . . . . . 1,018.0 0.0 920.0 989.4 0.0 948.0 Total . . . . . . . . . . . . . . . . . 21,506.0 2,812.0 7,500.0 34,263.8 5,323.2 10,477.5 Source: O. Ikki, et al., The Current Status of Photovoltaic Dissemination Programme in Japan (Tokyo, Japan, September 1998), Table 8. Japan Photovoltaic Energy Association data.

In addition, the single crystal silicon cell is thick com- field. Thus, single crystal silicon modules have an pared to thin film alternatives. Use of more material advantage over other PV flat-plate module technologies increases product cost. On the positive side, single in applications where space is at a premium.

crystal silicon modules still command an energy con-version efficiency premium per square meter over Another approach is amorphous silicon, which may be alternative PV products. In addition, crystal silicon is a viewed as a transitional technology, since it has a lower known material with years of proven performance in the energy efficiency than alternatives and since amorphous Energy Information Administration/ Renewable Energy 2000: Issues and Trends 27

silicon modules must be aged prior to sale to ensure that Table 8. U.S. National Photovoltaics Program their energy efficiency remains stable. Copper indium Goals % 2000-2005 diselenide (CIS) is the leading material for amorphous 1995 2000 2005 silicon technology. The current problem with CIS is Module Efficiency (percent) . . 7-17 8-18 10-20 availability; Siemens Solar is manufacturing only pre-commercial market conditioning volumes.17 For the CIS System Cost (1999 dollars per watt) . . . . . 7-15 5-12 4-8 market to develop, purchases in the 100 kW range are needed. To support such purchases, production in the System Life (years) . . . . . . . . 10-20 > 20 > 25 one megawatt per year range is needed. U.S. Cumulative Sales 1,000-(megawatts) . . . . . . . . . . . . . 175 500 1,500 United States National Photovoltaics Program Note: Table shows range of module efficiencies for commercial flat-plate and concentrator modules.

The National Photovoltaics Program, funded by the U.S. Source: U.S. Department of Energy, Photovoltaics -- Energy for the Department of Energy, involves national laboratories, New Millennium: The National Photovoltaics Program Plan 2000-2004, DOE/GO-10099-940 (Washington, DC, January 2000), p. 9.

universities, and industry stakeholders in cooperative research and development of photovoltaic systems to attain higher module energy efficiencies, lower system Fundamental Research. Support industry and uni-costs, and longer system life. The long-term goal of the versity research to characterize cell materials and program is to make photovoltaic electricity available at devices; conduct research to understand defects in an operating cost of $0.06/kWh. Current program goals conventional crystalline silicon and thin film were established by U.S.-based photovoltaic industry materials; and develop techniques to reduce members to establish a roadmap for future industry efficiency-limiting defects in cell material; increase development (Table 8).18 The roadmaps goal for ship- the efficiency of multijunction concentrating cells ments is 25 percent annual growth in shipments from and large-area, monolithically interconnected thin manufacturing facilities based in the United States. This films.

growth rate would result in at least 6 gigawatts-peak (GWp) installed worldwide by 2020 from manufacturing Advanced Materials and Devices. Develop next capacity based in the United States, including 3.2 GWp generation thin film technologies through cost-of domestic installations.19 The 3.2 GWp target assumes shared efforts with industry and universities. This (1) a constant U.S. share of worldwide annual shipments effort includes support of first-time manufacturing of 40 percent and (2) installation of 30 percent of U.S. and scale-up of thin film amorphous silicon, CIS, shipments in the United States in the year 2000, CdTe, and thin silicon. Develop high efficiency increasing to 50 percent by 2020. The expected appli- crystalline silicon devices, emphasizing manu-cation mix for the 3.2 GWp is the following: facturing methods that reduce cost.

50 percent alternating current (AC) distributed gen- Technology Development. Develop manufacturing eration (remote, off-grid power for applications methods that result in lower cost, higher efficiency including cabins, village power, and communi- modules and in lower cost PV system components cations) (e.g., batteries and inverters). This effort has in-cluded the Photovoltaic Manufacturing Technology

33 percent direct current (DC) and AC value appli-(PVMaT) initiative, which addresses systems engi-cations (consumer products such as cell phones, neering and reliability issues through activities calculators, and camping equipment), and such as testing, developing domestic and inter-

17 percent AC grid (wholesale) generation (grid- national standards and codes, and analyzing connected systems including BIPV systems).20 factors affecting stability of encapsulated materials and performance of cells in modules. Technology For FY2000, the Federal PV research and development development also includes: (1) developing ad-program is funded at a level of $65.9 million (Table 9). vanced PV building concepts, tools, and modeling The program is divided into three areas: procedures; (2) motivating introduction of PV into 17 Personal communication between Don Osborn (SMUD) and William R. King (SAIC), March 3, 2000.

18 Proceedings from the U.S. Photovoltaics Industry PV Technology Roadmap Workshop (Energetics, Inc., ed.), National Center for Photovoltaics (Chicago, IL, September 1999).

19 Ibid., p. A4.

20 Ibid.

28 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 9. U.S. Federal Photovoltaic R&D Budget (Thousand Dollars)

FY 1999 FY 2000 FY 2001 Program Area Actual Appropriation Request Fundamental Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10,761 14,221 20,300 Advanced Materials and Devices . . . . . . . . . . . . . . . . . . . . . 25,836 27,000 27,000 Technology Development . . . . . . . . . . . . . . . . . . . . . . . . . . 33,964 24,691 34,700 Partners for Technology . . . . . . . . . . . . . . . . . . . . . . . . . . 3,800 500 2,000 Introduction Million Solar Roofs Initiative . . . . . . . . . . . . . 1,500 1,500 3,000 International Clean Energy Initiative . . . . . . . . . . . . . . . . . 0 0 4,000 Total Budget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70,561 65,912 82,000 Source: FY 2001 Congressional Budget.

building systems through cost-shared projects Figure 8. Federal Photovoltaic R&D Budgets, (Partnerships for Technology Introduction) and United States, Japan, and Germany, support of the Million Solar Roofs Initiative; and (3) 1981-1999 accelerating introduction of photovoltaic power as 140 a rural electrification option for developing Budget (Million Nominal U.S. Dollars) 120 countries by developing prototype systems, ad- United States vancing the concept of international equipment 100 standards, and developing tools for analyzing 80 distributed photovoltaic opportunities (Inter- Japan national Clean Energy Initiative). 60 40 The Partnerships for Technology Introduction, Million Germany Solar Roofs Initiative, and International Clean Energy 20 Initiative elements of the Technology Development 0

budget address market stimulation through funding of 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 cost-shared projects, prototype systems, and activities to promote formation of Million Solar Roofs partnerships.

Sources: United States -- FY 1999 budget from: FY 2001 Congressional None of the $1.5 million for the Million Solar Roofs Budget, Energy Supply, Solar and Renewable Resources Technologies, Initiative is an end-use incentive. Photovoltaic Systems, pp. 44-57. FY 1981 through FY 1998 budgets from: Historical data from National Photovoltaics Program records.

Germany (Federal Department of Education, Science, Research and Japanese and German National Photovoltaic Technology budget) and Japan (Sunshine PV Program budget) --

Development Programs Historical data from Jack L. Stone, National Renewable Energy Laboratory, National Center for Photovoltaics.

The Japanese and German development programs have provided competition for the United States over the years. For instance, during the 8-year period from 1981 to 1988, the German and Japanese Federal PV R&D U.S. and International Demand budgets increased, while the U.S. Federal budget fell (Figure 8). Recent funding data show the willingness of In 1999, worldwide shipments of PV cells and modules the Japanese government to spend relatively large totaled 201 MW,22 a 30-percent increase over 1998 world-amounts on direct market stimulation for end uses to wide shipments of 155 MW. U.S. manufacturers shipped promote their building photovoltaic program. They are just under 51 MWp of the total 1998 worldwide photo-funding market stimulation at a rate over four times that voltaic cell and module shipments. Factors motivating spent by either the United States or German programs photovoltaic sales included Federal government and (Table 10). Data indicate that the Japanese PV pro- State tax incentives, utility rebate programs, green motional budget rose steadily from $53 million in 1995 pricing programs, and donor agency programs to install to $132 million in 1998.21 photovoltaic systems in developing economies.

21 O. Ikki, K. Tomori, and T. Ohigashi, The Current Status of Photovoltaic Dissemination Programme in Japan (Tokyo, Japan: Resources Total System Co. Ltd., September 1998).

22 P. Maycock, PV News, Vol. 19, No. 3 (March 2000).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 29

Table 10. Research, Development, Demonstration, Building Integrated Photovoltaics (BIPV). These are PV and Market Stimulation Budget arrays mounted on building roofs or facades. For resi-Comparison, Fiscal Year 1998 dential buildings, analyses have assumed BIPV capa-(Million U.S. Dollars) cities of up to 4 kWp per residence. Systems may consist of conventional PV modules or PV shingles. This market United Program Area States Japan Germany segment includes hybrid power systems, combining diesel generator set, battery, and photovoltaic generation R&D . . . . . . . . . . . . . . . . . . . 64.7 56.1 38.3 capacity for off-grid remote cabins.

Demonstration . . . . . . . . . . . -- 21.4 --

Market Stimulation . . . . . . . .

132.5 18.4 Non-BIPV Electricity Generation (grid interactive and Total Budget . . . . . . . . . . . 64.7 210.0 56.7 remote). This includes distributed generation (e.g.,

standalone PV systems or hybrid systems including

-- = Not applicable.

In FY 1998, about $30 million of the U.S. $64.7 million R&D budget diesel generators, battery storage, and other renewable was spent on a combination of market stimulation-related activities technologies), water pumping and power for irrigation (market transformation, research initiatives, application-specific systems, and power for cathodic protection. The U.S.

research, and manufacturing process research). These expenditures Coast Guard has installed over 20,000 PV-powered navi-are included in the R&D budget for the United States because their objective is related more to R&D than to market stimulation. Market gational aids (e.g., warning buoys and shore markers) stimulation amounts shown for Japan and Germany reflect payment of since 1984.24 subsidies to reduce the cost of photovoltaic systems.

Sources: International Energy Agency, Trends in Photovoltaic Communications. PV systems provide power for remote Applications in Selected IEA Countries Between 1992 and 1998 (IEA-PVPS 1-07:1999) (Paris, France, October 1999), p. 6. R&D budgets for telecommunications repeaters, fiber-optic amplifiers, Japan (Sunshine PV Program budget) and Germany (Federal rural telephones, and highway call boxes. Photovoltaic Department of Education, Science, Research and Technology budget) modules provide power for remote data acquisition for from Jack L. Stone, National Renewable Energy Laboratory, National both land-based and offshore operations in the oil and Center for Photovoltaics.

gas industries.

Over 80 percent of 1998 shipments by U.S. manu- Transportation. Examples include power on boats, in facturers went to the following end uses: remote and cars, in recreational vehicles, and for transportation grid interactive electricity generation (45 percent); com- support systems such as message boards or warning munications (16 percent); transportation, e.g., power on signals on streets and highways.

boats, in cars, in recreational vehicles, and transpor-tation support systems (13 percent); and water pumping Consumer Electronics. A few examples are calculators; (9 percent). Key market niches encompassed by these watches; portable and landscaping lights; portable, light-end uses include building integrated photovoltaics weight PV modules for recreational use; and battery promoted by utilities and national climate change or chargers.

green power initiatives; other village, rural, or dis-tributed generation applications in both developed and Market growth in each segment is affected by counter-emerging economies; water pumping and irrigation vailing factors. The primary factor thwarting growth is systems, communications, and consumer products. the installed cost per kilowatt of the photovoltaic system, which often causes the cost of electricity (e.g.,

The following sections characterize these markets and cents per kilowatthour) from such systems to be higher discuss factors that influence demand. than the cost of electricity produced by fossil-fired or hydropower generation alternatives. National and inter-U.S. Demand national research efforts focus on ways to reduce the cost of photovoltaic systems.

The U.S. market is characterized by several niches that accounted for 15 MWp of cell and module shipments Cost-Effective Markets from manufacturing facilities in the United States in 1998. The domestic U.S. market includes the following Near-term market growth is occurring where the end-segments, defined by application:23 use is in a remote location or the measurable cost of 23 Kyocera discusses several of these applications on its website at http://www.kyocerasolar.com/industrial/ (March 2000).

24 National Renewable Energy Laboratory, Photovoltaics: Advancing Toward the Millennium, DOE/GO-10095-241 (Golden, CO, May 1996),

pp. 14-15.

30 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

electricity from alternative generation technologies is actual applications to ensure successful transition of the high enough for photovoltaic systems to be cost- product to the market place (e.g., PV:Bonus, TEAM-UP effective. U.S. distributors have identified markets (Technical Experience to Accelerate Markets in Utility where photovoltaic power is cost-effective now, without Photovoltaics), and PVUSA) (Table 11). Another variant subsidies. Examples include the following: (1) rural on this approach is public policy initiatives designed to telephones and highway call boxes, (2) remote data support photovoltaic sales with subsidies or appeals to acquisition for both land-based and offshore operations green consumers willing to pay a premium for clean in the oil and gas industries, (3) message boards or photovoltaic power.

warning signals on streets and highways, and (4) off-grid remote cabins, as part of a hybrid power system TEAM-UP Program including batteries.25 In the United States, the Federal TEAM-UP program, a The current installed cost of photovoltaic systems ranges government-industry cost-shared program managed by from $0.20 to $0.50 per kilowatthour, depending on the Utility Photovoltaic Group (UPVG), is an example of factors such as the volume purchased and the level of market conditioning support. TEAM-UP is not a large solar insolation. Therefore, the electric price of the next program; the first three rounds of competitively best alternative must be no lower than this range for PV awarded installations will total more than 7.5 MW in 31 to be cost-effective. High electric prices tend to be found states.27 For grid-connected systems, the subsidies under where there is no cost-effective access to the electric grid this program are negotiated depending upon program (e.g., remote applications markets, including distributed size and have averaged about 20 percent of total system generation, telecommunications, navigational aids, and installed cost.28 In the United States, utility programs to cathodic protection). Diesel generator sets are the subsidize PV system deployment are motivated by alternative to photovoltaic electricity in some of these individual states electric utility restructuring and dereg-markets. In remote applications, diesel generator sets ulation activities.

may be at a disadvantage to PV because these systems bear high costs of hauling fuel to the site, storing fuel, For example, in California, revenues from a public bene-and maintaining equipment. fit charge are used to fund renewable energy projects, including photovoltaic projects. A public benefit charge In the longer term, it will take a combination of whole- is an amount embedded in the electricity rate paid by sale system price below $3.00/W and large volume consumers to cover public goods programs that would dealers for PV to be cost-effective in the residential grid- not otherwise be funded by deregulated utilities. The connected market. PV installed system costs must fall to state, through the California Energy Commission, a range where they are competitive with current retail manages activities in investor-owned utility service electric rates of $0.08 to $0.12/kWh in the residential territories; municipal utilities such as the Sacramento market and $0.06 to $0.07/kWh in the commercial Municipal Utility District (SMUD) and the Los Angeles market.26 Department of Water and Power (LADWP) manage their own photovoltaic programs. Other states are Photovoltaic Green Power considering renewable energy portfolio legislation to require a certain percentage of generation from renew-U.S. Federal programs such as Million Solar Roofs and able resources.

programs in states such as California emphasize the advantage of photovoltaic power as a clean sustainable Buy-Down Programs power source, one that promotes lower environmental emissions. Programs are a mix of those that promote California and Maryland are examples of states with growth of photovoltaic power market share (e.g., buy-down programs for photovoltaic systems. The Cali-Million Solar Roofs, PV Pioneer programs, Solar Power fornia Energy Commissions (CECs) Emerging Renew-Hosting and Ownership programs, and Emerging ables Buy-Down Program offers cash rebates for systems Renewables Buy-Down Program) and those that support purchased from eligible providers listed on the pro-PV product development, testing, and operation in grams web site. Eligible technologies are photovoltaic 25 For example, Kyocera discusses such applications on its website at http://www.kyocerasolar.com/industrial/ (March 2000).

26 Personal communication between Don Osborn (Sacramento Municipal Utility District) and William R. King (SAIC) (March 3, 2000).

27 Utility Photovoltaic Group, What Is TEAM-UP?. See website http://www.ttcorp.com/upvg/team_mn.htm (March 2000).

28 Ibid.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 31

32 Table 11. Examples of Photovoltaic Technology Market Development Initiatives Inception Date %

Initiative Sponsor(s) Completion Date Objective Strategy Results U.S. Department of Energy 1993 -- ongoing Develop prototype PV Innovative product designs for building Developed products including a (DOE) products to replace applications. Fund product development. flexible solar shingle and alternating PV:Bonus conventional windows, current (AC) PV modules.

skylights, and walls.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends Utility industry-DOE cost- 1994 -- 2000 Demonstrate and validate PV Market conditioning through demonstration. 4.5 MW installed under Round One sharing partnership system hardware installations Competitively procure, install, and and Two solicitations. Total 7.4 MW managed by Utility for various utility/energy demonstrate 50 MW of PV systems. installed capacity (2300 PV TEAM-UPb Photovoltaic Group service provider applications. Awards made to ventures that will build a systems) by October 2000.

(UPVG) Build owner and customer PV system and sell to end-users.

confidence in systems.

U.S. Department of Energy June 26, 1997 -- Reduce greenhouse gases Encourage installation of one million solar Motivating formation of partnerships 2010 and other emissions. Create energy systems on U.S. rooftops by 2010. committed to installed PV on Million Solar high-tech jobs. Keep U.S. PV rooftops. Examples of partnership Roofsc industry competitive. activities include the SMUD, LADWP, and Spire Solar Chicago PV programs.d Co-sponsors include 1986 -- 2000 Enable utilities to evaluate Market conditioning through demonstration. In 1998, monitoring activities various State and Federal grid-connected PV system Evaluate various PV technologies within a covered 26 PV systems with agencies and various performance, reliability, and systems context using three grid- combined 2.3 MW capacity in 10 PVUSAe electric utilities.f cost and to assess system connected pilot test stations in different U.S. locations.

operations & maintenance parts of the United States.

(O&M) requirements.

Sacramento Municipal 1993 -- on-going Reduce price of PV generated Mass purchase. SMUD purchases and As of year end 1999, about 550 Utility District (SMUD) power. installs PV system on volunteering residential and commercial rooftop customers roof and operates the system PV systems (total capacity about 2 for 10 years with all the solar electricity MW).h About 35 church and school sold to the customer at regular SMUD rooftop systems and parking lot PV Pioneer Ig rates. Volunteers pay an additional $4.00 a systems (1.5 MW total capacity) month, which is decreased if rates under the Neighborhood PV increase. Pioneers version of PV Pioneer I.i System costs have declined from

$7.70/W to less than $4.25/W.

Sacramento Municipal 1999 -- on-going Reduce price of PV generated Subsidized purchase. SMUD enables 250 signed letters of commitment Utility District power. customers to purchase a rooftop PV with virtually no marketing. First system at a substantial discount and system installed April 1999. By year PV Pioneer IIj receive credit on their electric bill for the end 1999, first 50 systems installed energy the system produces under a net or scheduled for installation.k metering arrangement.

See notes at end of table.

Table 11. Examples of Photovoltaic Technology Market Development Initiatives (Continued)

Inception Date %

Initiative Sponsor(s) Completion Date Objective Strategy Results Los Angeles Department of May 1998 100,000 systems on Mass purchase. LADWP installs and 15 customers (40 kW total capacity) to Solar Power Water and Power (LADWP) -- on-going residential rooftops in LA City owns the PV system on the customer date. Includes 14 customers with 2.5 kW Hostingl by the year 2010 volunteers roof. systems and one 5 kW system.m Energy Information Administration/ Renewable Energy 2000: Issues and Trends Los Angeles Department of December 31, 1998 100,000 systems on Subsidized purchase. Customer 35 customers (100 kW total capacity) to Solar Power Water and Power -- on-going residential rooftops in LA City owns the PV system on his/her roof date.o Ownershipn by the year 2010 and is billed by LADWP for electricity on a net metering basis.

California Energy March 20, 1998 -- Increase use of renewable Subsidized purchase. Provides cash As of March 14, 2000, 622 reservation Commission on-going electricity. Over 30 MW of rebates of up to $3,000/kW, or 50 requests received, including 471 (CEC) power possible under the percent of the system price, completed or approved projects.

Emerging program. Most assumed to whichever is less. Completed or approved projects include Renewables be PV; but PV, solar thermal, 2.9 MW of power from 428 PV systems, Buy-Down fuel cell, and small wind 41 wind systems, and 2 fuel cell systems Programp,q systems (no larger than 10 with 400 kW combined capacity. $4.2 kW capacity) are eligible. million paid for 282 completed projects;

$3.8 million encumbered for 189 approved projects.

a U.S. Department of Energy, Photovoltaic Energy Program Overview: Fiscal Year 1998, DOE/GO-10099-737 (Washington, DC, March 1999).

b Utility Photovoltaic Group, 4.5 Megawatts of PV and Counting. . . :Technical and Business Experience of TEAM-UP Program Partnerships (Washington, DC, November 1999).

c U.S. Department of Energy, http://www.eren.doe.gov/millionroofs/ (December 1999).

d A tally of partnerships may be found at Million Solar Roofs, Current State and Community Partnerships, http://www.eren.doe.gov/millionroofs/tally.html (May 2000).

e Photovoltaics for Utility System Applications, http://www.pvusa.com/index.html (December 1999), and SMUD, 1998 PVUSA Progress Report, 1999, (Sacramento, CA, 1999), pp. 1, 3, and 6.

f Co-sponsors include DOE; Electric Power Research Institute; Department of Defense; various utilities and national labs; New York State Energy Research and Development Authority; City of Austin, Texas; and the Solar Energy Industries Association. PVUSA is managed by the California Energy Commission and the Sacramento Municipal Utility District. See website http://www.pvusa.com (December 1999).

g Sacramento Municipal Utility District, http://www.smud.org/home/pv_pioneer/index.html (December 1999).

h Donald Osborn, Sustained Orderly Development and Commercialization of Grid-Connected Photovoltaics: SMUD as a Case Example, pre-print, Advances in Solar Energy, Vol. 14, 2000 American Solar Energy Society (Boulder, CO, May 2000), p. 8.

I Donald Osborn, Sustained Orderly Development and Commercialization of Grid-Connected Photovoltaics: SMUD as a Case Example, pre-print, Advances in Solar Energy, Vol. 14, 2000 American Solar Energy Society (Boulder, CO, May 2000), p. 11.

j Sacramento Municipal Utility District, http://www.smud.org/home/pv_pioneer/index.html (December 1999).

k Donald Osborn, Sustained Orderly Development and Commercialization of Grid-Connected Photovoltaics: SMUD as a Case Example, pre-print, Advances in Solar Energy, Vol. 14, 2000 American Solar Energy Society (Boulder, CO, May 2000), p. 11.

l Los Angeles Department of Water and Power, http://www.ladwp.com/whatnew/solaroof/solaroof.htm (December 1999).

m Personal communication between Robert McKinney (LADWP Solar Power Program Manager) and William R. King (SAIC), May 24, 2000.

n Los Angeles Department of Water and Power, http://www.ladwp.com/whatnew/solaroof/solaroof.htm (December 1999).

o Personal communication between Robert McKinney (LADWP Solar Power Program Manager) and William R. King (SAIC), May 24, 2000.

p Information from Sandy Miller, Manager, California Energy Commission Emerging Renewables Buy-Down Program (May 22, 2000).

q California Energy Commission, Emerging Renewables Buy-Down Program, http://www.energy.ca.gov/greengrid/index.html (March 8, 2000).

33

systems, wind turbines with maximum output of 10 kW, Under the PV Pioneer II program, the end user pur-fuel cells, and solar thermal systems. This program is chases a grid-connected BIPV system at a discounted per only available to customers of the following investor- kilowatt rate. The end user uses electricity from the owned utilities: Pacific Gas & Electric (PG&E), San BIPV system under a net metering arrangement with Diego Gas & Electric (SDG&E), Southern California SMUD. SMUD and LADWP bill customers who own Edison (SCE), and Bear Valley Electric Company. The their BIPV systems on a net metering basis, so the value Maryland Solar Roofs Program provides $2.00/W cost- of electricity equals the price the customer would pay sharing in the year 2000 for residential photovoltaic for electricity purchased from the utility.

systems. The Maryland program estimates that this would cover 40 percent of installed system cost. The Solar Power Hosting and Ownership Programs cost-share amount declines in subsequent years.29 LADWPs PV programs, the Solar Power Hosting Municipal Utility Programs Program and the Solar Power Ownership Program, are similar to SMUDs.30 Under the Hosting Program, SMUD and LADWP, both municipal utilities, have LADWP installs and maintains the BIPV system; the end photovoltaic system deployment programs because they user pays nothing. Under the Ownership Program, the get to spend their public benefit program funds. Both end user installs and owns a BIPV system and uses programs are similar. In California, utilities embed a electricity from the system under a net metering public benefit charge in the rate charged for electricity. arrangement with LADWP. The end user does not This charge funds programs, such as renewable tech- purchase the BIPV system through LADWP; LADWP nology market development, that would not be pursued just subsidizes the purchase and facilitates system normally in a deregulated utility environment. Munici- interconnection with the grid.

pal utilities are allowed to keep the revenue generated by this charge to spend on public benefit programs, such International Demand as renewable technology deployment programs, within their service territory. In contrast, public benefit pro- Shipments of photovoltaic cells and modules from gram revenue generated by shareholder-owned utilities manufacturing facilities in the United States and other in California is collected in a central pool. These funds countries supply growing international demand.

are available for CEC-sponsored energy projects, such as Growing markets include those where factors such as photovoltaic system buy-downs. high electricity prices and subsidies or other incentives improve the cost-effectiveness of PV systems. In several PV Pioneer I and II countries, average residential electricity prices are high compared to the United States (Table 12). These prices SMUD runs the PV Pioneer I and PV Pioneer II pro- represent those for grid-connected customers. The grams. Under PV Pioneer I, the end user allows SMUD following sections provide examples of these and other to install a grid-connected BIPV system. The end user factors that are motivating demand.

pays $4 per month to SMUD. This fee is decreased if the electricity rate increases and is eliminated if the rate Japan increases at least 15 percent. SMUD agrees to install and operate the system for 10 years, after which SMUD may The Ministry of International Trade and Industry (MITI)

(1) sell the system to the customer at an attractive rate promotes photovoltaic sales primarily through programs and convert the customer to the PV Pioneer II program; that promote growth of the residential BIPV market. The (2) ask for an extension of the agreement, perhaps at ministrys targets for installed PV capacity across all reduced rates; or (3) remove the system and repair the applications are 400 MW by the year 2000, and 5,000 roof. MW by the year 2010.31 Much of this capacity will 29 C. Cook, The Maryland Solar Roofs Program: State and Industry Partnership for PV Residential Commercial Viability Using the State Procurement Process, Second World Conference on Photovoltaic Solar Energy Conversion, Vienna, Austria. See website http://www.energy.state.md.us/paper.htm (July 1998).

30 Los Angeles Department of Water and Power, Solar Electricity Rooftop Program. See website http://www.ladwp.com/whatnew/

solaroof/solaroof.htm, March 2000. Personal contact between Robert McKinney (LADWP Program Manager) and William R. King (SAIC),

March 2000.

31 O. Ikki, K. Tomori, and T. Ohigashi, The Current Status of Photovoltaic Programme in Japan (Tokyo, Japan: Resources Total System Co.,

Ltd., September 1998), Table 3.

34 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 12. Examples of Countries with High implemented by Japanese companies, including capacity Residential Electricity Prices Relative to additions that result when these companies purchase the United States, 1997 companies previously incorporated in other countries.

Electricity Price Country (dollar per kilowatthour) The MITI BIPV program, through its New Energy United States . . . . . . . . . . . . . . . 0.085 Foundation, plans to equip 70,000 homes with 3 kW systems by 2000 (210 MW at 3 kW/system) and install Other OECD BIPV on half of new homes by 2010.32 As of March 31, Countries 1999, BIPV systems were installed on 28,000 homes (84 Japan . . . . . . . . . . . . . . . . . . . . . 0.207 MW at 3 kW/system). MITI motivates demand for the Denmark . . . . . . . . . . . . . . . . . . . 0.195 BIPV systems through an incentive program that pays Austria . . . . . . . . . . . . . . . . . . . . 0.169 half the cost difference between installed system cost per Belgium . . . . . . . . . . . . . . . . . . . . 0.168 kW and $3,100/kW for BIPV systems up to 10 kW Spain . . . . . . . . . . . . . . . . . . . . . 0.163 Germany . . . . . . . . . . . . . . . . . . . 0.161 capacity. The program requires that electric utilities Italy . . . . . . . . . . . . . . . . . . . . . . . 0.159 purchase excess electricity from residences at the going Portugal . . . . . . . . . . . . . . . . . . . 0.156 residential rate through net metering.

Switzerland . . . . . . . . . . . . . . . . . 0.136 France . . . . . . . . . . . . . . . . . . . . 0.134 Germany Ireland . . . . . . . . . . . . . . . . . . . . . 0.131 Netherlands . . . . . . . . . . . . . . . . 0.130 By year-end 1997, Germany had close to 10,000 grid-United Kingdom . . . . . . . . . . . . . 0.125 connected PV systems (34 MW total capacity).33 Cata-Luxembourg . . . . . . . . . . . . . . . . 0.124 lysts for PV system market growth included financial incentives (Federal and State), rate-base incentives, and Non-OECD Countries green pricing. Incentives contributed to 45 percent of Grenada . . . . . . . . . . . . . . . . . . . 0.193 1997 PV systems.

Suriname . . . . . . . . . . . . . . . . . . 0.171 Barbados . . . . . . . . . . . . . . . . . . 0.167 As of 1998, 3,500 residences had BIPV systems. The Uruguay . . . . . . . . . . . . . . . . . . . 0.157 economics of these installations benefitted from govern-Argentina . . . . . . . . . . . . . . . . . . 0.139 ment subsidies and a high price paid by the utility for Peru . . . . . . . . . . . . . . . . . . . . . . 0.138 excess electricity produced by each system.34 Jamaica . . . . . . . . . . . . . . . . . . . 0.135 Chile . . . . . . . . . . . . . . . . . . . . . . 0.121 In 1999, the German government initiated a 100,000 Panama . . . . . . . . . . . . . . . . . . . 0.121 Roofs Program with the goal of installing 300 to 500 Source: Energy Information Administration, International MW of BIPV systems over the period 1999 through Electricity Prices for Households, http://www.eia.doe.gov/emeu/

2005.35 Program cost is expected to be about $600 iea/elecprih.html (October 20, 2000).

million.36 In 1999, installation of 6,000 3-5 kW arrays was expected;37 actual home installations were about 35 be in BIPV systems. Assuming 400 MW installed by percent less&3,834 grid-connected arrays (10.1 MW) 2000, the annual demand from 2001 through 2010 would from program initiation through February 2000.38 be 460 MW per year. This amount helps explain the Planned annual installations will increase to more than current PV manufacturing capacity additions being 32,000 in the programs final year.39 The program offers 32 M. Dunn, U.S. Department of Energy, Office of Intelligence, International Solar Cells Outlook 99, NIS-8(U)99-102 (Washington, DC, April 1999), p. 11.

33 Dr. H. Gabler and V.U. Hoffman (Fraunhofer-Institute for Solar Energy Systems ISE), Dr. Klaus Heidler (The Solar Consultancy),

Financing Germanys PV expansion, The Sustainable Energy Industry Journal, Issue 8 (Vol. 3, No. 2) (1998), p. 16.

34 M. Dunn, U.S. Department of Energy, Office of Intelligence, International Solar Cells Outlook 99,NIS-8(U)99-102 (Washington, DC, April 1999), p. 10.

35 Ibid.

36 International Energy Agency, Trends in Photovoltaic Applications in Selected IEA Countries Between 1992 and 1998, IEA-PVPS 1-07:1999, (Paris, France, October 1999), p. 12.

37 M. Dunn, U.S. Department of Energy, Office of Intelligence, International Solar Cells Outlook 99, NIS-8(U)99-102 (Washington, DC, April 1999), p. 10.

38 P. Maycock, 100,000 Roofs Serves 3834 Roofs, PV News, Vol. 19, No. 4 (April 2000), p. 3.

39 M. Dunn, U.S. Department of Energy, Office of Intelligence, International Solar Cells Outlook 99, NIS-8(U)99-102 (Washington, DC, April 1999), p. 10.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 35

a 10-year low interest loan with repayment starting in ecotariff plants have been built, including 22 photo-the third year.40 voltaic plants. RWE Energy takes credit for the CO2 reduction.

The new Renewable Energy Law,41 passed February 25, 2000, is already prompting interest on the part of Other European Activity companies involved in the photovoltaics industry. It guarantees fixed tariffs for green electricity to the grid Switzerland. Up to 25 percent of the installed cost of a and provides a national incentive of 0.99 deutsche marks PV system is subsidized. More than 170 public schools (DM) per kWh ($0.51 per kWh) over 20 years for elec- have rooftop PV systems.45 Other activities include over tricity from renewable sources, including photovoltaics. 1,000 grid-connected 3 kW residential systems, 500 kW This incentive may be combined with zero interest loans on Mont Soliel, and 600 kW on highway sound available under the 100,000 Roofs program. barriers.46 The Swiss government has promoted photo-voltaic systems under its Energy 2000 project.

RWE Energie.42 RWE Energie, the largest energy service company in Germany, has built two PV power plants, The Netherlands. In 1997, the government initiated a each 350 kW, one on the Moselle River and one at Lake program to increase use of renewable energy. Goals for Neurath in the Rhenish lignite-mining area. The com- photovoltaic systems are 10 MW by 2000 and 250 MW pany operates a 1 MW plant jointly with Spanish by 2010.47 partners, near Toledo, Spain,43 The plant is one of the largest in Europe. In a 500-household PV complex, 50 percent of the electricity (1.3 MW/year) will be provided by 12,000 In mid 1996, RWE Energy initiated two consumer square meters (m2) of PV panels (20 m2 per house). The incentive programs, KesS SOLAR and Ecotariff, to complex is being developed by REMU, a Dutch electric promote renewable energy, including photovoltaics:44 power company, and is sponsored by the European Union and Dutch government. It includes both resi-KesS SOLAR. The consumer receives DM 2,000 (about dential and commercial installations. Residents pay 50

$1,030) for purchasing a residential solar system (solar percent of the panel cost. Generated electricity belongs collectors, PV, or electric heat pumps). RWE Energy has to the homeowner, who is compensated using net paid DM 20 million (about $10.3 million) under this metering. The project is motivated by global warming program. worries; the elevation of much of the countrys land is below sea level.48 Ecotariff (green pricing). The consumer purchases at least 100 kWh per year at a premium of 20 pfennigs/

kWh (about $0.10/kWh) over the normal retail price. India RWE Energie matches the contribution. Amounts are Through Winrock Internationals Renewable Project used to build new plants equivalent to the green Support Office (REPSO) in India, USAID supports PV kilowatthours. RWE Energy made DM 20 million (about projects including the following:49

$10.3 million) available under this scheme. Fifteen thousand customers have used this plan, purchasing 2.6 SELCO Photovoltaic Electrification Private Limited million kWh of renewable electricity. Twenty-four (SELCO), Bangalore. Under a conditional grant of Rs. 5 40 International Energy Agency, Trends in Photovoltaic Applications in Selected IEA Countries Between 1992 and 1998, IEA-PVPS 1-07:1999, (Paris, France, October 1999), p. 12.

41 P. Maycock, New Renewable Energy Law to Trigger Solar Boom in Germany, PV News, Vol. 19, No. 4 (April 2000), p. 3.

42 In this section, German deutsche marks (DM) are converted to equivalent U.S. dollars at a rate of 1.94 DM/US dollar.

43 Dr. Munch, A Partnership with Our Customers to Promote Renewable Energy, The Sustainable Energy Industry Journal, Issue 8 (Vol.

3, No. 2) (1998), p. 27.

44 Ibid.

45 M. Dunn, U.S. Department of Energy, Office of Intelligence, International Solar Cells Outlook 99, NIS-8(U)99-102 (Washington, DC, April 1999), pp. 13-14.

46 P. Maycock, The World Photovoltaic Market: 1975-1998 (Warrenton, VA: PV Energy Systems, Inc., August 1999), p. 40.

47 M. Dunn, U.S. Department of Energy, Office of Intelligence, International Solar Cells Outlook 99, NIS-8(U)99-102 (Washington, DC, April 1999), p. 13.

48 R. Curry, Photovoltaic Insiders Report, Vol. XIX, No. 2 (February 2000), p. 2.

49 U.S. Agency for International Development, USAID Activities in Indias Southern States: Tamil Nadu, Karnataka, Kerala, and Andhra Pradesh. See website http://www.info.usaid.gov/india/states/south.htm (March 2000), pp. 5-6.

36 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

million ($140,000), SELCO will promote commerciali- agency. The total cost of Rs 32.14 million ($750,000) zation of residential PV lighting systems in Andhra is financed by the Ministry of Non-Conventional Pradesh and Karnataka. Over 2,500 systems have been Energy Sources (MNES) and Indian Renewable sold. SELCO has started making repayments to Winrock Energy Development Agency Limited (IREDA) as reflows,50 which can be used for other renewable and includes World Bank funding.52 energy activities.

Fifteen PV streetlights installed in Sanjay Gandhi Polyene Film Industries (PFI), Chennai. Under a Biological Park. The parks medical clinic also has conditional grant of Rs. 4.3 million ($100,000), PFI will a PV system that ensures uninterruptible elec-install 100 PV water pumping systems for irrigation. The tricity.53 systems will be used by poor farmers and tribal people in District Nellore, Andhra Pradesh and Tamil Nadu.

The grant will be repaid by PFI up to 1.4 times in semi- Peoples Republic of China annual installments starting 2 years from the date of the The World Bank has signed a renewable energy develop-conditional grant. The systems use 800 Wp DC motors ment agreement for the Peoples Republic of China.

powered by multicrystalline thin-film Solarex PV Included in the agreement is a $15 million Global modules. Environment Facility (GEF) grant to install 10 to 12 MW Center for Technology Development NGO Resource of photovoltaics in 400,000 households.54 The total $444 Center (CTD-RC), Bangalore. Under a conditional grant million renewable energy project also supports instal-of Rs. 5.6 million (approximately $130,000), CTD-RC, in lation of 190 MW wind turbines (Table 13).55 collaboration with SELCO, will commercialize resi-dential PV lighting systems in rural areas of Karnataka. The GEF grant will fund a $1.50/Wp installed system Cooperative banks will act as financial intermediaries. payment to Chinese PV system companies for systems The end-user will pay 20 percent of the total installed 10 Wp or greater in capacity. The $15 million grant system cost up front. The remaining 80 percent of would, therefore, cover 10 MW of installed PV capacity system cost will be financed by a load to the end-user meeting the 10 Wp minimum system capacity. This guaranteed by CTD-RC, and repaid in convenient grant is given to these companies to (1) improve product installments. quality, (2) improve warranties and service, (3) strengthen business capabilities and marketing efforts.56 Examples of other PV projects in India include the following: Additionally, $7 million as a GEF grant and $4 million

A 50 kW PV power plant commissioned on Kad- from other sources, for $11 million total, are allocated mat Island in the Arabian Sea, in Lakshwadeep, for a PV market development program (awareness pro-India. The power plant serves the Water Sports grams, demos, market development assistance) and for Institute and surrounding cottages and is the first institutional strengthening (PV quality assurance and PV facility to serve sporting activity. On the Bitra project management capabilities).57 and Bangaran Islands in Lakshwadeep, 25 kW and 10 kW PV power plants, respectively, meet resi- The following photovoltaic system market development dential lighting loads.51 Examples of other PV barriers have been identified for the Peoples Republic of projects in India include the following: China:58

Two grid-connected PV plants approved for the High cost of PV systems. A 20 Wp system costs about State of Punjab by the Punjab Energy Development $200, including value-added tax (VAT), making these 50 Reflows are revenues from projects that are paid back to the group that originally provided project funding. Then, the group can use the funds for other projects.

51 P. D. Maycock, Unique Solar Plant Commissioned in Lakshwadeep, PV News, Vol. 19, No. 3 (March 2000), p. 6.

52 P. D. Maycock, Ministry Approves 2 Grid Interactive PV Units, PV News, Vol. 19, No. 3 (March 2000), p. 6.

53 P. D. Maycock, Biological Park Gets Solar PV for New Years Day, PV News, Vol. 19, No. 3 (March 2000), pp. 6-7.

54 Personal communication between Susan Bogach (The World Bank) and Peter Holihan (DOE/EIA) (March 2000).

55 The World Bank, Project Appraisal Document on a Proposed Loan in the Amount of US$100 million and a Proposed GEF Grant of US$35 million equivalent to the Peoples Republic of China for a Renewable Energy Development Project, Report No. 18479-CHA (Washington, DC, May 5, 1999),

p. 6.

56 Ibid., p. 7.

57 Ibid., pp. 7-8.

58 Ibid., p. 5.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 37

Table 13. Funding for Photovoltaics/Wind World Bank China Project Technology Funding Source Amount Global Environment Facility (GEF) Grant $15 million Photovoltaics Funding from other sources (power and PV companies; banks; $129.9 million consumers)

IBRD loan to the PRC government $100 million GEF Grant $20 million Wind Funding from other sources (power and wind companies; banks; $179.1 million consumers)

Total Funding $444 million Source: The World Bank, project appraisal document on a proposed loan in the amount of US $100 million and a proposed project GEF grant of US

$35 million equivalent to the Peoples Republic of China for a Renewable Energy Development Project, Report No. 18479-CHA (May 5, 1999), pp. 7-8.

systems very expensive for Chinese consumers. Such (1) Shell Renewables-ESKOM joint venture (in the consumers, including those in urban areas, do not have Eastern Cape); (2) BP-ESKOM (northern KwaZulu-easy access to credit and usually cannot afford cash Natal); (3) Electricite de France; and (4) NUON (The purchases.59 Netherlands) in partnership with RAPS (South Africa).

Poor quality of products and services. Locally made To ensure that the consortia charge an affordable price modules sold by Chinese PV system companies are not for BIPV electricity, the government pays at least 50 certified, and their performance is often overrated. To percent of the investment cost ($450 to $500). The reduce system cost, smaller systems are sold without remainder of the investment is covered by each con-controllers, a practice that can shorten battery life. Poor sortium using equity or debt financing. The Shell service support after installation can lead to low system Renewables-ESKOM joint venture is an example of how availability, since suppliers of replacement parts are the program will work.63 Each customer will pay $30 for often distant from the installation. installation of a 50 Wp system, large enough to run a small black and white TV, radio, and three to four lights.

South Africa Community-owned and operated companies will operate and maintain each system. Customers prepay The South African government has initiated a rural the local company an $8 monthly fee for service. Upon electrification program with goals for installation of payment, the company issues a card used to operate a BIPV systems. The foundation for the initiative is the prepayment meter integrated into the systems charge governments White Paper on Energy Policy, which controller. The system and access to electricity are establishes universal access to electricity as primary protected against theft by (1) integrating an intelligent South African energy policy goal. About one-third of switching device into the module and battery that South African households have no access to grid deactivates them if the system is disconnected, and (2) electricity, and one to two million of these are too far controlling access to electricity with a prepayment meter from the grid60 for grid extension to be a consideration. that deactivates the system if payments are not made.

Initiated in early 1999, the goal of the BIPV program is Other end-uses for photovoltaics in South Africa installation of 350,000 systems.61 The program will be include:64 implemented by seven private utility consortia, each awarded an exclusive service territory in which it will School PV electrification program operated by install and operate approximately 50,000 BIPV systems. ESKOM. ESKOM installed 1,200 systems (400 and Service territories are awarded using a competitive 900 Wp arrays) to provide light and power. About bidding process. Awards already made include:62 16,000 schools are without electricity.

59 Despite cash shortages, cash sales have grown steadily over the period 1996 to 1999, with continued growth expected.

60 R. Karottki and D. Banks, PV Power and Profit? Electrifying Rural South Africa, Renewable Energy World, Vol. 3/No. 1 (January 2000),

p. 51.

61 Ibid.

62 Ibid., p. 54.

63 Ibid., p. 54.

64 Ibid., p. 52.

38 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Independent Development Trust (IDT). The IDT efficiency development activity in Africa (Lesotho, South has provided PV-based electricity for about 210 Africa, Zimbabwe, Angola, Malawi, and Namibia).

rural clinics (light, vaccine refrigeration, nurses homes). Examples of other Energy Account projects are:

Rural telephone systems operated by Telkom Syria (Project No. SYR/97/E01). Decentralized (national company). Over 2.5 years, Telcom has rural electrification with PV (Rural Electrification purchased 84,000 PV modules rated 32 and 55 Wp Programme) cottage industries established to use for solar-powered wireless systems. excess electricity in summer months since PV systems sized to meet winter electrical loads when Multi-Country Activities Promoted by International solar insolation is lowest67 (3-year project, January Assistance Organizations 8, 1997 to January 8, 2000), $553,700.

U.S. Agency for International Development. During Fiscal Years 1998 and 1999, USAIDs renewable energy

Sudan (Project No. SUD/90/E01 and SUD/90/010).

Rural electrification of at least 50 communities with program installed over 4,000 photovoltaic systems in PV; encourage commercialization of solar energy Brazil, India, Indonesia, the Philippines, Guatemala, and (5-year project, January 12, 1992 to January 12, South Africa.65 1997), $1,800,000.

United Nations Development Program. The United Nations Development Program supports photovoltaic Near-Term Industry Prospects projects under the Bureau for Development and Policy (BDP)/Sustainable Energy and Environment Division In the near-term, the worldwide photovoltaic market (SEED)/Energy and Atmosphere Programme could well grow at an annual rate of 15 to 25 percent.

(EAP)/Energy Account. The Energy Account was Capital cost subsidies, and tax and financial incentives, established in 1980. Since September 1, 1994, it has been driven by the Japanese and German solar building under UNDP/BDP/SEED/EAP. Primary sources of programs, are driving global photovoltaic power market financial support for the Energy Account are The growth. In the long-term, larger manufacturing facilities Netherlands Directorate for International Co-operation being constructed in the United States and abroad are (DGIS), the Government of Japan, and the OPEC Fund expected to achieve economies of scale that reduce the for International Development. cost of manufacturing photovoltaic cells, enabling photo-voltaic power to be cost-effective in more markets Under the Energy Account, the FINESSE (Financing without subsidies. These facilities would have capacities Energy Services for Small Scale End-users) program over 20 MW.

assists countries in identifying and promoting tech-nically feasible and economically viable renewable Manufacturing capacity to meet this demand is being energy technologies. Initiated in 1989 jointly by The constructed in Japan, Germany, and the United States.

World Bank, DOE, DGIS, and UNDP, the programs Photovoltaic cells from U.S.-based manufacturing capa-objective is to provide small loans to small-scale end- city are shipped worldwide, including Japan and users without incurring the high overhead costs for Germany. Such shipments should continue because (1) administering small loans. Large multilateral financing global capacity, including U.S.-based capacity, is needed organizations sell loans wholesale to commercial banks, to meet the world market growth rate; (2) shipment utilities, or NGOs, which make loans at market rates to costs currently do not affect competitiveness; (3) the small users.66 FINESSE was instrumental in the forma- United States has the technically qualified labor required tion of Asia Alternative Energy Program (ASTAE) in for cell production; (4) U.S. vendors provide high-1991. The amount of current PV activity is unknown; quality materials needed for manufacturing cells; and (5) however, there is current renewables and energy U.S.-based research programs are on the cutting edge of 65 U.S. Agency for International Development, Remarks by Ambassador Harriet C. Babbitt (Deputy Administrator), International Conference on Accelerating Grid-Based Renewable Energy Power Generation for a Clean Environment. See website http://www.info.usaid.gov/press/spe_test/speeches/2000/world_bank.html (March 7, 2000), p. 2.

66 United Nations Development Programme, FINESSE Concept. See website http://www.undp.org/seed/eap/activities/finesse.html (February 2000), p. 1.

67 United Nations Development Programme, FINESSE Concept. See website http://www.undp.org/seed/eap/activities/finesse.html (February 2000), p. 2.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 39

new photovoltaic cell technology and manufacturing shipments has decreased from 45 percent in 1995 to 30 techniques. Evidence of the cutting edge is the copper percent in 1999. This has been caused by Japanese-based indium diselenide production capacity being developed PV manufacturing firms, who have increased local by Siemens Solar in California. manufacturing capacity in response to heavy govern-ment support for the integration of PVs into buildings.

Conclusions Future U.S. success in manufacturing cells and modules for export lies in the availability of a highly skilled The world PV market for cells and modules has grown manufacturing work force, high-quality materials, and rapidly since 1994, due principally to heavily subsidized a willingness to send highly trained technicians to work programs for PV use in Japan and Germany. Continued with end users. Near-term growth in U.S. cell and near-term growth is heavily dependent on retention of module production for export is highly dependent on these subsidies. foreign governments retaining their PV end-user support programs. U.S. Federal support for PV use is U.S. manufacturers have shared in the rapidly relatively modest, and most near-term domestic growth expanding world markets, with U.S. cell and module is expected to occur in unsubsidized niche markets or in shipments rising from 26 MW in 1994 to 61 MW in 1999. response to State and local programs. Even in these Much of the increase in U.S. shipments has gone to areas, continued cost reductions will be necessary to export markets, principally Japan and Germany. How- sustain 15-25 percent annual growth in U.S. PV cell and ever, the U.S. share of world PV cell and module module production for the next several years.

40 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

The Impact of Environmental Regulation on Capital Costs of Municipal Waste Combustion Facilities: 1960-1998 period the groundwork for future regulatory approaches Introduction was established. In 1963 the Clean Air Act was passed, and during the 1960s, particulate standards for all Growth in the municipal waste combustion industry incinerators were promulgated under the law. In 1970, slowed dramatically during the 1990s after very rapid the U.S. Environmental Protection Agency (EPA) was growth during the 1980s.1 This leveling of growth is formed. Despite EPAs growing attention to airborne attributed to three primary factors: (1) the Tax Reform pollutants, it and other governmental bodies perceived Act of 1986, which made capital-intensive projects such municipal waste combustion favorably. As many sub-as municipal waste combustion facilities more expensive standard local landfills were closing, municipal waste relative to less capital-intensive waste disposal alter-combustion was considered a technologically advanced native such as landfills; (2) the landmark 1994 Supreme method of reducing the volume of waste. In addition, Court decision (C&A Carbone, Inc. v. Town of Clark-after the Arab oil embargoes in the 1970s, the concept of stown2), which struck down local flow control ordinances generating energy from waste was given impetus by that required waste to be delivered to specific municipal favorable tax and utility regulations. Thus, in sum, this waste combustion facilities rather than landfills that may period saw the birth of the environmental movement in have had lower tipping fees; and (3) increasingly strin-the United States and the attendant focus on air and gent environmental regulations that increased the capital water pollution control. EPAs regulatory approach and cost necessary to construct and maintain municipal framework was established during this period. How-waste combustion facilities. The Energy Information ever, given the facts that the municipal waste com-Administration (EIA) has already published articles bustion industry was in its infancy and that it was seen pertaining to the first two factors.3 This paper focuses on as an improved waste disposal alternative to landfilling, the third factor and attempts to quantify and isolate the few regulatory barriers stood in its path. Actually, tax variables affecting the cost of constructing and retro-and utility regulatory policy provided incentives to fitting municipal waste combustion facilities.

build such facilities.

Background The second period, 1982-1990, marked the growth phase of the municipal waste combustion industry, due Between 1960 and 1998, Federal regulations governing primarily to the existence of various tax and investment plant operations changed considerably. This paper subsidies, as well as acceptance of the technology by divides the 38-year time frame into three different Federal and local governments. EPA continued to focus regulatory periods. The first period, 1960 to 1981, was a its regulatory attention on the air emissions of these time when relatively low-level regulatory attention was plants. Of particular concern were the carcinogenic paid to waste incineration facilities. Yet during this effects of dioxins and furans4 produced by the 1

This article comes from an unpublished report: Eileen B. Berenyi, The Impact of Federal Regulation on Capital Costs of Municipal Waste Combustion Facilities: 1980-1998, Governmental Advisory Associates, Inc., prepared for the Energy Information Administration, U.S. Department of Energy.

2 C&A Carbone, Inc. v. Town of Clarkstown, New York, No. 114, S. Ct. 1677 (1994).

3 Two of the factors are discussed in the following documents and the third is the focus of this paper: J. Carlin, The Impact of Flow Control and Tax Reform on Ownership and Growth in the U.S. Waste-to-Energy Industry, in Energy Information Administration, Monthly Energy Review, DOE/EIA-0535(94/09) (Washington, DC, September 1994), and Public Policy Affecting the Waste-to-Energy Industry and Flow Control and the Interstate Movement of Waste: Post-Carbone, in Energy Information Administration, Renewable Energy Annual 1996, DOE/EIA-0603(96) (Washington, DC, March 1997).

4 Furans and dioxins are trace emissions from the combustion of commonly used materials such as paper and plastics.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 41

combustion process, the toxicity of incinerator ash, and separation and recovery of municipal waste streams ash disposal methodology and testing. By 1987, EPA prior to combustion.

proposed new source performance standards (NSPS) for waste incinerators. Best available control technology Furthermore, in November 1990, Congress enacted the (BACT) was upgraded through the use of acid gas Clean Air Act Amendments of 1990 to the Clean Air Act scrubber/baghouse combinations as well as the instal- of 1977. These amendments directed EPA to develop lation of controls on nitrous oxide production. As air new emission guidelines for existing MWCs and NSPS pollution control technology improved, EPA imple- for new MWC facilities. Five years later, after much mented more stringent standards, forcing municipal discussion, the EPA published air emission guidelines waste combustion facilities to upgrade or install new air for existing MWCs. The new guidelines covered not only pollution control systems. large facilities (plants with capacities greater than 248 tons per day), but also contained requirements for As a concurrent development during this period, in 1986 smaller facilities. While the requirements applying to Congress passed the Tax Reform Act. Prior to 1986, smaller facilities were under challenge, they have been Federal financial incentives for the municipal waste com- modified and were implemented in 1999.

bustion industry included grants for feasibility studies and pilot projects, investment tax credits, favorable tax The new regulations require an aggressive approach to treatment for equipment depreciation, and the ability to the reduction of toxic emissions through a combination qualify for public financing. The Tax Reform Act of 1986 of air pollution control systems, improved monitoring of removed or curtailed most of these incentives for pro- emissions, application of tested combustion methods, spective facilities, creating a negative impact on the personnel training, and front-end materials separation industry by constraining the availability of low-cost programs. These regulations set numerical limits for capital and limiting the favorable tax treatment afforded sulfur dioxide, hydrogen chloride, cadmium, lead, and to the industry. In essence, with the removal of tax mercury emissions. Additionally, more stringent limits protection, municipal waste combustion facilities had to were set for dioxins and furans as well as for nitrogen rely more heavily on tip fees and revenues generated oxides, fugitive fly, and bottom ash. Facilities were from energy sales. With both of these revenue sources required to adopt maximum achievable control tech-facing downward pressure in the 1990s, the financial nology (MACT) to reach acceptable levels of air viability of many projects has been under stress.5 emissions and install continuous emission monitoring Coupled with the increased regulatory costs associated (CEM) systems to track and report emissions on a with meeting BACT, these changes in the tax law periodic basis. MACT includes scrubber/baghouses, as affected the financial viability of many plants. well as mercury and nitrous oxide control systems. The implementation deadline for large facilities to meet these The last period, from 1991 to 1998, represents a time of criteria was December 2000.

intense regulatory activity by EPA, focusing on air emissions of municipal waste combustion projects and The result of this renewed emphasis on air emissions the toxicity of ash produced as a residue of incineration. control has been twofold. First, a number of small, aging In addition, with the decline in revenues from energy projects have shut down, possibly as a result of sales and tipping fees, the adoption of waste recycling, calculating that it was no longer economically feasible to and the growth of modern code compliant large operate, given the large capital investment necessary to landfills, municipal waste combustion no longer fulfilled comply with new Federal regulations. Second, existing its earlier function as a viable disposal technology and a projects are undergoing or are planning significant source of alternative energy. By 1989, EPA began the upgrades to their air pollution control and combustion process of upgrading its NSPS for municipal waste systems.

combustors (MWCs), as municipal waste combustion facilities came to be called. In its final rule of 1991, EPA Prior to a discussion of the methodology and findings, proposed standards for air emissions control. Later several points relevant to this analysis must be noted.

rulings also incorporated requirements for a ban on First, no standard annual reporting mechanism exists by the combustion of lead acid batteries and for materials which municipal waste combustion projects report 5

Data from the Energy Information Administration survey Form EIA-860B, Annual Electric Generator Report % Nonutility, and nonpublished analysis from the Office of Coal, Nuclear, Electric and Alternate Fuels indicate the weighted average capacity factor of the municipal waste combustion facilities in three of the four regions (South, West, and North Central) has dropped below the 85-percent norm (to almost as low as 70 percent in the North Central Region) for the industry during 1998.

42 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

capital or operating costs and additional capital invest- for the given year to control for size of facility. To create ments made over time. Second, no sufficient measure of this profile, the Engineering News Record (ENR) intensity or change in the Federal regulatory environ- industrial building index was used to inflate both initial ment exists. Indeed, even attempting to categorize capital costs and additional capital costs to 1999 dollars, regulatory periods is fraught with difficulty. No fool- thereby removing the effects of inflationary price proof method exists to distinguish where one regulatory increases over time.6 A depreciation factor was added to regime begins and another ends, as final rules by the more accurately represent the value of capital stock at EPA may be challenged in court, modified, or over- any given point in time. For the purposes of this study, turned. Even when dates are published, the determi- a straight-line 25-year depreciation was used, which is nation of when a given regulatory policy will take effect an industry standard. The depreciation factor was is judgmental. Plant owners respond in different ways. applied both to the original capital costs as well as to the Some will act in advance of implementation, making additional capital expenditures made during the changes to their facilities prior to the date; others will relevant time periods.

seek exemptions or attempt to obtain time extensions.

Underlying most of the analysis presented in this paper Upon the creation of this profile, the behavior of capital is the notion that time will be a substitute (albeit an costs of municipal waste combustion projects can be imprecise one) for regulatory period. viewed over time, both in aggregate and separated by technology type or other variables. As technology type was shown to have an impact on capital costs, the first breakdown was done by technology.

Methodology Technology Used for Waste Combustion To assess the regulatory impact on capital costs of municipal waste combustion facilities, a viable database All municipal waste combustors incinerate the waste was constructed from data on municipal waste com- and use the resultant heat to generate steam, hot water, bustion facilities. These data were abstracted from the or electricity. Projects rely on three basic types of Governmental Advisory Associates Resource Recovery technologies: mass burn, modular, and refuse-derived Yearbook series. While information pertaining to 1982 fuel (RDF). Pyrolysis and anaerobic digestion represent through 1998 was available from all Yearbooks, the data waste disposal processes that have yet to be com-were reformatted to be compatible over the 16-year mercially developed in the United States. Although a observation period. There have been seven survey number of such facilities have been built (Table 1), none periods between 1982 and 1998. For a plant coming on of them remain operational or commercially viable.

line in 1982 and still operating as of 1998, there are seven possible observations for any given variable. While Mass burning technologies are most commonly used in certain data remain constant, such as original capital the United States. This group of technologies process cost or year operations commenced, other characteristics raw municipal solid waste (MSW) as is, with little or are dynamic, changing periodically. These variables no sizing, shredding, or separation prior to combustion.

include actual tons processed, gross and net electricity At most plants, large bulky items such as white goods, output, additional capital investment, operation and e.g., large appliances, batteries and/or hazardous maintenance costs, owner, and operator. materials are either prohibited or removed from waste receiving areas by crane operators and other personnel.

Any project in operation as of 1980 is included in the Waste materials are typically deposited in a pit or on a data set. Appendix A lists the projects in the study, and tipping floor and the refuse is fed into individual includes basic information about each facility. Once a furnaces by overhead cranes (or front-end loaders in project closes down, it falls out of the database. Thus, the case of smaller facilities). The wastes are burned in at any period of time, the database consists of projects of one or more furnaces of differing designs, and heat mixed vintages&some recent and others near the end of produced by the combustion process is used to create their operational life. A capital profile for each project steam for use as an energy product. The steam can be was then constructed; profiles contain both initial and sold directly to industrial or institutional customers additional capital costs. Appendix B outlines the and/or used to power a turbine for the generation of definition and construction of the capital cost profile in electricity, which is typically sold to an investor-owned detail. Capital costs were divided by design tons per day or municipal utility.

6 Building Cost Index History (1916-1999), Engineering News Record, Vol. 242, No. 12 (March22/March29,1999), p. 99.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 43

Table 1. Years Projects Began And Ceased removed by screening. In some RDF plants, air classi-Operation fiers, trommel screens, or rotary drums are employed to further process the fuel products, by eliminating Began Operation additional non-combustible materials.

Mass Year Burn Modular RDF Pyrolysis Total All waste combustion systems, to greater or lesser

1980 . . . . 12 15 9 1 37 degrees, generate an ash residue that is buried in 81-84 . . . . 5 19 7 1 32 landfills. The ash residue is composed of two basic 85-88 . . . . 26 23 12 -- 61 components: bottom ash and fly ash. Bottom ash refers 89-92 . . . . 27 1 9 -- 37 to that portion of the unburned waste that fall to the 93+ . . . . . . 7 1 1 -- 9 bottom of the grate or furnace. Fly ash, on the other Total . . . . 77 59 38 2 176 hand, represents the small particles that rise from the furnace during the combustion process; they are Ceased Operation generally removed from flue-gases using air pollution Mass control equipment such as fabric filters and scrubbers.

Year Burn Modular RDF Pyrolysis Total Most research has implicated fly ash as the major

1980 . . . . -- 3 1 -- 4 environmental hazard with respect to ash residue, given 81-84 . . . . 2 1 4 1 8 that heavy metals and organic compounds tend to be 85-88 . . . . 2 6 2 1 11 concentrated in the fly ash, rather than in the bottom ash. In recent years, lined ash monofills have been 89-92 . . . . 2 11 3 -- 16 developed to better isolate this potentially harmful 93+ . . . . . . 8 14 13 -- 35 residue from groundwater supplies.

Total . . . . 14 35 23 2 74 RDF = Refuse-Derived Fuel.

Source: Based on database developed by Governmental Advisory Data Description Associates (Westport, Connecticut).

To carry out the study, a database of 176 municipal waste combustion projects (universe) was created. The Modular facilities employ one or more small-scale com- database initially contained any project that operated for bustion units to process lesser quantities of wastes than at least 1 year commencing in 1980. Two projects were mass burn refractory7 or mass burn waterwall com- ultimately dropped from the database, as they relied bustors.8 This type of plant is usually pre-fabricated and upon a unique technology. Data were collected through can be shipped fully assembled or in modules. Steam is the use of a telephone survey conducted by Govern-most commonly generated from the combustion process, mental Advisory Associates, Inc., using a detailed and many modular plants utilize a two-chamber design interview protocol. Selected aspects of the interview to accomplish this task. Flue gases, which contain format have changed over the 16 years it has been incompletely burned materials, are then channeled into administered. However, the variables selected for the a secondary chamber where final combustion takes purposes of this study have remained the same. For each place. The steam can be sold and/or used to generate plant included in the database, the following variables electricity, not unlike other mass burning designs. were extracted:

The refuse-derived fuel (RDF) technologies employ a Name of Facility two-stage production-incineration system. Wastes are State and Region Where Located pre-processed to produce a more homogeneous fuel Year Commenced Operation product (RDF), than raw MSW. The RDF is either sold to Year Shut Down (if applicable) outside customers or burned on-site in a dedicated Type of Technology (mass burn, modular, RDF) furnace. The refuse is usually shredded to reduce Tons per Day, Design particle size for burning in semi-suspension or sus- Energy Product (i.e. electricity, steam or both) pension-fired furnaces. Ferrous metals can be recovered Gross Power Output Rating in Megawatts (MW) using magnetic separators. Glass, grit, and sand may be Pounds per Hour of Steam Produced 7

Conventional technology used by older mass-burn facilities; energy is recovered in a boiler that is downstream from the combustor process.

8 In the waterwall design, the walls of the furnace consist of closely spaced tubes that circulate water, which cools the furnace walls and absorbs thermal energy to produce steam or electricity.

44 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Original Capital Cost and Year Incurred Table 2 shows the distribution of plants by technology Additional Capital Modification Costs by type and region.9 The Northeast and South regions have Year Incurred had the preponderance of municipal waste combustion Public or Private Sector Ownership facilities. The majority of facilities operating in the Public or Private Sector Operation Northeast are mass burn; the largest proportion of plants in the South are modular. These breakdowns Descriptive statistics were obtained for all the facilities relate to the entire database. At any point in time, the in the database, which are categorized by technology regional distribution may look somewhat different, type. Table 1 summarizes basic data on the plants, given that some plants have shut down, and others came showing the years plants began and ceased operation by on line.

technology type. A large number of facilities (61) com-menced operation in the 1985-1988 time period. Between Table 3 provides further summary statistics with respect 1989 and 1992, the number of projects coming on line to the plants. On average, the initial capital cost of a dropped by almost 40 percent to 37. In the years sub- facility, indexed to 1999 dollars, is $77 million. Addi-sequent to 1992, only nine additional projects came on tional capital investment per plant averages $22 million line. Also, the data show that the dominant technology in 1999 dollars. The average year a project began oper-shifted over time. Among 69 plants that began operation ations was 1985, with a design capacity of 718 tons per through 1984, 34 (49 percent) were modular facilities. day. The average duration of plant operations is 10.8 After 1984, of the 107 plants that came on line, only 25 years, and the average power output rating for elec-(23 percent) were modular facilities. The dominant tech- tricity is 28.3 MW. Steam output is 177,248 pounds per nology from 1985 to 1998 was mass burn. Sixty of these hour. With respect to each characteristic, a considerable plants were built, comprising 56 percent of the projects range is evident between the minimum and maximum coming on line during this period. Reliance on RDF values.

technology wavered somewhat over the time period. Of the 69 total projects built through 1984, 23 percent used Prior to breaking down the data to examine the impact RDF processes. Of the plants coming on line after 1984, of Federal environmental regulations on capital costs, it about 21 percent used the RDF technology. is useful to show the evolution of the composition of the group of facilities in operation at each point in time.

Table 1 also indicates the number of projects that ceased Tables 4 through 6 show the number of firms (by num-operation by time period and technology type. Each ber of years of operation) operating in each calendar successive time period had an increasing number of year from 1975 to 1998 for each of the three technology closures, with the largest amount (35) occurring since types. (Table 4 actually traces back to calendar year 1992. Of the total sample of 176 municipal waste com- 1965.)

bustion facilities in operation from 1980 to 1998, 74 have closed. Categorization by technology type, 14 facilities The key features of the tables are the diagonals (see, (19 percent) that closed were mass burn, 23 facilities (31 for example, shaded area in Table 4) from a non-zero percent) were RDF, and 35 facilities (47 percent) were element in the row labeled with a number and the modular. Both pyrolysis facilities also ceased operation. column and row totals. The diagonal down and to the The high percentage of modular facility closures may be right from any element contains the numbers of facilities due to age. Most were built between 1980 and 1988 and in a cohort (of a particular vintage) that are still oper-have or are reaching the end of their useful life. ating in the calendar year indicated by the column label.

However, the disappearance of modular facilities may The column total represents the number of firms in also be related to the imposition of new air pollution operation for the year. If one picks a particular calendar requirements promulgated since 1991. The additional year (column), the numbers indicate the mix of capital costs associated with the implementation of new vintages of the facilities operating in that year.10 technology may be too onerous for plant owners to bear, given the level of expected revenues.

9 The four regions include the following States: Northeast: CT, ME, MA, NH, NJ, NY, PA, RI, VT; South: AL, AR, DE, DC, FL, GA, KY, LA, MD, MS, NC, OK, SC, TN, TX, VA, WV; North Central: IL, IN, IA, KS, MI, MN, MO, NE, ND, OH, SD, WI; West: AK, AZ, CA, CO, HI, ID, MT, NV, NM, OR, UT, WA, WY.

10 While examining these tables, it is important to remember that facility capacity is not taken into account. If old facilities are replaced by larger scale operations and the hypothesis of increasing returns to scale is indeed true, this could lead to a negatively sloped capital profile or possibly offset increases due to retrofitting.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 45

Table 2. Number and Percent of Projects by Type of Technology and Region Northeast South North Central West Total Technology Number Percent Number Percent Number Percent Number Percent Number Percent Mass Burn . . . . . . . . . . 37 62 24 38 9 26 7 41 77 44 Modular . . . . . . . . . . . . . 13 22 30 47 11 31 5 29 59 34 RDF . . . . . . . . . . . . . . . 10 17 8 13 15 43 5 29 38 22 Pyrolysis . . . . . . . . . . . . -- % 2 3 -- % -- -- 2 1 Total . . . . . . . . . . . . . 60 100 64 100 35 100 17 100 176 100 RDF = Refuse-Derived Fuel.

Notes:

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Totals may not equal the sum of components due to independent rounding.

Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

Table 3. Summary Statistics for Total Municipal Waste Combustion Sample Variable Mean Value Minimum Maximum Number of Plants Initial Capital Cost (1999 Dollars) Per Plant . . . . . . . . . . . . $77,073,438 $1,032,339 $550,385,843 176 Adjusted Additional Capital Costs Per Plant (1999 Dollars) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $22,238,254 $62,157 $263,396,562 70 Year Began Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1985 1965 1997 176 Tons Per Day Design (tons) . . . . . . . . . . . . . . . . . . . . . . . . 718.2 13 4,000 176 Average Years of Operation (years) . . . . . . . . . . . . . . . . . 10.8 1 31 176 Gross Rated Output for Electricity (MW) . . . . . . . . . . . . . . 28.3* 0.5 90 89 Steam Production (pph) . . . . . . . . . . . . . . . . . . . . . . . . . . . 177,248* 2,500 823,000 151 MW = Megawatts.

pph = pounds per hour.

Includes those facilities that are burning only MSW as a fuel. All plants that are co-firing coal and MSW are excluded from this number.

Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

Finally, the row totals indicate the number of facilities (Table 4) with modular (Table 5) projects, reveals several operating with various years of experience, represented differences. First, as of 1998, there are considerably by the row labels. To determine the number of facilities fewer modular plants, 24, than mass burn (63). The that have closed for each technology type, one can decline in modular plant numbers began in 1990, as subtract the column total in the latest year of operation, opposed to 1996 for mass burn plants. Twenty-seven 1998, from the first row total, which represents the total mass burn facilities began operating in the 1990-1998 number of plants with at least 1 year of operating period, as opposed to one modular plant during the experience. same time period. Of the 59 modular facilities that began operations since 1975, 35 ceased operations by 1998.

Examining Table 4 (mass burn), one observes that as of 1998, 63 plants have been in operation. This total is RDF facilities represent the smallest total in the data-down from a high of 68 in 1995. Subtracting the 63 base. This type of facility came on line in 1975 and facilities in operation in 1998 from the 77 plants that increased in number slowly through 1991. Reaching its operated for at least 1 year, one sees that 14 mass burn peak in 1990/1991 (29 plants), numbers have since facilities have been closed. A comparison of mass burn declined to 15 operating plants, equaling the 1986 total.

46 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 4. Number of Firms by Years of Operating Experience and Calendar Year of Operation, Mass Burn Projects Years Calendar Year Oper-ating 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 Total 1 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 2 1 4 3 8 11 7 5 10 5 0 2 5 0 0 0 77 2 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 2 1 4 3 8 11 7 5 10 5 0 2 5 0 0 77 3 0 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 2 1 4 3 8 11 7 5 10 5 0 2 5 0 77 Energy Information Administration/ Renewable Energy 2000: Issues and Trends 4 0 0 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 2 1 4 3 8 11 7 5 10 5 0 2 5 77 5 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 2 1 4 3 8 11 7 5 10 5 0 2 72 6 0 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 2 1 4 3 8 11 7 5 10 5 0 70 7 0 0 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 2 1 4 3 8 10 6 5 10 5 68 8 0 0 0 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 2 1 4 2 8 10 6 5 10 62 9 0 0 0 0 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 1 1 4 2 8 10 6 5 51 10 0 0 0 0 0 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 1 0 0 0 2 2 0 1 1 4 2 7 10 6 45 11 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 0 0 0 0 2 2 0 1 1 4 2 7 10 38 12 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 0 0 0 0 2 2 0 1 1 4 2 7 28 13 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 2 2 0 0 2 0 0 0 0 0 2 2 0 1 1 4 2 21 14 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 1 2 0 0 2 0 0 0 0 0 2 1 0 0 1 3 15 15 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 1 2 0 0 2 0 0 0 0 0 2 1 0 0 1 12 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 1 2 0 0 2 0 0 0 0 0 2 1 0 0 11 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 1 2 0 0 2 0 0 0 0 0 2 1 0 11 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 1 2 0 0 2 0 0 0 0 0 2 1 11 19 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 1 2 0 0 2 0 0 0 0 0 2 10 20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 1 2 0 0 2 0 0 0 0 0 7 21 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 2 0 0 0 0 5 22 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 2 0 0 0 5 23 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 2 0 0 5 24 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 1 0 2 0 6 25 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 2 5 26 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 3 27 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 3 28 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 3 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 2 30 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 31 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 32 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 33 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Total 2 2 3 3 3 5 7 7 7 9 9 10 10 10 10 12 14 14 15 15 19 21 28 39 46 51 59 64 63 63 68 64 64 63 Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

47

48 Table 5. Number of Firms by Years of Operating Experience and Calendar Year of Operation, Modular Calendar Year Energy Information Administration/ Renewable Energy 2000: Issues and Trends Years Oper-ating 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 Total 1 2 1 0 2 1 9 8 7 2 2 4 10 2 7 1 0 0 0 0 0 0 0 1 0 59 2 0 2 1 0 2 1 9 8 7 2 2 4 10 2 7 1 0 0 0 0 0 0 0 1 59 3 0 0 2 1 0 1 1 9 8 7 2 2 4 9 2 7 1 0 0 0 0 0 0 0 56 4 0 0 0 2 1 0 1 1 8 8 7 2 2 4 9 2 7 1 0 0 0 0 0 0 55 5 0 0 0 0 2 1 0 1 1 8 7 7 2 2 4 9 2 7 1 0 0 0 0 0 54 6 0 0 0 0 0 0 1 0 1 1 8 6 6 2 2 4 9 2 7 1 0 0 0 0 50 7 0 0 0 0 0 0 0 1 0 1 1 8 5 6 2 2 4 9 1 7 1 0 0 0 48 8 0 0 0 0 0 0 0 0 1 0 1 1 8 5 6 2 2 4 9 1 7 1 0 0 48 9 0 0 0 0 0 0 0 0 0 1 0 1 1 7 5 6 1 2 3 9 0 7 1 0 44 10 0 0 0 0 0 0 0 0 0 0 1 0 1 1 7 4 4 0 1 3 9 0 7 1 39 11 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 6 4 4 0 1 2 9 0 7 36 12 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 3 4 3 0 1 2 8 0 24 13 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 2 4 3 0 1 2 7 22 14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 3 2 0 1 2 12 15 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2 3 2 0 1 9 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2 2 2 0 7 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 2 5 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3 19 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 Total 2 3 3 5 6 12 20 27 28 30 33 42 42 47 47 45 39 37 32 30 27 25 25 24 Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

Table 6. Number of Firms by Years of Operating Experience and Calendar Year of Operation, RDF Projects Years Calendar Year Oper-ating 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 Total 1 1 1 1 2 4 0 3 0 2 2 2 2 2 6 5 3 1 0 0 0 0 0 1 0 38 2 0 1 1 1 2 4 0 3 0 2 2 2 2 2 6 5 3 1 0 0 0 0 0 0 37 Energy Information Administration/ Renewable Energy 2000: Issues and Trends 3 0 0 1 1 1 1 4 0 3 0 2 2 2 1 2 6 5 3 1 0 0 0 0 0 35 4 0 0 0 1 1 1 1 4 0 2 0 2 2 2 1 2 6 3 3 1 0 0 0 0 32 5 0 0 0 0 1 1 1 0 4 0 2 0 2 2 2 1 2 6 3 3 0 0 0 0 30 6 0 0 0 0 0 1 1 0 0 3 0 2 0 1 2 2 1 2 5 3 2 0 0 0 25 7 0 0 0 0 0 0 1 1 0 0 3 0 2 0 1 2 2 1 2 5 3 2 0 0 25 8 0 0 0 0 0 0 0 1 1 0 0 3 0 2 0 1 2 2 1 2 5 3 2 0 25 9 0 0 0 0 0 0 0 0 1 1 0 0 3 0 2 0 1 2 2 1 2 5 3 1 24 10 0 0 0 0 0 0 0 0 0 1 1 0 0 3 0 2 0 1 1 2 1 2 5 3 22 11 0 0 0 0 0 0 0 0 0 0 1 1 0 0 3 0 2 0 1 1 2 1 2 5 19 12 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 3 0 2 0 1 0 2 1 2 13 13 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 3 0 2 0 1 0 2 1 11 14 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 3 0 2 0 1 0 0 8 15 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 2 0 1 0 1 0 6 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2 0 1 0 1 5 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 1 0 3 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 1 3 19 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 2 20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 21 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 22 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 23 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 24 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 Total 1 2 3 5 9 8 11 9 11 11 13 15 17 21 26 29 29 27 24 24 19 19 20 15 RDF = Refuse-Derived Fuel.

Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

49

Of the 38 facilities, operating since 1975, 15 were still 1. As one moves from the second time period to the operating in 1998. latest one, the number of modular facilities coming on line drop off drastically. In the earliest time Examining the column labeled 1998 in each of the period, modular facilities are the technology of tables, it is apparent that three different mixtures of choice. By the latest time period, only one project vintages are represented. The mass burn table has the began operation.

most entries for projects with 1 year to 12 years of operation, and combined with the low attrition rate, 2. By definition, modular projects always cluster at represented the youngest fleet of facilities. The modular the low end of tonnage throughput, regardless of table shows somewhat the opposite mixture of plants; the vintage of the plant. As can be observed from those still operating cluster between year 11 and year 19 tonnages along the horizontal axis, no daily design due to the high attrition and low entry rates. The RDF tonnage exceeds 600 TPD.

table shows no facility operating in 1998 with less than 9 years of experience. 3. Adjusted capital costs for the modular facilities show similar distributions across time. There do not appear to be any scale economies across any of the time periods. Additionally, a minimal obser-Analysis and Findings vable increase in initial capital costs is evident Three major analyses of the data were conducted to across time periods, due perhaps in part to the assess the impact of Federal environmental regulations smaller combustors, which were initially exempted on municipal waste combustion plants. The first con- from air pollution control requirements.

sisted of breaking down initial capital costs (adjusted for inflation) of each project by technology type and vintage. The top row shows the mass burn projects. Several The second consisted of regressing initial capital cost per findings are prominent:

ton by technology type, vintage, and other selected variables. The third incorporated the concept of the 1. While modular projects may be the losing tech-capital profile, assessing its change over time across all nology, the opposite is true for mass burn projects.

facilities and facilities disaggregated by technology type. As one moves from the first time period to the last, mass burn is certainly the technology of choice.

The majority of projects began operating between Breakdown of Unit Initial Capital Cost by 1982 and 1989; in addition, more mass burn Technology Type and Vintage facilities came on line in the last time period than for both modular and RDF projects.

For the first level of analysis of the relationship between key variables, the sample was broken down by tech- 2. On average, costs appear to rise over time, nology type and vintage of the facility (determined by controlling for inflation. This may be due to the year the project began operation). Average capital increasing requirements for air pollution control cost per ton indexed in 1999 dollars was graphed against add-ons.

size in terms of design tons per day (TPD) for each technology and vintage category, using the three major 3. Evidence of economies of scale is apparent. As technology types. In addition, the year the plant began plants become larger, the initial capital cost per ton operations was divided into three categories, which appears to decrease. This is particularly noticeable roughly correspond to three differing regulatory en- in the middle time period and somewhat apparent vironments prevailing over the 38-year period, 1960 in the later time period.

through 1998. The three periods are 1960 through 1981, 1982 through 1990, and 1991 through 1998. The basic The RDF projects, represented in the third row of concept behind this classification was an attempt to graphs, present less clear-cut patterns. This is partially characterize Federal regulatory intensity prevailing at a due to the nature of these types of plants. Some plants given time, and to determine if change in unit capital include dedicated boilers on site; others do not. Thus, cost could be observed across these different time data for this type of project are not as homogeneous as categories. the other two technology types. Several observations stand out:

The results are shown in Figure 1. If one looks initially at the middle row, which contains data on modular 1. By the 1991-1998 period, RDF was no longer a facilities, one observes that: technology of choice. During the first two time 50 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Figure 1. Initial Capital Costs by Technology Type and Time Period Operations Began Time Period: 1960-1981 Time Period: 1982-1990 Time Period: 1991-1998 360 360 360 320 320 320 280

Mass Burn 280 280 Million 1999 Dollars Million 1999 Dollars Million 1999 Dollars 240 240 240

200

200 200

160 160 160

120 120 120

80 80 80

40 40 40

0 0 0 0 500 1,000 1,500 2,000 2,500 3,000 0 500 1,000 1,500 2,000 2,500 3,000 0 500 1,000 1,500 2,000 2,500 3,000 Tons Per Day Tons Per Day Tons Per Day 360 360 360 320 320 320 280 280 280 Million 1999 Dollars Million 1999 Dollars Million 1999 Dollars Modular 240 240 240 200 200 200 160 160 160

120 120 120

80 80 80

40

40 40

0 0 0 0 100 200 300 400 500 600 700 800 0 100 200 300 400 500 600 700 800 0 100 200 300 400 500 600 700 800 Tons Per Day Tons Per Day Tons Per Day 360 360 360

320 320 320 280 280 280 Million 1999 Dollars Million 1999 Dollars Million 1999 Dollars 240 240 240 RDF 200 200 200 160 160 160 120 120 120

80 80 80 40 40 40

0 0 0 0 1,000 2,000 3,000 4,000 5,000 0 1,000 2,000 3,000 4,000 5,000 0 1,000 2,000 3,000 4,000 5,000 Tons Per Day Tons Per Day Tons Per Day RDF = Refuse-Derived Fuel.

Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

periods, its use increased slightly, which can be To further examine the issue of economies of scale, viewed as neither winning or losing. another measure of output&gross megawatts pro-duced&was used. Initial capital cost dollars/megawatt

2. Costs tend to rise in relation to size. On average, was plotted against tons per day. The results are shown costs appear to increase somewhat over the first in Figure 2. In this figure, a downward slope is evident.

two time periods. From 1991-1998, variation in cost Capital costs per megawatt appear to decrease as design make any conclusion difficult. No economies of tons per day increase.

scale appear evident. In fact, it appears that initial capital cost is directly related to size.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 51

Figure 2. Initial Capital Costs in 1999 Dollars per pattern holds true even for plants built in the late 1980s, Megawatt by Tons Per Day indicating that reasons other than pure equipment replacement were forcing the additional capital expen-12,000 ditures.

(Thousand 1999 Dollars Per Megawatt) 10,000 Finally, Figure 5 summarizes total additional capital dollars spent by municipal waste combustion facilities in Initial Capital Costs 8,000 each of the three basic time periods. In 1999 dollars, the 6,000 total for 1960-1981 was approximately $9.2 million, for 1982 to 1990 it was $367 million, and for 1991-1998 it was 4,000

$1.17 billion.

2,000 Estimation of Linear and Log Linear Regression Models Using Initial Capital 0

0 1,000 2,000 3,000 4,000 Costs Tons Per Day Based on the categorizations above, initial linear regres-sions were estimated, which hypothesized that the initial Source: Energy Information Administration. capital cost of a facility (adjusted for inflation) per daily ton is related to the type of technology employed, the Figure 3 uses the same breakdowns as Figure 1, except size of the project (in terms of design tons per day), and that it uses adjusted additional capital costs per ton the region of the country in which the plant is sited. In instead of initial capital costs. Additional capital costs addition, it was hypothesized that public sector owner-encompass expenditures made after the construction of ship or operation might affect initial capital costs.

the plant for retrofit, upgrade, expansion, or additional Regressions were therefore tried with variables of public investment. As reflected on the graphs, the most activity sector ownership and operation, but these variables with respect to additional investments occurs among were not significant and were therefore dropped. While middle age plants, i.e., those built between 1982 and capital costs were adjusted for inflation (all escalated to 1999 dollars, using the ENR Building Index), no attempt 1990. These plants are still young enough to continue was made at this point to incorporate changes to the operating without major rebuilding, yet may need to facility over time. Each plant only had one data entry, its invest in environmental control or other upgrades. As start date of operation (scaled down by subtracting 1960 might be assumed, the oldest plants show less pro-from the start date, as 1960 was the earliest start date in pensity to make additional capital investments. Costs the database), its size and its original cost of con-may simply outweigh investment returns. Finally, the struction at that point. Only plants employing the three newest projects also reflect a low level of additional basic technologies discussed above were included.

investment, which is to be expected as these projects incorporate the latest environmental and technological improvements during construction. The estimated equation was as follows:

However, while Figure 3 shows the pattern of additional UNIT.CAP = + 0*OP + 1*NCEN + 2*OWN + 3*RDF capital investment by plant vintage, it does not reflect at + 4*NOEA + 5*OPYR + 6*TPD + 7*MBMOD + 8*SOU what time the capital investment was made. If the life of a boiler is 20 years, the additional investments may have where, been made to replace a boiler at the end of its lifespan or UNIT.CAP = initial capital expenditure/design tons in response to regulatory requirements. per day indexed to 1999 dollars OPYR = year operations began minus 1960 ( values Figure 4 plots the year an additional capital investment going from 0 to 38) was made by the year the plant became operational. TPD = design tons per day of refuse processed when What is interesting are the number of dots at or above plant was built the 1990 line on the y-axis. Despite plant vintage, most OWN = ownership type dummy variable additional expenditures came in 1990 or after. This OP = operating entity type dummy variable 52 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Figure 3. Additional Capital Costs Per Ton by Technology Type and Time Period Operations Began Time Period: 1960-1981 Time Period: 1982-1990 Time Period: 1991-1998 120 120 120 Additional Capital Costs in Additional Capital Costs in Additional Capital Costs in 100 100 100 Mass Burn 80 80 80

60 Thousand 1999 Dollars 60 60 Thousand 1999 Dollars Thousand 1999 Dollars

40 40 40

20 20 20

0

0 0 500 1,000 1,500 2,000 2,500 3,000 3,500 0 500 1,000 1,500 2,000 2,500 3,000 3,500 0 500 1,000 1,500 2,000 2,500 3,000 3,500 Tons Per Day Tons Per Day Tons Per Day 120 120 120 Additional Capital Costs in Additional Capital Costs in Additional Capital Costs in 100 100

100 80 80 80 Modular 60 60 Thousand 1999 Dollars Thousand 1999 Dollars 60 Thousand 1999 Dollars

40 40

40

20 20

20

0 0

0

0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700 Tons Per Day Tons Per Day Tons Per Day 120 120 120 Additional Capital Costs in Additional Capital Costs in Additional Capital Costs in 100 100 100

80 80 80 60 60 Thousand 1999 Dollars 60 Thousand 1999 Dollars Thousand 1999 Dollars RDF 40 40 40

20

20 20

0

0 0

0 1,000 2,000 3,000 4,000 5,000 0 1 2 3 4 5 0 1,000 2,000 3,000 4,000 5,000 Tons Per Day Tons Per Day Tons Per Day RDF = Refuse-Derived Fuel.

Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

TYPE OF TECHNOLOGY = DUMMY VARIABLE With the dummy variables in the equation, the base case RDF = 1, if plant is RDF; 0, if not for technology (RDF=0, MBMOD=0) is Mass Burn and MBMOD = 1, if plant is modular; 0 if not the base case for region (NCEN=NOEA=SOU=0) is the REGION = DUMMY VARIABLE West, which includes the following states: Alaska, NCEN = 1, if plant located in North Central Arizona, California, Colorado, Hawaii, Idaho, Montana, Region (IL, IN, IA, KS, MI, MN, MO, NE, Nevada, New Mexico, Oregon, Utah, Washington, and ND, OH, SD, WI); 0, if not. Wyoming.

NOEA =1, if plant is located in the Northeast (CT, ME, MA, NH, NJ, NY, PA, RI, VT); 0, if The overall results from the regression are provided in not. Table 7. The resultant multiple R-squared is 0.34, indi-SOU = 1, if plant located in South (AL, AR, DE, cating that approximately 34 percent of the variation in DC, FL, GA, KY, LA, MD, MS, NC, OK, SC, initial capital cost is explained by its start date, size, TN, TX, VA, WV); 0, if not. technology and region of the country, as well as public Energy Information Administration/ Renewable Energy 2000: Issues and Trends 53

Figure 4. Year of Additional Capital Cost by Year Figure 5. Total Additional Capital Costs by EPA Plant Began Operating Regulatory Time Period 98 1,400

96

1,200 94

Year of Additional Capital Investment

1999 Dollars (Millions) 92

1,000 90

88 800

86

600 84 82 400 80

200 78

76 0 74 1960-1981 1982-1990 1991-1998 72 Time Period 70 7071727374757677787980818283848586878889909192939495 Source: Based on database developed by Governmental Advisory Year Plant Began Operating Associates (Westport, Connecticut).

Source: Based on database developed by Governmental Advisory This equation, including all facilities, regardless of tech-Associates (Westport, Connecticut). nology, explains more of the variation in initial capital costs than the first regression, about 41 percent of the sector ownership and operation. Both the ownership and variation in initial capital costs per ton as opposed to 34 operation variables are statistically insignificant and are percent. In this equation, the base cases were mass burn, excluded from future analysis. Highly significant is the Northeast region, and the first EPA regulatory OPYR, which is positively correlated with capital cost. period (MB=0, NOEA=0, and EPAREG1=0). This As project vintage (controlling for inflation) increases by configuration is repeated in all subsequent tables. Using one, initial capital cost per ton increases by approxi- a log linear format, one observes that relative to mass mately $4,000. Also significant is the dummy variable burn facilities, both RDF and modular projects are less denoting modular facilities. With all other variables costly across all time periods. In addition, project constant relative to the null case of mass burn, modular vintage is associated with a significantly positive impact facilities are less costly per ton by about $17,000. The on cost. In this format, the South, West, and North third significant variable is the SOU regional dummy Central regions have a significant impact (at least at variable. Finally, while not highly significant, tons per approximately the 0.10 level of significance) on cost day carries a negative value. This finding indicates that relative to the Northeast, all showing that costs are less increases in design tons per day (across all facilities) are in these regions. Examining the EPA regulatory periods, associated with corresponding decreases in initial capital one observes that relative to the very early years of costs per ton, suggesting that economies of scale exist. municipal waste combustion facilities (prior to 1982) when there was a minimal level of environmental While the equation points to certain relationships, a regulation, later regulatory periods had a positive but second equation was tested. This equation stipulates a statistically insignificant impact (at the 0.10 level) on log-linear relationship between the variables and initial initial capital costs.

capital cost. In addition, the non-significant variables of public and private sector ownership and operation were However, while this equation explains somewhat more dropped. To assess the significance of the EPA regula- of the variations in plant capital costs, 59 percent of the tory period two dummy variables were created. The cost variation is still not explained by the stated first, EPAREG2, takes the value 1 for plants com- variables. One aspect that may confound the analysis is mencing operations between 1982 to 1990 and 0 for all the fact that technology types are mixed. As the graphs others. The second, EPAREG3, takes the value 1 for in Figure 1 show, different technology types behave all plants commencing operation during the third differently if one looks at initial unit capital costs over regulatory period (1991 and later) and takes the value time and size. In particular, RDF facilities appear to 0 for all others. The null case for this variable is the behave according to a different cost model than do mass first regulatory period, representing the years prior to or modular facilities.

1982. Table 8 shows the results.

54 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 7. Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities Coefficients Value Std. Error t value Pr(>ltl)

Intercept . . . . . . . . . . . . . . . . . . . . -3226.5292 15112.0506 -0.2140 0.8312 NCEN . . . . . . . . . . . . . . . . . . . . . . -24347.8245 8725.8862 -2.7900 0.0059 MBMOD . . . . . . . . . . . . . . . . . . . . -17152.5854 8039.0935 -2.1340 0.0344 WEST . . . . . . . . . . . . . . . . . . . . . . -16895.7814 11312.6400 -1.4940 0.1373 OPYR . . . . . . . . . . . . . . . . . . . . . . 3690.2840 522.5420 7.0620 0.0000 RDF . . . . . . . . . . . . . . . . . . . . . . . -12608.0754 8407.7334 -1.5000 0.1357 SOU . . . . . . . . . . . . . . . . . . . . . . . -16573.1629 7303.8606 -2.2690 0.0246 TPD . . . . . . . . . . . . . . . . . . . . . . . -3.4365 4.5756 -0.7510 0.4537 OPYR = year operations began minus 1960 (values from 0 to 38)

TPD = design tons per day of refuse processed when plant was built TYPE OF TECHNOLOGY = DUMMY VARIABLE RDF = 1, if plant uses refuse-derived fuel; 0, if not MBMOD = 1, if plant is modular; 0 if not REGION = DUMMY VARIABLE NCEN = 1, if plant located in North Central Region; 0, if not SOU = 1, if plant located in South; 0, if not WEST = 1, if plant located in West; 0, if not With the dummy variables in the equation, the base case for technology is Mass Burn and the base case for region is the Northeast.

Notes: Residual standard error: 38624.46237 on 160 degrees of freedom. Multiple R-Squared: 0.3398.

F-statistic: f= 11.76654 on 7 and 160 degrees of freedom, the Pr(>f) is 0.0000.

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

Tables 9-11 show the results obtained by running the log 1982, the second regulatory period (1982-1990) is linear equation displayed in Table 8, disaggregating the associated with a 29-percent increase in cost, and the sample by technology type. third regulatory period with a 53-percent increase in cost. With the Northeast as the base case, one observes As shown in Table 9, looking only at mass burn facilities, from the table that plants located both in the North the regression equation in log linear form explains 64 Central region and in the South region have significantly percent of the variation in cost. Highly significant lower initial capital costs than those in the Northeast.

variables are tons per day, the initial year of operation, and at a lesser level of significance, the dummy variables Table 10 illustrates the results for the same equation run for the second and third EPA regulatory periods. Tons for modular facilities. NCEN is the only statistically per day has an inverse relationship to cost, indicating significant variable. This result can be inferred by the that holding all other variables constant, a 10-percent graphs in Figure 1. By definition, there is little variation increase in tons per day is associated with a 1.3-percent in tons per day across these facilities.

decrease in initial capital cost per daily ton. Approxi-mately a 3-year or a 10-percent increase in project Finally, Table 11 delineates the results for RDF projects.

vintage (or the year the project began operation) is These projects appear to behave differently than mass associated with a 5.9-percent increase in unit costs.11 burn facilities and the modular projects. First, the sign Similarly, the EPA regulatory periods are associated on tons per day is significantly positive, indicating not with increasing costs. Compared to the years prior to only are scale economies not present, but that the 11 Project vintage is measured by a variable that takes a value from 1 to 38, with 38 representing the newest plants, 1 the oldest.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 55

Table 8. Log Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities Coefficients Value Std. Error t value Pr(>ltl)

Intercept . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.0079 0.6399 14.0780 0.0000 EPAREG2 . . . . . . . . . . . . . . . . . . . . . . . . . . 0.2061 0.1354 1.5220 0.1299 EPAREG3 . . . . . . . . . . . . . . . . . . . . . . . . . . 0.2833 0.1993 1.4220 0.1570 LOPYR(ln) . . . . . . . . . . . . . . . . . . . . . . . . . . 0.7229 0.2050 3.5240 0.0006 LTPD(ln) . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0240 0.0439 0.5480 0.5848 MBMOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.1998 0.1236 -1.6170 0.1078 NCEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.3204 0.1176 -2.7240 0.0072 RDF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.4783 0.1139 -4.1970 0.0000 SOU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.1792 0.0977 -1.8340 0.0685 WEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.2423 0.1472 -1.6460 0.1018 LTPD = (In) design tons per day of refuse processed when plant was built LOPYR = (In) vintage of facility (year commenced - 60)

TYPE OF TECHNOLOGY = DUMMY VARIABLE RDF = 1, if plant uses refuse-derived fuel; 0, if not MBMOD = 1, if plant is modular; 0, if not REGION = DUMMY VARIABLE NCEN = 1, if plant located in North Central Region [states below]; 0, if not SOU = 1, if plant located in South [states below]; 0, if not WEST = 1, if plant located in West [states below]; 0, if not EPA Regulatory Period = DUMMY VARIABLE EPAREG2 = 1, if plant commenced operations between 1982 and 1990; 0, if not EPAREG3 = 1, if plant commenced operations in 1991 or later; 0, if not Notes: Residual standard error: 0.50754 on 159 degrees of freedom. Multiple R-Squared: 0.4087.

F-statistic: f = 12.20832 on 9 and 159 degrees of freedom, the Pr (>f) is 0.0000.

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

contrary is true. This result runs counter to the findings investments made after initial construction. Using the for mass burn and modular projects. Second, project capital profile, outlined in Appendix B and graphing vintage does not have a statistically significant effect, capital profile in each year of operation against time, one nor does the EPA regulatory period under which it might expect any of three basic types of investment began operating. Similar to findings for other type of behavior and thus shapes to the graph. If the firm facilities, projects located in the Northeast region are expects EPA regulations to increase costs beyond its more costly on a per-ton basis than those of other ability to maintain some profit level, no additional regions, significantly more so with respect to the West investment would be made by the facility and the capital and North Central regions. profile for that project would be a negatively sloped line.12 If EPA regulations have no effect on capital/unit Average Costs Per Ton Over Time Using the capacity and the firm expects to maintain operations, the capital profile will be reflected in a downward sloping Capital Profile Construct line due to the depreciation of the equipment. This Although the prior breakdowns did appear to show a downward slope will continue until some replacement variation in capital cost behavior of facilities of differing is required. At this time, the profile will increase by the technologies over time, they did not factor in capital amount of the replacement investment, then continue to 12 The firm would ultimately go into noncompliance and would be forced to cease operations.

56 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 9. Log Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities: Mass Burn Coefficients Value Std. Error t value Pr(>ltl)

Intercept . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2452 0.5312 19.2870 0.0000 EPAREG2 . . . . . . . . . . . . . . . . . . . . . . . . . 0.2949 0.1807 1.6320 0.1072 EPAREG3 . . . . . . . . . . . . . . . . . . . . . . . . . 0.5262 0.2131 2.4690 0.0160 LOPYR(ln) . . . . . . . . . . . . . . . . . . . . . . . . . 0.5943 0.1770 3.3570 0.0013 LTPD(ln) . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.1271 0.0421 -3.0200 0.0035 NCEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.2255 0.1271 -1.7740 0.0805 SOU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.1356 0.0866 -1.5680 0.1214 WEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0385 0.1415 0.2720 0.7862 LTPD = (In) design tons per day of refuse processed when plant was built LOPYR = (In) vintage of facility (year commenced - 60)

REGION = DUMMY VARIABLE NCEN = 1, if plant located in North Central Region [states below]; 0, if not SOU = 1, if plant located in South [states below]; 0, if not WEST = 1, if plant located in West [states below]; 0, if not EPA Regulatory Period = DUMMY VARIABLE EPAREG2 = 1, if plant commenced operations between 1982 and 1990; 0, if not EPAREG3 = 1, if plant commenced operations in 1991 or later; 0, if not Notes: Residual standard error: 0.32564 on 69 degrees of freedom. Multiple R-Squared: 0.6368.

F-statistic: f = 17.28255 on 7 and 69 degrees of freedom, the Pr (>f) is 0.0000.

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

decline in linear fashion. This should be reflected on a or (c) increasing capital investments not associated with graph as a horizontal sawtooth pattern about some increased capacity. The first possibility is unlikely. While stationary level of capital. If, however, EPA regulations technological innovations have occurred with respect to increase the necessary capital required per unit capacity, grate configuration, boiler lining, tubing, and furnace one should observe a sawtooth pattern with an upward design, these advancements constitute only marginal trend. This upward trend would represent the rate of improvements with respect to cost. Over the 1980 to capital accumulation for meeting emissions standards. 1998 period, no major new technology has been imple-mented on a widespread basis. Thus, new technological Figure 6 shows the overall trend of average capital costs breakthroughs with embedded higher capital cost do not per design ton for municipal waste combustion projects explain rising costs.

over time, from 1975 to 1998, using the capital profile.

As discussed in a previous section, the capital profile A second explanation may be growing capital ineffi-incorporates both an inflation and a depreciation factor, ciency. This explanation is difficult to rule out com-as well as additional investments made over time, also pletely. While environmental regulation affecting the adjusted for inflation and depreciation over time. industry was changing and becoming increasingly Despite these adjustments, the curve has an overall stringent over the entire period under study, tax and upward slope. Since 1975, the average capital costs per PURPA regulations created strong financial incentives, design ton of waste have been generally increasing. making MWC projects attractive investments until 1987.

As has been discussed, with the enactment of the Tax Several explanations exist for this finding. The upward Reform Act of 1986, tax incentives were severely cost trend may be a reflection of (a) fundamental shifts curtailed. Thus, financial incentives, which may have in technology; (b) increasing inefficiency in the industry; introduced capital inefficiencies in the market prior to Energy Information Administration/ Renewable Energy 2000: Issues and Trends 57

Table 10. Log Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities: Modular Coefficients Value Std. Error t value Pr(>ltl)

Intercept . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3974 1.8134 5.1820 0.0000 EPAREG2 . . . . . . . . . . . . . . . . . . . . . . . . . 0.2274 0.1927 1.1800 0.2435 EPAREG3 . . . . . . . . . . . . . . . . . . . . . . . . . 0.3479 0.4580 0.7600 0.4510 LOPYR(In) . . . . . . . . . . . . . . . . . . . . . . . . . 0.7582 0.6074 1.2480 0.2177 LTPD(ln) . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.1299 0.0850 -1.5260 0.1331 NCEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.3588 0.1808 -1.9850 0.0525 SOU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.2277 0.1469 -1.5500 0.1273 WEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.0208 0.2280 -0.0910 0.9277 LTPD = (In) design tons per day of refuse processed when plant was built LOPYR = (In) vintage of facility (year commenced - 60)

REGION = DUMMY VARIABLE NCEN = 1, if plant located in North Central Region [states below]; 0, if not SOU = 1, if plant located in South [states below]; 0, if not WEST = 1, if plant located in West [states below]; 0, if not EPA Regulatory Period = DUMMY VARIABLE EPAREG2 = 1, if plant commenced operations between 1982 and 1990; 0, if not EPAREG3 = 1, if plant commenced operations in 1991 or later; 0, if not Notes: Residual standard error: 0.39587 on 51 degrees of freedom. Multiple R-Squared: 0.2784.

F-statistic: f = 2.81021 on 7 and 51 degrees of freedom, the Pr (>f) is 0.0149.

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

1987, can no longer be used as an explanation for the the product&producing energy with a lower level of increase in capital costs. emissions&this positive benefit does not directly offset the cost of the additional investment required.

A final reason for the rising capital costs depicted in Figure 6 may be the increasing level of capital invest-ments made over the period, which were not associated Average Capital Cost (Using Capital Profile) Per with an appreciable increase in capacity, nor additional Ton Over Time by Technology Type technological efficiency. Air pollution control add-ons, implemented in response to changing mandates incor- Average capital profiles per ton over time are shown by porated in the Clean Air Act, may have had this effect. technology type in Figure 7 (mass burn), Figure 8 Reduction of air emissions can be achieved by mon- (modular) and Figure 9 (RDF). Analyzing the sample, itoring the composition of the refuse that is burned, one observes the differing behavior of each technology improving combustor technology to achieve a more type. In Figure 7, mass burn facilities show a steep complete burn, thereby lowering noxious emissions and positive slope throughout the mid- to late 1980's, which cleaning up the emissions from the plant. then flattens, assumes a gradual rise and then begins to decline. The steep slope may reflect the myriad of new All three approaches are mandated by EPA. Require- projects that came on line in the 1980s. Averages are ments are clear in terms of the level of back-end air pushed up by new projects entering the mix, which pollution control equipment that must be in place. By contributes to a lesser proportion of older facilities.

adding on this type of equipment, a plant increases the These facilities, with a large amount of depreciated level of investment, but does not affect throughput. capital stock, tend to have a downward influence on While pollution control equipment changes the nature of average total cost per ton. The dramatic rise could also 58 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 11. Log Linear Regression Results Using Initial Capital Costs of Municipal Waste Combustion Facilities: RDF Coefficients Value Std. Error t value Pr(>ltl)

Intercept . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7998 5.0866 0.5500 0.5869 EPAREG2 . . . . . . . . . . . . . . . . . . . . . . . . . -0.1660 0.6667 -0.2490 0.8054 EPAREG3 . . . . . . . . . . . . . . . . . . . . . . . . . 0.0192 1.0688 0.0180 0.9858 LOPYR(ln) . . . . . . . . . . . . . . . . . . . . . . . . . 1.9705 1.7056 1.1550 0.2589 LTPD(ln) . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.3582 0.1120 3.1980 0.0037 NCEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.6244 0.3211 -1.9450 0.0631 SOU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.0710 0.3983 -0.1780 0.8599 WEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -1.2786 0.4597 -2.7820 0.0101 RDF = Refuse-Derived Fuel.

LTPD = (In) design tons per day of refuse processed when plant was built LOPYR = (In) vintage of facility (year commenced - 60)

REGION = DUMMY VARIABLE NCEN = 1, if plant located in North Central Region [states below]; 0, if not SOU = 1, if plant located in South [states below]; 0, if not WEST = 1, if plant located in West [states below]; 0, if not EPA Regulatory Period = DUMMY VARIABLE EPAREG2 = 1, if plant commenced operations between 1982 and 1990; 0, if not EPAREG3 = 1, if plant commenced operations in 1991 or later; 0, if not Notes: Residual standard error: 0.72437 on 25 degrees of freedom. Multiple R-Squared: 0.5363.

F-statistic: f = 4.12972 on 7 and 25 degrees of freedom, the Pr (>f) is 0.0038.

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

be linked to favorable tax treatment and financing not significantly affected capital costs of these facilities.

and/or increased investment in capital stock. One upward spike exists from 1989 to 1991. This marked increase coincides with the beginning of more stringent In addition, new projects tend to be more costly than emission standards and could represent the exit of those of a previous era and are already embedded with facilities that were no longer viable and therefore had up-to-date control and boiler technology. The spike in lower capital costs per unit of output. The exiting of costs during the 1993-1995 period possibly reflects the older facilities during this period might have caused implementation of the 1991 Air Pollution Control regu- average costs to increase. The final downturn could be lations for larger projects. It is still too early to determine associated with the continued depreciation of existing if the downward turn in the slope during the most facilities, without the entry of new projects.

recent years is an ongoing trend or just a temporary halt in additional investments. It does likely reflect the fact RDF facilities average capital cost/unit output shows a that no new projects are coming on line, so average cost rather distinct pattern. The increase in 1981 is associated increases are solely reflective of additional investments with entry of four new facilities. The gentle negative made in upgrades and modifications. slope from 1988 through 1994 seems to indicate a slow depreciation of total capital among the RDF facilities.

With respect to modular facilities, shown in Figure 8, However, averages began to rise as of 1995, perhaps average total capital costs/TPD rose gradually across indicating a response among existing projects to the time, beginning in 1978. It appears that regulations have newly promulgated EPA regulations.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 59

Figure 6. Average Total Capital Costs Adjusted for Figure 7. Average Total Capital Costs Adjusted for Depreciation by Year: All Projects Depreciation by Year: Mass Burn 120,000 100,000 100,000 80,000 1999 Dollars per TPD 1999 Dollars per TPD 80,000 60,000 60,000 40,000 40,000 20,000 20,000 0,000 0

75 77 79 81 83 85 87 89 91 93 95 97 0,0000 75 77 79 81 83 85 87 89 91 93 95 97 Year Year TPD = Tons Per Day.

Source: Based on database developed by Governmental Advisory TPD = Tons Per Day.

Associates (Westport, Connecticut). Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

Figure 8. Average Total Capital Costs Adjusted for Figure 9. Average Total Capital Costs Adjusted for Depreciation by Year: Modular Depreciation by Year: RDF 100,000 120,000 80,000 100,000 Total Average Costs per TPD 1999 Dollars per TPD 80,000 60,000 60,000 (1999 Dollars) 40,000 40,000 20,000 20,000 0,0000 0,000 0

75 77 79 81 83 85 87 89 91 93 95 97 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 Year Year TPD = Tons Per Day. RDF = Refuse-Derived Fuel.

Source: Based on database developed by Governmental Advisory TPD = Tons Per Day.

Associates (Westport, Connecticut). Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

Regressions Using the Capital Profile 1998. (For plants no longer operating, the capital profile The regressions cited above used initial capital costs per would represent capital expenditures through their final design TPD indexed to 1999 dollars as the dependent year of operation.) The construction of the capital profile variable. The following log-linear regressions use the has already been discussed elsewhere in this paper.

same independent variables, but introduce the concept Suffice it to say that this profile includes both initial of the capital profile as the dependent variable. The capital costs and additional capital expenditures made capital profile provides a snapshot of capital expendi- over the life of the facility, depreciated and then indexed tures of a facility as of its most recent year of operation. to 1999 dollars. This approach results in one data point For plants currently in operation, that year would be per plant, which summarizes both the original capital 60 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

investment and the additional expenditures (capital lower capital profile (lower annualized costs per additions) over the life of the project (see Appendix B). ton) than those in the Northeast region. This difference is statistically significant at least at a 0.05 Similar to the regression involving only initial capital level, except for projects in the Western region.

costs, the equation was estimated for each of the three major technology types and is as follows: 3. The coefficient for LOPYR, which represents project vintage, is a positive number and is highly TOTUNIT.CAP = 0*+ 1*LTPD + 2*SOU + 3*LOPYR + significant in the equation. Because LOPYR is

4*NCEN + 5*WEST + 6*EPAREG2+ 7*EPAREG3 based on the year the facility began operation minus 1960, the newer the project vintage, the where, larger the number. Thus, the later the project came on line, the greater the total unit capital costs TOTUNIT.CAP (ln)= capital profile in last operating associated with it. This increase may be related to year/design tons per day indexed to 1999 dollars. additional capital requirements of regulations.

LTPD (ln) = tons per day SOU = dummy variable for region, 1 if in South, 0 if

4. Finally, with respect to the dummy variables repre-in other region senting EPA regulatory periods, both EPAREG2 LOPYR (ln)= Vintage of facility (year commenced and EPAREG3 have a statistically significant im-operation - 60) pact on total capital costs. As compared with the NCEN = dummy variable for region, 1 if in North base case of plants built during the earliest EPA Central, 0 if in other region regulatory period, total capital cost rises with each WEST = dummy variable for region, 1 if in West, 0 subsequent period. The second EPA regulatory if in other region period increases costs by 83 percent, compared to EPAREG2 = Dummy Variable EPA Regulatory the initial period; the third regulatory period is Period: 1= 1982-1990, 0, if not associated with a 182-percent increase.

EPAREG3 = Dummy Variable EPA Regulatory Period: 1=1991 or later, 0, if not.

Modular facilities appear to exhibit substantially dif-This regression equation is estimated for the sample of ferent behavior, as shown in Table 13. The equation firms in operation between the years 1975 and 1998. explains only 29 percent of the variation in total costs, Tables 12, 13, and 14 summarize the results of estimation which is consistent with the nature of these types of of this regression for each of the three technology types. facilities. Modular units tend to be smaller in design capacity and are available in somewhat fixed incre-Looking across technology types, one finds that the most ments. Additionally, expansion possibilities are quite robust equation as measured by the multiple R-Squared limited by design. Several factors may explain the is that for mass burn facilities (Table 12). Each variable findings:

is statistically significant at the 0.05 level, with the exception of the Western region. The equation explains 1. Retrofitting or additional capital costs invested in about 75 percent of the variation in unit total capital these projects may be minimal. As earlier graphs costs. The estimated equation exhibits the following showed, average total capital costs were relatively characteristics: flat over time. Thus, there was little variation in capital costs to explain.

1. The negative coefficient for LTPD reflects the increasing returns to scale effects, which were

2. Furthermore, a number of modular projects began hypothesized. As the designed capacity of the to drop out over time, without making required facility is increased, the number of constant dollars additional investments. This fact would tend to capital required per ton per day design declines. A negate the effect of both vintage as well as the EPA 10-percent increase in tonnage results in about a 2-regulatory period.

percent decrease in capital costs/TPD. This con-stitutes some slight scale economies for the mass burn plants. This finding is similar to the result of As shown in Table 14, the regression model also has the regression using initial capital costs. only moderate explanatory power for RDF projects, accounting for about 45 percent of the variation in total

2. As with the earlier estimations, projects in the capital costs per tons per day. The equation yields the South, North Central, and Western regions have a following findings:

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 61

Table 12. Log Linear Regression Results Using Capital Profile Estimates of Municipal Waste Combustion Facilities: Mass Burn Coefficients Value Std. Error T Value Pr(>ltl)

Intercept . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2263 0.5321 19.2180 0.0000 EPAREG2 . . . . . . . . . . . . . . . . . . . . . . . . 0.6021 0.1809 3.3270 0.0014 EPAREG3 . . . . . . . . . . . . . . . . . . . . . . . . 1.0376 0.2135 4.8600 0.0000 LOPYR (ln) . . . . . . . . . . . . . . . . . . . . . . . . 0.4738 0.1773 2.6720 0.0094 LTPD (ln) . . . . . . . . . . . . . . . . . . . . . . . . . -0.1687 0.0422 -4.0000 0.0002 NCEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.4176 0.1273 -3.2790 0.0016 SOU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.1816 0.0868 -2.0920 0.0401 WEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.0712 0.1418 -0.5020 0.6173 LTPD = (In) design tons per day of refuse processed when plant was built LOPYR = (In) vintage of facility (year commenced - 60)

REGION = DUMMY VARIABLE NCEN = 1, if plant located in North Central Region [states below]; 0, if not SOU = 1, if plant located in South [states below]; 0, if not WEST = 1, if plant located in West [states below]; 0, if not EPA Regulatory Period = DUMMY VARIABLE EPAREG2 = 1, if plant commenced operations between 1982 and 1990; 0, if not EPAREG3 = 1, if plant commenced operations in 1991 or later; 0, if not Notes: Residual Standard Error: 0.32621with 69 degrees of freedom. Multiple R-Squared: 0.7482.

F-Statistic: f = 29.29123 on 7 and 69 degrees of freedom. the Pr (>f) is 0.0000.

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

1. Vintage is associated with a statistically significant As shown with previous equations, results for this effect on total capital costs. An increase in project category of facilities demonstrate different patterns.

vintage of 10 percent is associated with a 52-percent increase in total capital costs. RDF facilities encompass a variety of front-end prepar-ation technologies as well as boiler technologies. For

2. Contrary to mass burn and modular projects, tons example, in some instances, RDF is mixed with other per day is associated with a small, however fuels and burned to generate energy; in other cases, it is statistically insignificant, positive effect on total used exclusively as a fuel. It is possible that the capital costs. producers in this category are sufficiently diverse so as to render any simple description essentially useless.

3. Similar to findings for other technologies, plants in the Northeast region have the highest capital costs.

The coefficients of each of the regional variables Conclusion are negative, the North and the West significantly so. The finding of major significance is that unit capital costs (capital costs per design ton), while controlling for

4. Both the second and the third regulatory periods inflation and adding in a depreciation factor, increase for are associated with reduced total costs (though firms of the same vintage as time progresses. In other only the second period cost reductions are words, at any given point in time, facilities of later statistically significant), relative to the earliest EPA vintages (built at a later time) have higher capital costs period (prior to 1982). This finding runs counter to per ton than do projects built in prior years. This finding results obtained for both mass burn and modular holds true in pooled equations including facilities of all facilities. technologies, as well as for mass burn facilities. The 62 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 13. Log Linear Regression Results Using Capital Profile Estimates of Municipal Waste Combustion Facilities: Modular Coefficients Value Std. Error T Value Pr(>ltl)

Intercept . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7098 1.8400 4.7340 0.0000 EPAREG2 . . . . . . . . . . . . . . . . . . . . . . . . 0.1454 0.1956 0.7430 0.4606 EPAREG3 . . . . . . . . . . . . . . . . . . . . . . . . 0.4240 0.4647 0.9120 0.3658 LOPYR (ln) . . . . . . . . . . . . . . . . . . . . . . . . 0.8634 0.6163 1.4010 0.1673 LTPD (ln) . . . . . . . . . . . . . . . . . . . . . . . . . -0.1232 0.0863 -1.4270 0.1597 NCEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.3154 0.1834 -1.7190 0.0916 SOU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.2660 0.1490 -1.7850 0.0802 WEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.0167 0.2313 -0.0720 0.9428 LTPD = (In) design tons per day of refuse processed when plant was built LOPYR = (In) vintage of facility (year commenced - 60)

REGION = DUMMY VARIABLE NCEN = 1, if plant located in North Central Region [states below]; 0, if not SOU = 1, if plant located in South [states below]; 0, if not WEST = 1, if plant located in West [states below]; 0, if not EPA Regulatory Period = DUMMY VARIABLE EPAREG2 = 1, if plant commenced operations between 1982 and 1990; 0, if not EPAREG3 = 1, if plant commenced operations in 1991 or later; 0, if not Notes: Residual Standard Error: 0.40168 with 51 degrees of freedom. Multiple R-Squared: 0.2930.

F-Statistic: f = 3.01944 on 7 and 51 degrees of freedom. the Pr (>f) is 0.0099.

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

relationship, while still positive, is not statistically sig- scale economies are present. In both regressions using nificant for modular and RDF facilities when the sample initial capital costs and total capital costs, size of the is disaggregated. The results point to the effect of plant was significantly related to cost and carried a changing regulations and the increased capital invest- negative coefficient. Thus, as design tonnage increased, ment necessary to meet air emissions and other costs tended to decrease, holding all other factors environmental standards. constant. Furthermore, the study shows that plants with different technologies behave differently over time. Con-Furthermore, it appears that at least for mass burn fronted with regulatory hurdles, the mass burn projects facilities, EPA regulatory periods are significantly have tended to integrate changes into their capital base, associated with total capital expenditures at a facility. despite higher average capital costs that have resulted.

Controlling for region and vintage, plant owners and Modular plants, however, have opted to cease oper-operators invest more capital dollars in a facility as one ations. Currently, across all technologies, construction of moves across regulatory periods. However, at this new facilities has slowed nearly to a halt. Looking to the point, it cannot be conclusively stated that capital cost future, mass burn and RDF projects may begin to drop increases are due to environmental regulation alone. The out in greater numbers, mimicking the behavior of the issue of regulatory impact remains highly complicated, modular projects.

given the fact that different firms will respond differ-ently to the same set of regulations. One company may opt to stall, another to challenge the regulations in court, To reach a firm conclusion about the direct impacts of a third to comply in advance with potential change, a regulation and other factors, additional data on both fourth to close the facility. capital and operating costs of municipal waste com-bustion projects is necessary. Both capital and operating Several secondary conclusions are also evident. Par- costs must be documented in a consistent manner across ticularly with mass burn facilities, some indications of the facilities selected, and precise dates of capital Energy Information Administration/ Renewable Energy 2000: Issues and Trends 63

Table 14. Log Linear Regression Results Using Capital Profile Estimates of Municipal Waste Combustion Facilities: RDF Coefficients Value Std. Error T Value Pr(>ltl)

Intercept . . . . . . . . . . . . . . . . . . . . . . . . . . -5.4745 6.6658 -0.8210 0.4192 EPAREG2 . . . . . . . . . . . . . . . . . . . . . . . . -1.6887 0.8736 -1.9330 0.0646 EPAREG3 . . . . . . . . . . . . . . . . . . . . . . . . -1.6308 1.4006 -1.1640 0.2553 LOPYR (ln) . . . . . . . . . . . . . . . . . . . . . . . . 5.1677 2.2352 2.3120 0.0293 LTPD (ln) . . . . . . . . . . . . . . . . . . . . . . . . . 0.1901 0.1468 1.2950 0.2071 NCEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . -1.2188 0.4208 -2.8970 0.0077 SOU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.1640 0.5219 -0.3140 0.7560 WEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . -1.5653 0.6024 -2.5980 0.0155 LTPD = (In) design tons per day of refuse processed when plant was built LOPYR = (In) vintage of facility (year commenced - 60)

REGION = DUMMY VARIABLE NCEN = 1, if plant located in North Central Region [states below]; 0, if not SOU = 1, if plant located in South [states below]; 0, if not WEST = 1, if plant located in West [states below]; 0, if not EPA Regulatory Period = DUMMY VARIABLE EPAREG2 = 1, if plant commenced operations between 1982 and 1990; 0, if not EPAREG3 = 1, if plant commenced operations in 1991 or later; 0, if not Notes: Residual Standard Error: 0.94926 with 25 degrees of freedom. Multiple R-Squared: 0.4541.

F-Statistic: f = 2.97038 on 7 and 25 degrees of freedom. the Pr (>f) is 0.0207.

Northeast: Connecticut, Massachusetts, Maine, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia North Central: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Source: Based on database developed by Governmental Advisory Associates (Westport, Connecticut).

additions and changes and reasons for these changes that there are two outputs directly related to each other.

would have to be provided The first is energy, be it electricity or steam. The second is waste disposal, or tons of waste diverted from other However, even if such data became available, the appli- forms of disposal. Standard methods of estimation cation of a traditional cost function raises a number of would have to be adjusted to account for the multiple issues, which have been mentioned throughout this output problem.

document. Most notably is the modeling of firm behavior with respect to the decision to retrofit, replace A third issue is the modeling of the entire pollution equipment, or exit the industry entirely due to the control process and level of outputs. There are, after all, impact of the cost of EPA regulations on profitability. If various technologies and approaches addressing pol-two firms are identical with exact cost structures, and if lution reduction. Emissions reduction and technological one firm opts to replace equipment and upgrade in change, with attendant changes in levels of input and response to regulations and the other decides not to output with respect to air pollution control, are a third replace equipment, then the two firms become different output of a municipal waste combustion project. These and this divergence must be measured. This difference inputs and outputs must be included or accounted for in could be due to geographic location, variations in the a cost estimation function.

regional energy market, or external factors.

This paper has raised initial methodological issues and A second major issue discussed is the measurement of identified further work that must be done to model the outputs of a municipal waste combustion facility. A cost economic behavior of these unique types of facilities.

function relates unit inputs (capital and labor) to unit Hopefully, additional research will be conducted, which outputs. Defining outputs of a municipal waste com- will shed further light on the relationship between cost bustion project is made more complicated by the fact and regulation.

64 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Appendix A. List of Projects Included in Sample Table A1. List of Projects Included in Sample Site State Technology TPD Year Begun Year Closed Adirondack Resource Recovery Facility . . . . . . . . . . . . . . . . NY MB 400.00 1992 Akron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OH RDF 1,000.00 1979 1995 Alaska Solid Waste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AK RDF 200.00 1991 1995 Albany (Answers) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY RDF 800.00 1981 1995 Albany Steam Plant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 600.00 1981 1994 Alexandria/Arlington . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VA MB 975.00 1988 Ames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IA RDF 200.00 1975 Anoka County, Elk River . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN RDF 1,500.00 1989 Auburn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME MOD 200.00 1981 1990 Auburn-(Mid-Maine Waste Action) . . . . . . . . . . . . . . . . . . . . . ME MB 200.00 1992 Babylon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 750.00 1989 Baltimore (Monsanto) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD Py 1,000.00 1976 1981 Baltimore County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD RDF 1,200.00 1976 1991 Barron County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WI MOD 80.00 1986 Batesville . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AR MOD 100.00 1981 1996 Bay County Energy System . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 510.00 1987 Bellingham/Ferndale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WA MOD 100.00 1986 1997 Blytheville . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AR MOD 70.00 1975 1980 Braintree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA MB 240.00 1970 1983 Bridgeport RESCO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CT MB 2,250.00 1988 Bristol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CT MB 650.00 1988 Broward County-North . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 2,250.00 1991 Broward County-South . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 2,250.00 1991 Camden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NJ MB 1,050.00 1991 Carthage/Panola . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TX MOD 40.00 1986 Cassia County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ID MOD 50.00 1980 1991 Cattaraugus County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MOD 112.00 1983 1992 Center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TX MOD 40.00 1986 Central Mass, Millbury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA MB 1,500.00 1988 Charleston County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SC MB 644.00 1989 Chicago NW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IL MB 1,600.00 1970 1996 Cleburne . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TX MOD 115.00 1986 Collegeville (St. John's University) . . . . . . . . . . . . . . . . . . . . . MN MOD 65.00 1981 1987 Columbus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OH RDF 2,000.00 1984 1995 Commerce . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CA MB 360.00 1987 Concord Regional . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NH MB 500.00 1989 Crossville . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TN MOD 60.00 1978 1980 Dade County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL RDF 3,000.00 1986 Davis County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . UT MB 400.00 1988 Delaware County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PA MB 2,688.00 1992 Delaware Reclamation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DE RDF 1,000.00 1984 1993 Detroit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MI RDF 4,000.00 1989 Duluth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN RDF 400.00 1981 Durham . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NH MOD 108.00 1980 1996 Energy Information Administration/ Renewable Energy 2000: Issues and Trends 65

Table A1. List of Projects Included in Sample (Continued)

Site State Technology TPD Year Begun Year Closed Dutchess County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 506.00 1988 Dyersburg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TN MOD 100.00 1980 1992 Easton WMS Town . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PA RDF 300.00 1986 1988 Essex County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NJ MB 2,277.00 1991 Fairfax County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VA MB 3,000.00 1990 Fergus Falls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN MOD 94.00 1988 Fisher Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MI MOD 100.00 1985 Fort Dix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NJ MOD 80.00 1986 Fort Leonard Wood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MO MOD 75.00 1982 1991 Fort Lewis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WA MOD 120.00 1997 Fort Rucker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AL Py 50.00 1984 1988 Franklin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . KY MOD 75.00 1986 1988 Gahanna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OH RDF 1,000.00 1981 1984 Galax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TN MB 55.00 1986 1993 Gatesville . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TX MOD 13.00 1980 1991 Glen Cove . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 225.00 1983 1991 Gloucester Coun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NJ MB 575.00 1995 Hampton County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SC MOD 270.00 1985 1993 Hampton/NASA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SC MB 200.00 1980 Harford County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD MOD 360.00 1993 Harrisburg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PA MB 720.00 1971 Harrisonburg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VA MOD 100.00 1982 Haverhill & Lawrence RDF . . . . . . . . . . . . . . . . . . . . . . . . . . . MA RDF 901.00 1985 1998 Haverhill (Mass Burn) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA MB 1,650.00 1989 Heartland Recycling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IA RDF 100.00 1988 1993 Hempstead . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 2,505.00 1989 Harrisburg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PA MB 720.00 1971 Harrisonburg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VA MOD 100.00 1982 Haverhill & Lawrence RDF . . . . . . . . . . . . . . . . . . . . . . . . . . . MA RDF 901.00 1985 1998 Haverhill (Mass Burn) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA MB 1,650.00 1989 Heartland Recycling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IA RDF 100.00 1988 1993 Hempstead . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 2,505.00 1989 Hempstead (Parsons and Whittemore) . . . . . . . . . . . . . . . . . NY RDF 2,000.00 1978 1980 Hennepin Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN MB 1,200.00 1990 Henrico County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VA RDF 250.00 1983 1988 Hillsborough County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 1,200.00 1987 Honolulu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HI RDF 2,160.00 1990 Humboldt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TN RDF 50.00 1989 1992 Huntington . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 750.00 1991 Huntsville . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AL MB 690.00 1990 Indianapolis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IN MB 2,362.00 1988 Jackson County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MI MB 200.00 1987 Jacksonville Naval Air Station . . . . . . . . . . . . . . . . . . . . . . . . FL MOD 40.00 1980 1983 Johnsonville . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SC MOD 50.00 1981 1985 Kent County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MI MB 625.00 1990 Key West . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MOD 150.00 1986 La Crosse County(French Island) . . . . . . . . . . . . . . . . . . . . . WI RDF 400.00 1993 66 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table A1. List of Projects Included in Sample (Continued)

Site State Technology TPD Year Begun Year Closed Lake County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 528.00 1991 Lakeland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL RDF 300.00 1983 Lancaster County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PA MB 1,200.00 1991 Lane County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OR RDF 500.00 1978 1982 Lee County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 1,200.00 1995 Lewisburg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MOD 60.00 1980 1990 Lisbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CT MB 500.00 1995 Long Beach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MOD 200.00 1988 MERC Biddeford . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME RDF 607.00 1987 MacArthur, Islip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 518.00 1990 Madison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WI RDF 250.00 1979 1993 Marion County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OR MB 550.00 1986 Mayport Naval Station . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MOD 50.00 1979 1993 McKay Bay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 1,000.00 1985 Miami . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OK MOD 108.00 1982 1993 Miami International Airport . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MOD 60.00 1983 1991 Mid-CT-Hartford . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CT RDF 2,000.00 1988 Milwaukee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WI RDF 1,600.00 1977 1982 Montgomery County-Conshocken PA . . . . . . . . . . . . . . . . . . PA MB 1,200.00 1992 Montgomery County-MD . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD MB 1,800.00 1995 Montgomery County (North)-OH . . . . . . . . . . . . . . . . . . . . . . OH MB 300.00 1987 1996 NH/VT S.W. Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NH MB 200.00 1987 Nashville . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TN MB 1,120.00 1974 New Hanover County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NC MOD 450.00 1984 New York (Betts Ave.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 1,000.00 1965 1996 Newport News (Ft. Eustis) . . . . . . . . . . . . . . . . . . . . . . . . . . . VA MOD 40.00 1980 1988 Niagara Falls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 2,500.00 1980 Norfolk MB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VA MB 360.00 1967 1986 Norfolk Naval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VA MB 160.00 1976 1986 North Andover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA MB 1,505.00 1985 North Little Rock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AR MOD 100.00 1976 1989 North Slope Borough . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AK MOD 100.00 1981 Oceanside . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 750.00 1965 1984 Olmstead County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN MB 200.00 1987 Oneida County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MOD 200.00 1985 1995 Onondaga County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 990.00 1995 Osceola . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AR MOD 50.00 1980 Oswego County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MOD 200.00 1986 PERC Orrington . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME RDF 1,100.00 1988 Palestine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TX MOD 25.00 1980 1991 Palm Beach County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL RDF 2,000.00 1989 Park County-Livingston . . . . . . . . . . . . . . . . . . . . . . . . . . . . . UT MOD 75.00 1981 1986 Pascagoula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MS MOD 150.00 1985 Pasco County S.W.R.R.F . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 1,050.00 1991 Perham . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN MOD 116.00 1986 1998 Pidgeon Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DE MOD 600.00 1987 1993 Pinellas County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FL MB 3,150.00 1983 Energy Information Administration/ Renewable Energy 2000: Issues and Trends 67

Table A1. List of Projects Included in Sample (Continued)

Site State Technology TPD Year Begun Year Closed Pittsfield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA MOD 240.00 1981 Polk County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN MOD 103.00 1988 Pope-Douglas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN MOD 80.00 1988 Portland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME MB 500.00 1988 Portsmouth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NH MOD 200.00 1982 1987 Ramsey/Washinton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN RDF 1,200.00 1987 Red Wing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN MOD 72.00 1982 Robbins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IL RDF 1,200.00 1997 1998 Robertson County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TN RDF 50.00 1990 1995 Rochester (Monroe County) . . . . . . . . . . . . . . . . . . . . . . . . . . NY RDF 2,000.00 1979 1984 S.W.R.R.F. (Baltimore) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD MB 2,250.00 1985 SEMASS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA RDF 1,800.00 1988 SERRF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CA MB 1,380.00 1988 Salem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VA MOD 100.00 1978 1994 Saugus RESCO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA MB 1,500.00 1974 Savage (Richards Asphalt) . . . . . . . . . . . . . . . . . . . . . . . . . . MN MOD 57.00 1982 1995 Savannah . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GA MB 500.00 1987 Siloam Springs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AR MOD 18.00 1975 1980 Sitka . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AK MB 24.00 1985 1998 Skagit County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WA MB 178.00 1988 1994 Southeast Resource Recovery Facility . . . . . . . . . . . . . . . . . CT MB 600.00 1992 Southeast Tidewater Energy Project . . . . . . . . . . . . . . . . . . . VA RDF 2,000.00 1988 Spokane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WA MB 800.00 1991 Springfield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA MOD 408.00 1988 St. Croix County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WI MOD 115.00 1987 1995 Stanislaus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CA MB 800.00 1989 Sumner County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TN MB 200.00 1981 Tacoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WA RDF 530.00 1979 1998 Tacoma Steam Plant #2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WA RDF 300.00 1990 1998 Thief River Falls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MN RDF 100.00 1985 1998 Tuscaloosa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AL MOD 300.00 1984 1993 Union County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NJ MB 1,440.00 1994 University City . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NC MB 235.00 1989 1995 Wallingford . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CT MOD 420.00 1989 Walter B. Hall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OK MB 1,125.00 1986 Warren Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NJ MB 450.00 1988 Waukesha . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WI MB 175.00 1971 1991 Waxahachie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TX MOD 50.00 1982 1991 Westchester RESCO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NY MB 2,250.00 1984 Westmoreland County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PA MOD 50.00 1988 Wheelabrator Falls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PA MB 1,500.00 1994 Windham . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CT MOD 108.00 1981 1994 Yankton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SD RDF 100.00 1989 1992 York County . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PA MB 1,344.00 1991 MB = Mass Burn.

MOD = Modular.

RDF = Refuse-Derived Fuel.

TPD = Tons Per Day.

68 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Appendix B. Rationale for the Use of a Capital Profile The standard econometrics method employed in analy- Estimation of a cost function presents a number of sis of firm costs is estimation of the cost function.13 The additional difficulties:

basic premise is that the cost of production for a profit maximizing firm can be summarized as a function of 1. Detailed operating data on each facility do not input prices and output levels. Under certain restric- exist. In particular, the series of rental rates for tions, one can recover all information regarding pro- capital, i.e., the price per unit time of service of one duction technology from such a function.14 To apply this years worth of burning capacity for one ton per methodology one must have observations on each of the day, would have to be constructed from the raw input prices and output levels over a sequence of time data.

periods.

2. The owners and operators of the MWC facilities MWC facilities present somewhat unique complications, are sometimes public entities and may have which make the estimation of a cost function difficult. objectives other than profit maximization.

Unlike most firms, a municipal waste combustion facility has multiple outputs which are a) energy in the 3. The capital demand function derived from the cost form of electricity or steam and b) the diversion of solid function is the cost minimizing level of capital, waste from alternative disposal sites. The levels of these which depends on the actual level of output, not outputs are not independent or even jointly produced by productive or design capacity. However, capital a single process. Kilowatt hours of electric power or additions for the purpose of air emissions reduc-pounds of steam generated by the facility depend tion are based on the design capacity of the waste directly on the quantity (and to some extent, the quality) combustion boilers. Thus, if one uses actual output of the material burned during the combustion stage. as an output measure, and therefore, a lower However, the quantity of material is also a measure of tonnage number than capacity, in conjunction with waste diversion or level of waste disposed. In equation a capital cost that is dependent on design capacity, form: the effects of EPA regulations may be overstated.

Cost = C(wageLabor,rentCapital,Solid Waste, kWh(Solid Waste)) 4. No model or function relates time to regulatory changes. One needs to explicitly incorporate time The last term in the equation (Solid Waste), is in into the estimation process to allow for the deter-parenthesis to show the nesting of waste quantity in the mination of any differential in capital cost between quantity of energy produced. The interrelationship pre-EPA and EPA years. Normally, time may between the two terms makes estimation of this cost be associated with changes in the quality of inputs, function more complicated than that of a single output technology changes, or productivity changes. In or joint production from a single process.15 If it were the case of MWC facilities and other like industries, possible to estimate a straightforward cost function, one time is also related to regulatory shifts.

could then derive the capital demand, as a function of input prices and output levels.

13 This methodological approach was developed by Keith A. Heyen, Governmental Advisory Associates, Inc.

14 See, for example, Varian, H., Microeconomic Analysis, 3rd edition (New York, New York: W.W. Norton and Company, 1992).

15 Generation from a single process is generally assumed in applications where the outputs are similar in nature, e.g., local and toll service in telecommunications. See, e.g., Evans, D. S., and Heckman J.J., Multiproduct Cost Function Estimates and Natural Monopoly Tests for the Bell System, In D. S. Evans, ed., Breaking Up Bell (Amsterdam, New York: North-Holland, 1983).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 69

The Capital Profile Model at time t. The initial capital investment (and, therefore, capacity) decision is not explicitly modeled, since that To address these problems it was deemed necessary to decision depends on local waste disposal needs and forgo direct estimation of the cost function and focus landfill availability. What is of interest for the present only on the capital equipment component. Actual capital purposes is an estimate of purchased is substituted for capital required based on a level of inputs and outputs. One major drawback of this approach is that facilities may be overcapitalized due to 0 Ct

0t

technology tax or other investment incentives. Such overcapital-

start year ization may result in the purchasing of an excess of air capacity pollution control equipment, since the level of pollution control is based on boiler design capacity and not actual Specifically, one seeks to observe the change in capital tonnage throughput. investment per facility, given its technology, design capacity, and the year it began its operation.

The information available on the capital stock includes two types of measures that contain random components: The above model does not allow the making of definitive initial capital investment and additional capital invest- statements regarding a causal relationship between EPA ment. In each period, the firm (facility owner or emissions standards and firm capital costs. Rather, the operator) must decide if it is necessary to augment the goal is to find evidence of an association between the capital stock and, if so, by how much. One such model two.16 As mentioned above, the limitations inherent in for this process would take the following form: survey data and the irregular sampling interval of this particular survey required the researchers to abstract from the model described above.17 The simplified Investment:

structure entailed construction of a sequence of actual Ct Ct if Ct < C t capital stock dollar figures, Ct. This sequence is used as It (1) the dependent variable in a regression in order to 0 if Ct Ct estimate the change of capital expenditures over time, controlling for technology type and capacity, as an Capital Stock:

approximation to the slope of interest as follows:

Ct Ct 1 #(1 /(t ,y )) It 0 Ct 0t

technology

where,

start year capacity Ct is the actual capital Ct* is the required capital, and is a function of The regression methodology employed herein is based capacity, technology type, year of initial operation, on several important assumptions:

and EPA standards

= depreciation factor 1. As of the time period of interest, 1980- 1998, EPA y = initial time period of operation regulations, particularly in the latter period, incor-t = current time period porated the concept of Best Available Control Technology (BACT) type and have a direct effect Ct* represents the physical capital necessary to achieve only on the capital equipment necessary for opera-energy production (and waste diversion) at levels up to tion. Neither technology type nor capacity is the design capacity of the facility for a given technology affected by the type of air pollution control equip-type and vintage and to meet EPA emissions standards ment selected.

16 What would be required to test claims of causality is a structural model of the decision process at the firm level. See Rust, John P.,

Optimal Replacement of GMC Bus Engines: An Empirical Model of Harold Zurcher, Econometrica, Vol. 55, No. 5, 1987, and Kennet, D.

Mark, A Structural Model of Aircraft Engine Maintenance, Journal of Applied Econometrics, Vol. 9, 1994, for examples of these kinds of structural models of capital equipment used in production processes.

17 More precisely, estimation of this model would require annual observations on those factors that affect Ct*. The resulting stochastic specification of Ct would generate some form of a discrete/continuous choice model. The discrete component being whether or not to invest and the continuous component would be the amount of additional investment. The structure of such models is discussed in, e.g., Heyen, K.A., Semiparametric Estimation of Discrete/Continuous Choice Models, Ph.D. dissertation, University of Wisconsin - Madison, 1992.

70 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

2. The expenditure on additional equipment to meet present setting, one is interested in the quantity of EPA standards depends only on the design com- physical capital in dollars expended that is required to bustion capacity of boilers at the facility. produce some level of output at each point in time and in changes to this investment amount over time,

3. The combustion technology has not changed in any adjusting for normal depreciation and inflation. The substantial way over the time period of interest.18 concept of a capital profile is borrowed from the labor economics literature, wherein the researcher is interested

4. The technologies employed at the facilities can be in construction of an earnings profile or path over time divided into three groups: mass burn, modular, for an individual. This profile is then analyzed, assessing and refuse-derived fuel. Within each group the the impact of education, experience and other demo-firms differ only by number of years in operation, graphic or socio-economic factors on the level of initial year of operation, and capacity. earnings. The objective is to characterize and test for changes in the slope of the profile over time.

5. Firms invest in capital equipment to expand capa-city, replace deteriorated equipment or to modify Applying this concept to MWC facilities, one assesses current facilities to meet EPA emissions standards. changes in capital expenditures over time. If the slope is positive, i.e., there is increased expenditure per unit Assumptions (1) and (2) imply that the type of addi- capacity over the range of years in which EPA regula-tional capital investments for the purpose of meeting tions forced a modification of facilities, holding constant EPA standards will be relatively narrow for a given the technology type and age of the facility, then there is facility, since it is determined by the principle of best an indication of an impact of regulation on capital available technology.19 Assumptions (3) and (4) allow for spending. The positive slope does not provide con-treatment of all facilities in the same vintage/year cohort clusive evidence, but points to the EPA regulations as a as similar. Facilities are only allowed to differ over a possible cause for increasing capital outlays on the small number of characteristics. In addition, assump- facilities.

tions (1) through (3), incorporate the notion that replace-ment investment does not materially affect productivity To make meaningful comparisons between firms of or capacity. various sizes, it is necessary to construct the capital profile on a per unit of output capacity basis. This Underlying these assumptions is the contention that a enables one to superimpose time paths for large and facility is not reinvesting to lower costs or to increase small facilities on the same diagram. If there exist productivity. Rather, reinvestment occurs to replace increasing returns to scale effects, this should appear as worn out equipment or to incorporate additional pol- the larger firm having the lower capital/unit capacity lution control systems. A firms decision to enter or to profile. To distinguish replacement investment from net exit the business is not considered here, and its decision additions to capital, a method for accounting for capital to operate in a given period is predicated on the depletion is needed. The industry standard is to use a expected profitability of the facility during that period. boiler lifetime of 25 years, so a straight-line depreciation Under the model presented here, if a firm operates factor of 0.04 was used.20 To obtain a measure of capital profitably, the capital investment amount during that equipment in place, a price index for energy facility period is determined by the vintage of the facility, the construction is used to deflate expenditures.

need to replace equipment, and the prevailing pollution control regulations. The method for construction of the capital profile is summarized as:

J Ij Under these assumptions, it is reasonable to consider the time path of the capital stock for each facility. In the Ct #[1 /#(t tj ) ] (2)

M j 0 Pt j

18 This statement refers to efficiency at the combustion stage. It is assumed that new designs incorporate the current emissions control technology and are more efficient when considering both outputs (combustion and emissions).

19 A structural model of capital investment would include expectations of future emission standards. The BACT assumption and uncertainty about innovations in emissions control technology make long-term planning difficult to model in this context. The planning aspect is ignored so firms make year-to-year decisions.

20 One might consider the use of a straight-line method to be inappropriate in this case because tax incentives and accelerated depreciation methods were available for use by the firms. These considerations are important for the viability decision by the owners. Once the decision to operate is made, what is needed here is the most accurate measure of actual physical capital in place at each point in time.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 71

where, The capital profiles of each facility, as constructed in the J = the number of additional capital investments previous chart, were used as the dependent variable in Ij = jth investment, I0 is the initial investment a sequence of regressions. The estimation of a linear tj = year of investment j regression implies not only that the slopes are constant, Pt but that the scale effects and number of years in j

operation move the capital profile up or down by a fixed

= ENR Building Price Index for time period tj factor over the entire time period. This is somewhat

= constant depreciation factor restrictive but does provide a good first look at the behavior of capital equipment in place.21 As an example, consider a facility in which there is an initial investment of $1,000 and one subsequent addition When viewing the regression results, it is important to of $500 in the next year using 0.10 as the depreciation understand that all the data points in the capital profile factor over a period of 4 years. If the price index is 1.0 in are not random. Equation (2) has imputed values for the first year and 1.05 in the second, then the deflated those time periods, t, where no additional investment is amounts are $1,000 and $476.19, respectively. The capital made.22 Specifically, actual data exist only for those profile would then be calculated as follows: years in which the facilities were surveyed. In non-survey years, cost values were imputed using the Addi- deflation and depreciation factor on the previously Initial Depre- tional Depre- existing data point. Thus, the values of Ct in these time Invest- ciated Invest- ciated Total Year ment Initial ment Additional Capital periods are deterministic, not missing. The resulting estimated function can not be interpreted as a con-1 1,000 1,000 1,000.00 ditional expectation function and should be regarded as 2 900 476.19 476.19 1,376.19 a summary of the sample information on the shape of 3 800 428.57 1,228.57 Ct*. The standard summary statistics for the regressions are presented for completeness and to indicate good-4 700 380.95 1,080.95 ness of fit.

The elements of the Total Capital column would then be divided by the design combustion capacity reported in the associated year.

21 One strategy is to write the regression coefficients as functions of the initial year of operation. This approach is equivalent to working with cohorts. A problem associated with implementation of this method is the small number of facilities starting in most years.

22 There is a vast literature detailing the types of remedies for missing data. For a summary of the basic issues see, e.g., Greene, W.H.,

Econometric Analysis (Upper Saddle River, New Jersey: Prentice-Hall, 1997).

72 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Forces Behind Wind Power by Louise Guey-Lee Zond, a subsidiary of Enron Wind Corporation (392 Introduction MW); NEG Micon (325 MW); and Vestas (159 MW).4 In the past several years, a number of new wind farms Less than 32 percent of new wind power construction have begun commercial operation. Industry sources was located in California in 1999.

have estimated that more than 900 megawatts (MW) of wind capacity was under construction in 1999. A major A number of recent events have triggered an interest in portion of this capacity was constructed outside of wind energy. Significant interest has arisen in the ability California, the birth place of the wind power industry in of renewable energy to survive as a viable energy the United States.1 While the economics of wind turbine source, compared with less expensive fossil fuels, as the technology is improving, it is generally not yet com- electric power industry moves from a regulated to a petitive with fossil fuels.2 Just as the outlook for wind competitive environment. Because renewable energy improves, it can also improve for other energy sources. sources are generally perceived to be more environ-Thus, despite the encouraging portrayal of wind mentally benign than other energy sources, much turbines, they face uncertainty in the future. This paper recently enacted and/or proposed Federal and State looks at the forces behind recent wind energy develop- legislation on electric competition contains provisions ment. encouraging consumption of renewable energy. Hence, in those instances, electric restructuring may actually promote renewable energy use rather than restrain it.

Current Status and Recent Events Wind energy, which is more economically competitive In 1997, wind power generation capacity of 1,579 MW than most other renewable energy options, should produced 3,254,117 megawatthours (MWh) of elec- benefit most from this effort.

tricity.3 More than 99 percent of generation was by independent power producers, and nearly all of it was Another event that increased interest in wind energy located in California. During 1998 and 1999, wind farm was the expiration of the federal production tax credit activity expanded into other States, motivated in part by for any projects beginning operation after June 30, 1999.

financial and regulatory incentives and, in the case of This tax credit was established by the Energy Policy Act Iowa and Minnesota, State mandates. Iowa, Minnesota, of 1992 and provided a 1.5 cent per kilowatthour tax and Texas each had capacity additions exceeding 100 credit for the first 10 years of the projects life. Since all MW that came on line in 1999 (Table 1). During 1999, projects in operation by June 30, 1999, would be eligible wind farm capacity that came on line consisted of state- for the tax credit, most of the capacity that came on line of-the-art wind turbines manufactured primarily by in 1999 came on by that date. Although the credit 1

For a brief history of early developments in the wind power industry, see Wind Energy Developments: Incentives in Selected Countries, in Energy Information Administration, Renewable Energy: Issues and Trends 1998, DOE/EIA-0628(98) (Washington, DC, March 1999). In the early years the Public Utility Regulatory Policies Act of 1978 (PURPA) was instrumental in creating a market for renewable power. It required utilities to purchase power from qualified facilities (including renewable nonutility generators) at prices that were more favorable than they are today. Now some restructuring proposals advocate repeal of PURPA in the belief that PURPAs provisions are inconsistent with the move to competitive electric markets.

2 For a complete assessment and assumptions, see Energy Information Administration, Annual Energy Outlook 2000, DOE/EIA-383 (2000) (Washington, DC, December 1999).

3 Energy Information Administration (EIA), Renewable Energy Annual 1999 With Data for 1998, DOE/EIA-0603(99) (Washington, DC, March 2000), Tables 4 and 5. See the EIA website http://www.eia.doe.gov/cneaf/solar.renewables/rea_data99/rea_sum.html (January 2001). Electric utilities had wind net generation of 5,977 megawatthours and nonutilities had wind gross generation of 3,248,140 megawatthours in 1997.

4 American Wind Energy Association, Wind Energy Projects Throughout the United States. See website http://www.awea.org/

projects/index.html (July 7, 2000).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 73

actually expired, it was reinstated in December 1999, it significant agreement was reached at that time, but is retroactive to July 1999, and extends until the end of future meetings are expected.

2001. The current schedule for new capacity is less ambitious than 1999, but substantial (Table 1). A total of nearly 400 MW of new wind power construction Table 1. United States Wind Energy Capacity (including a significant share of repowered capacity in by State, 1998, and New Construction, California) was expected for 2000. 1999 and 2000 (Megawatts)

Additionally, in June of 1999, the Secretary of Energy New Construction announced the start of a new initiative, Wind Powering Existinga State 1998 1999 2000 America. The stated goal of this program is to have 80,000 MW of wind power generation capacity in place Alaska . . . . . . . . . . . . . * .58 .10 b b by 2020 and have wind power provide 5 percent of the California . . . . . . . . . . 1,487 290.33 208.50 Nations electricity generation.5 Year-end 1998 wind Colorado . . . . . . . . . . . 0 16.00 0 power capacity was about 1,698 MW,6 so this goal repre- Hawaii . . . . . . . . . . . . . 20 0 39.75 sents an enormous increase in capacity additions. The Iowa . . . . . . . . . . . . . .

237.45 0.60 initiative is mentioned here because of its potential im- Kansas . . . . . . . . . . . . 0 1.50 0 portance and the attention it is drawing to wind energy. Maine . . . . . . . . . . . . . 0 0 6.10 However, the full impact of the program on wind energy Massachusetts . . . . . .

0 7.50 will be over the long-term future and is a concern more Michigan . . . . . . . . . . . 1 0 0 so for the Energy Information Administrations (EIA) Minnesota . . . . . . . . . . 129 139.56 32.00 Annual Energy Outlook, and less so for this paper, which Nebraska . . . . . . . . . . 0 1.32 0 covers the recent past and near-term future.7 New Mexico . . . . . . . . 0 0.66 0 New York . . . . . . . . . . 0 0 18.15 Another long-term impact on renewable energy sources Oregon . . . . . . . . . . . . 25 0 0 is concern over global warming and formulating a policy Pennsylvania . . . . . . . 0 0 26.17 to reduce greenhouse gases in accordance with the South Dakota . . . . . . . 0 0 0.75 Kyoto Protocol. A United Nations conference with rep-Tennessee . . . . . . . . . 0 0 1.98 resentatives from more than 160 countries met in Kyoto, Japan, in 1997 to negotiate binding limits for greenhouse Texas . . . . . . . . . . . . . 34 145.82 25.10 gas emissions for developed nations. Carbon dioxide is Utah . . . . . . . . . . . . . . 0 0 .23 the major greenhouse gas. The target for the United Vermont . . . . . . . . . . . 1 0 5.00 States is to reduce carbon dioxide to 7 percent below Wisconsin . . . . . . . . . . 0 21.78 0 1990 levels in the 2008-2012 time frame. Adopting a car- Wyoming . . . . . . . . . . 1 71.25 28.12 bon tax to accomplish this goal would increase the price Total 1,698 926.24 395.05 of fossil fuels (particularly coal) but have little impact on a Defined as net summer capability.

the cost of renewables, which have zero or net zero b Includes a substantial portion of repowered capacity.

carbon dioxide emissions. Assuming a carbon tax is im- * = Less than 0.5 megawatts capacity.

posed, analysis indicates that an increase in the con- NA = Not available.

-- = Not applicable.

sumption of renewable energy, led by wind, would Sources: 1998 Capacity: Energy Information Administration, make a significant contribution to achieving the targeted Renewable Energy Annual 1999 With Data for 1998, DOE/EIA-0603(99) level of reduced emissions.8 The next United Nations (Washington, DC, March 2000) and New Construction: Based on data Conference of Parties (COP) meeting to develop in American Wind Energy Association (AWEA), Wind Energy Projects Throughout the United States, http://www.awea.org/projects/index.html strategies to achieve the goals of the Kyoto Protocol was (July 7, 2000).

held in November 2000 in the Hague, Netherlands.9 No 5

For more details, see the Department of Energys website for this initiative: http://www.eren.doe.gov/windpoweringamerica.

6 Energy Information Administration, Renewable Energy Annual 1999 With Data for 1998, DOE/EIA-0603(99) (Washington, DC, March 2000).

7 For an update on the status of the Wind Initiatives activities, see U.S. Department of Energy, Wind Power Today, DOE/GO-102000-0966 (Washington, DC, April 2000).

8 Energy Information Administration, Impacts of the Kyoto Protocol on U.S. Energy Markets and Economic Activity, SR/OIAF/98-03 (Washington, D.C., March 1998).

9 Energy Information Administration, International Energy Outlook 2000, DOE/EIA-0484(2000) (Washington, DC, March 2000).

74 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

This paper is divided into two main sections followed by Factors Affecting Wind Turbine an appendix. The first section includes a technical Performance discussion of expectations for wind turbine performance and efforts to improve it. The second section provides an Wind Resources and Wind Turbine Machine overview of the world in which the wind power Basics 12 industry is developing. This discussion includes a broad view of the impact of electric power industry restruc- Winds are created by atmospheric temperature and turing, as well as Federal and State incentives. These two pressure variations caused by the sun heating air during main sections are supplemented by an Appendix of State the day, so general wind patterns coincide well with Wind Profiles that takes a snapshot of the status of electricity demand during the daytime. During night-electricity restructuring in each State, the type of time, temperature variations are lessened; therefore, incentives or green power programs available to wind, winds are less severe. Although geostrophic winds (or and status of wind energy development through 2000. global winds) winds determine the prevailing direction References are included so more current information can and magnitude in an area, the surface winds (up to an be obtained as needed.10 altitude of 100 meters) such as sea breezes and mountain winds are key factors in calculating the usable energy content of the wind at a particular site. Wind direction Wind Turbine Performance is influenced by the sum of global and local effects; when larger scale winds are light, local winds may dominate The following sections provide an overview of the tur- the wind patterns.

bine technology being installed in todays wind farms.

These turbines have generation capacities at or above The wind resource is seldom a steady, consistent flow.

225 kilowatts (kW).11 The discussion examines (1) wind It varies with the time of day, season, height above resource issues and related siting considerations, (2) ground, and type of terrain. An areas surface roughness factors affecting wind turbine performance, (3) physical and obstacles are also important determinants in wind and operational characteristics of wind farm turbines resource. High surface roughness and larger obstacles in and (4) operation and maintenance (O&M) consider- the path of the wind result in slowing the wind by ations. The discussion focuses on wind farm turbines creating turbulence. Wind speed generally increases manufactured by NEG Micon, Vestas, and Zond, as they with height above ground.

represent most of new installed capacity in the United States. The discussion indicates that each of their designs A wind turbine converts the force of the wind into a is equally adaptable to a variety of wind farm sites. The torque (turning force) that turns the turbine blades, discussion shows how O&M considerations can be which are connected to the shaft of an electric generator.

managed to ensure that the cost of O&M for a wind The amount of energy that the wind transfers to the farm can be controlled and minimized. blades depends on the density of the air, the blade area, and the wind speed. Wind speed determines how much A major caveat in evaluating information presented in energy is available for conversion to electricity. For wind this section is the availability of data. Performance data farm applications, developers seek sites with an annual on operating wind turbines are frequently proprietary average wind speed of at least 7.0 meters per second and extremely closely guarded. Thus, although some (15.7 miles per hour), measured at a wind turbine hub historical data are available, the data used in this chapter height above ground of 50 meters (164 feet).

are often based upon engineering sources and not actual commercial operational performance data.

10 While this paper acknowledges the importance of some obstacles to the development process, such as congestion on the transmission and distribution system and mitigation of environmental problems (avian mortality, noise and visual obstruction), the paper will focus on elements that support development rather than those that deter it. The latter issues are the subject of future study.

11 American Wind Energy Association, Wind Industry Members Directory: Wind Turbine Manufacturers and Dealers. See website http://www.awea.org/directory/wtgmfgr.html (October 2000). Vestas has a 225 kW turbine.

12 Unless noted otherwise, based on information in Danish Wind Turbine Manufacturers Association, Guided Tour on Wind Energy.

See website http://www.windpower.dk/tour/index.htm (1999).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 75

Wind power density, measured in watts per square Table 2. Definition of Classes of Wind Power meter of blade surface, is used to evaluate the wind Density for 50 Meter (164 Feet) resource available at a potential site. The wind power Hub Height density indicates how much energy is available for Wind Power Wind Power Density Speed a conversion by a wind turbine. The power available at a Class (W/m2) m/s (mph) given wind speed varies with the cube (the third power) 7.0 (15.7) %

of the average wind speed.13 Wind power developers 4 400 % 500 7.5 (16.8) think in terms of ranges of wind power density, termed 7.5 (16.8) %

wind power classes. Sites with a wind power class rating 5 500 % 600 8.0 (17.9) of 4 or higher are preferred for large-scale wind plants (see Table 2), which have installed capacity of at least 10 8.0 (17.9) %

6 600 % 800 8.8 (19.7)

MW.14 For any given wind power class, the wind power density range and wind speed range increases with hub 7 > 800 > 8.8 (19.7) height; a hub height of 50 meters is the approximate hub a Mean wind speed is based on the Rayleigh speed distribution of height for utility-scale turbines. For instance, NEG equivalent wind power density. Wind speed is for standard sea-level Micon turbine hub heights range from 40-55 meters for conditions. To maintain the same power density, speed increases 3 600 kW and 750 kW turbines, to 49-80 meters for their percent /1000 m (5 percent/5000 ft) of elevation.

W/m2 = Watts per square meter.

900 kW to 1.5 MW turbines.15 Depending on rotor Notes: Vertical extrapolation of wind speed from 10 meter baseline diameter, Vestas turbine hub heights range from 35-65 height based on the 1/7 power law.

meters for their 600 kW and 660 kW models, to 60-100 Source: D.L. Elliott, C.G. Holladay, W.R. Barchet, H.P. Foote, W.F.

meters for their 1.5 MW and 1.65 MW models.16 The Sandusky, Wind Energy Resource Atlas of the United States, DOE/CH 10093-4 (Washington, DC, October 1986), Table 1.1.

Zond turbine hub height is 53 meters for their 750 kW turbines, with an optional 65 meter height for the 48 meter and 50 meter rotor diameter versions of the 750 kW turbine.17 rated power. In contrast, the variable speed generator used in the Zond Z-48 design enables the turbine to The goal of wind turbine design is to convert as much of maintain rated output of 750 kW over the range of wind the power in wind, illustrated by the wind power classes speeds listed in Table 3, starting with 11.6 meters per in Table 2, into turbine generator power output. The second (the speed at which it first achieves 750 kW power curve for a wind turbine shows this relationship output), because the generator speed varies with wind of wind speed to turbine power output by plotting speed to maintain rated output. Power output per unit turbine power output (e.g., kilowatts) as a function of of rotor swept area offers a way to compare perfor-wind speed (e.g., meters per second). Power curve mance among wind turbines. Restated, the goal of wind values vary among turbines because turbine design turbine design is to obtain the highest value of power approaches differ. The impact of design on power curve output per unit of rotor swept area (Table 3) for the values is illustrated by comparing the wind speeds at lowest capital cost.

which various turbines achieve rated power. For instance, the Zond Z-48 turbine achieves 750 kW rated Siting Factors Affecting Wind Turbine power output at a lower wind speed (11.6 meters/sec- Performance ond) than does the NEG Micon Multi-power 48 (16 meters/second) (Table 3). The shape of the power curve Several performance factors contribute to the selection also varies with turbine design. For instance, the NEG of a wind farm site. Choosing a terrain with the least Micon Multi-power 48, which uses a generator that number of obstacles, least roughness, and the most operates at constant speed, produces less than 750 kW expansive views is generally a good practice. The orien-output at wind speeds less than or greater than 16 tations of trees and shrubs and erosion patterns along a meters/second (Table 3), the speed at which it achieves terrain provide clues to prevailing wind directions.

13 E. Eggleston, American Wind Energy Association, Wind Energy FAQ: How Can I Calculate the Amount of Power Available at a Given Wind Speed? See website http://www.awea.org/faq/windpower.html (February 1998).

14 Personal communication between Donald M. Hardy, PanAero Corporation (Lakewood, CO) and William R. King, SAIC, 1999.

15 NEG Micon turbine specifications. See website http://www.awea.org/directory/negmicon.html (October 23,2000).

16 Vestas turbine specifications. See website http://www.awea.org/directory/vestas.html (October 23, 2000).

17 Enron Wind Corporation turbine specifications. See website http://www.awea.org/directory/enronwind.html (October 23, 2000).

76 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 3. Utility-Scale Wind Turbines&Performance Comparison Power Output (kW) Power Output/Rotor Swept Area (W/m2)

Turbine Manufacturer/ Rotor Model Swept Wind Speed (meters/second) Wind Speed (meters/second)

(Rotor Diameter/ Area Rated Power) (m2) 11.6 14 15 16 17 11.6 14 15 16 17 NEG Micon/Unipower 64 NM 1500C/64 3,217 1,168 1,490 1,542 1,562 1,564 363 463 479 486 486 (64 meters/1500 kW)

Vestas/V66 (66 meters/1650 kW) 3,421 1161 1,549 1,616 1,641 1,650 339 453 472 480 482 NEG Micon/Multi-power 48 NM 750/48 1,824 610 730 746 750 745 334 400 409 411 408 (48.2 meters/750 kW)

Vestas/V47 (47 meters/660 kW) 1,735 569 651 660 660 660 328 375 380 380 380 Zond/Z-48 (48 meters/750 kW) 1,810 750 750 750 750 750 414 414 414 414 414 2

m = Square meters W/m2 = Watts per square meter Source: NEG Micon, Vestas, and Zond wind turbine specification sheets for design information (rotor diameter, swept area, and rated power output). Power output at different wind speeds from manufacturer contacts, 1999.

Meteorological data, preferably spanning periods cost, depending on the terms of the wind electricity greater than 20 years, are used to screen potential sites. purchase agreement between the wind farm developer Meteorologists collect wind data for weather forecasts and the electric utility. For example, the Southwest Mesa and aviation, and that information is often used to assess Wind Energy Project in Texas uses 700 kW NEG Micon an areas potential for wind energy. However, wind turbines, which produce 600 volt electricity.18 Electricity speeds and wind energy are not measured with great travels from the turbine to a field transformer to the enough precision when monitored for weather fore- wind farm substation to the utility transmission line.

casting to enable placement of turbines within a site. For Therefore, the following transmission capital must be example, wind speed is influenced by surface roughness, included in the project cost: field transformers, sub-obstacles, and contours of the local terrain. The impact station, and transmission lines to connect each element, of these factors may be estimated when screening for ending with connection to the utility line. Congestion on potential wind farm sites. the regional transmission system is also a consideration.

It would be undesirable to locate a new wind farm Land conditions, which affect the cost of site prep- where the transmission system would not accommodate aration, are a factor in wind farm economics and in site the power generated.

selection. The earth must be able to withstand the com-bined weight of a tower foundation and the tower, Once a potential site is selected, meteorological data are turbine, and rotor. The earth and geography leading to measured at points within the site as part of wind and including the site must be accessible to large, heavy turbine micrositing. Micrositing refers to the actual trucks and cranes used to haul wind turbine components placement of turbines within a wind farm site to opti-on to the site and to install the turbines. The cost of mize electricity production.

building a road to the site must also be factored into site selection. Capacity Factor Connection to the electric grid presents other issues that Capacity factor is defined as the actual annual wind must be addressed when choosing a wind farm site. farm energy output, in kilowatthours, divided by the Grid connection may be a component of total project rated maximum turbine output, in kilowatts, times 8,760 18 NEG Micon, Southwest Mesa Wind Energy Project: Development, Construction, and Installation of a 75 MW Wind Farm, video, 1999.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 77

hours/year. For a 100 kW turbine producing 175,000 may not have an economic advantage over a wind kWh in a year, the capacity factor would be: turbine with a lower factor. For example, compare two wind turbines with the same rotor diameter but different Capacity Factor generator capacities in a location with daily wind gusts

= ((175,000 kWh/year) /(100 kW x 8,760 or seasonal wind variations that are above the mean hours/year)) x 100 daily or seasonal speed. The turbine with the larger

= 20 percent generator may be more economical because it enables higher power output, thus more electricity, when the Factors affecting the magnitude of the capacity factor wind turbine can take advantage of higher wind speeds.

include wind resource intermittency, the wind farm This strategy would tend to lower the capacity factor, sites wind speed distribution, turbine design, and using less of the available capacity of a larger generator.

turbine reliability. The degree of wind resource inter- However, the strategy is economical if the value of the mittency may vary both daily and seasonally. For a electricity production can be increased more than the given turbine design, turbines sited where the wind incremental cost of the larger turbine over a smaller resource is more intermittent will have a lower capacity capacity turbine. The value of the electricity depends on factor. The wind farm sites wind speed distribution, daily or seasonal variations in electricity price. For and the associated average annual wind speed, affect instance, increased electricity production from a larger annual electricity output. The annual electricity output turbine has more value if produced during peak, rather for a wind turbine increases with average annual wind than off-peak, periods of a utilitys load curve.

speed, since more hours of operation at a higher wind speed mean a higher average kilowatt power output from the turbine. Thus, for a given turbine design, wind Physical and Operational Characteristics of farm sites with higher mean wind speeds have higher Wind Farm Turbines capacity factors. Historical data show wind farm To understand the advances in wind farm technology, capacity factors in the range of 25 percent to nearly 36 general knowledge of a wind turbine and its com-percent (Table 4). An objective of turbine design is to ponents is essential. Recent advances in component maximize annual power output, which would increase design in addition to site-specific optimization have been the capacity factor. Higher capacity factors, compared to instrumental in improving energy output and reducing Danish data and DOE 1997 baseline data for class 4 operation and maintenance costs. The text box that winds, are projected for the Zond Z-750 Series turbines follows on page 84 provides a brief summary of the (Table 4) because the Zond Z-750s variable speed gen- components in a wind turbine (see also Figure 1).

erator design, taller tower, and larger rotor swept area enable a greater amount of wind energy to be converted Physical Characteristics to electrical energy. Finally, an increase in turbine relia-bility would be reflected in an increase in the capacity During the past quarter century, extensive public- and factor. private-sector efforts were made to optimize wind turbine design, including development of advanced Annual electricity production can be estimated from the rotor blade materials, design concepts, advanced turbine turbine's power curve, which plots kilowatt output as a designs, and other wind energy conversion systems function of wind speed.19 Alternatively, electricity pro- (WECS) components, such as towers.

duction from wind turbines may be estimated by statistical means.20 This section discusses the results of these efforts and their impact on enabling wind farm developers to Contrary to conventional steam or nuclear power gen- optimize WECS design based on site requirements.

eration, the wind turbine with the larger capacity factor Information focuses on technology deployed by 19 Divide the kilowatt output that corresponds to the sites average wind speed by the turbines rated maximum output to estimate a capacity factor. Then multiply the estimated capacity factor by 8,760 hours0.0088 days 0.211 hours 0.00126 weeks 2.8918e-4 months per year to estimate annual electricity production. This estimated value is somewhat lower than the actual annual production because any percent increase in wind speed above the mean results in a power of three increase in the wind turbine electricity output See American Wind Energy Association, Wind Energy FAQ: How Does a Wind Turbines Energy Production Differ from Its Power Production? See website http://www.awea.org/faq/basicen.html (October 23, 2000).

20 The Weibull and Rayleigh probability density functions are commonly used to estimate annual electricity production when precise site data are lacking. Both distributions are variations of a bell curve. The Weibull distribution has two parameters: mean value and shape; the Rayleigh distribution is a Weibull distribution with the shape parameter equal to 2. See Danish Wind Turbine Manufacturers Association, Describing Wind Variations: Weibull Distribution. See website http://www.windpower.dk/tour/wres/weibull.htm (October 23, 2000).

78 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 4. Examples of Wind Farm Capacity Factors Turbine Description Turbine Wind Farm Location Wind Farm Manufacturer/ Max. Power Rotor Swept Capacity Factor (Developer) Capacity (MW) Model Output (kW) Hub Height (m) Area (m2) (percent)

Denmark 27.6-28.8a Micon 600b 40-70 1810-1452 28.5 (historical)c Denmark 19 Vestas 500d 40 1195-1521 25.2 (historical)e Hypothetical, Class 4 Windsf 25 DOE 1997 500 40 1,134 26.2 (based on baseline historical) technology Hypothetical, Class 6 windsg 25 DOE 1997 500 40 1,134 35.5 (based on baseline historical) technology Storm Lake II, Iowa (Enron)h 80 Zond Z-750 750 63 1,963 32 (historical) 38 (projected)

Lake Benton I, Minnesota 107 Zond Z-750 750 51 1,810 28 (historical)

(Enron)I 35 (projected) a Wind Turbine Performance Summary, WindStats Newsletter, Vol. 11, No. 1 through 4, four consecutive quarters of data from winter 1998 through autumn 1998, wind farm section of tables with Danish data. During the winter 1998 and spring 1998 quarters, 46 turbines were operating. During the summer 1998 and autumn 1998 quarters, 48 turbines were operating.

b NEG Micon. See website http://www.negmicon.dk/English/products/ (November 1999). The 600 kW turbine comes in two rotor diameters: 48 meter (1810 m2 swept area) and 43 meter (1452 m2 swept area). Hub height options for the 48 meter model are 46 meters, 60 meters, and 70 meters. Hub height options for the 43 meter model are 40 meters, 46 meters, and 56 meters.

c Wind Turbine Performance Summary, WindStats Newsletter, Vol. 11, No. 1 through 4, four consecutive quarters of data from winter 1998 through autumn 1998, wind farm section of tables with Danish data. An annualized average capacity factor was calculated by averaging the four seasonal capacity factors provided in the WindStats Newsletter.

d Turbine information for the Vestas 500 kW model from personal communication between Soren Christensen, Project and Sales Coordinator, Vestas-American Wind Technology, Inc., and William R. King, SAIC, November 1999. The 40-meter hub height is common in Denmark. The 500 kW turbine comes in three rotor diameters: 39 meters (1195 m2 swept area), 42 meters (1385 m2 swept area), and 44 meters (1521 m2 swept area).

e Wind Turbine Performance Summary, WindStats Newsletter, Vol. 11, No. 1 through 4, four consecutive quarters of data from winter 1998 through autumn 1998, wind farm section of tables with Danish data. An annualized average capacity factor has been calculated by averaging the four seasonal capacity factors provided in the WindStats Newsletter.

f U.S. Department of Energy (Office of Utility Technologies) and Electric Power Research Institute, Renewable Energy Technology Characterizations, TR-109496 (Washington, DC, December 1997), p. 6-12.

g U.S. Department of Energy (Office of Utility Technologies) and Electric Power Research Institute, Renewable Energy Technology Characterizations, TR-109496 (Washington, DC, December 1997), p. 6-12.

h Assumed Generation for Historical Capacity Factor: Energy Information Administration, Form EIA-900, Monthly Nonutility Power Report, Other Data: Enron Wind Corporation, See website http://www.wind.enron.com/newsroom/casestudies/stormlake.html (October 23, 2000). Note: Historical capacity factor is preliminary, calculated with preliminary generation data for 12 consecutive months during 1999 and 2000.

i Assumed Generation for Historical Capacity Factor: Energy Information Administration, Form EIA-900, Monthly Nonutility Power Report, Other Data: Enron Wind Corporation. See website http://www.wind.enron.com/ newsroom/casestudies/lb1.html (October 23, 2000). Note: Historical capacity factor is preliminary, calculated with preliminary generation data for 12 consecutive months during 1999 and 2000.

Source: Energy Information Administration.

Enron/Zond, Vestas, and NEG Micon, the current major Following are some of the major improvements that wind farm developers in the United States. have made these benefits possible:

Technology Advances for Improved Wind Farm Per- Airfoil Design. Over the past 20 years, inter-formance and Reliability. The current generation of national research efforts have led to new airfoils utility-scale wind turbines uses technology developed designed specifically for horizontal axis wind over the past 20 years. Advances in technology have turbines. In the United States, the Zond Energy resulted in lower installed cost per kilowatt of a wind Systems Z-750 series utility-scale turbines use turbine, improved turbine performance, and improved airfoil designs developed at the National Renew-turbine reliability and reduced maintenance cost. able Energy Laboratory (NREL). The results of Energy Information Administration/ Renewable Energy 2000: Issues and Trends 79

Turbine Component Function Nacelle Contains the key components of the wind turbine, including the gearbox, yaw system, and electrical generator.

Rotor blades Captures the wind and transfers its power to the rotor hub.

Hub Attaches the rotor to the low-speed shaft of the wind turbine.

Low speed shaft Connects the rotor hub to the gearbox.

Gear box Connects to the low-speed shaft and turns the high-speed shaft at a ratio several times (approximately 50 for a 600 kW turbine) faster than the low-speed shaft.

High-speed shaft with Drives the electrical generator by rotating at approximately 1,500 revolutions per minute (RPM). The mechanical mechanical brake brake is used as backup to the aerodynamic brake, or when the turbine is being serviced.

Electric generator Usually an induction generator or asynchronous generator with a maximum electric power of 500 to 1,500 kilowatts (kW) on a modern wind turbine.

Yaw mechanism Turns the nacelle with the rotor into the wind using electrical or other motors.

Electronic controller Continuously monitors the condition of the wind turbine. Controls pitch and yaw mechanisms. In case of any malfunction (e.g., overheating of the gearbox or the generator), it automatically stops the wind turbine and may also be designed to signal the turbine operator's computer via a modem link.

Hydraulic system Resets the aerodynamic brakes of the wind turbine. May also perform other functions.

Cooling system Cools the electrical generator using an electric fan or liquid cooling system. In addition, the system may contain an oil cooling unit used to cool the oil in the gearbox.

Tower Carries the nacelle and the rotor. Generally, it is advantageous to have a high tower, as wind speeds increase farther away from the ground.

Anemometer and wind Measures the speed and the direction of the wind while sending signals to the controller to start or stop the vane turbine.

similar research by European manufacturers are Center has developed test procedures to assess incorporated into the blade design of European compliance with standards. For instance, their test turbines. NRELs airfoils, when used with stall- procedures to assess compliance with power regulated turbines, have produced 23 percent to 30 quality, structural load, blade structural load, percent more electricity annually in the field. power performance, and noise standards have been accepted by the American Association of

Structural Testing Improvements. Structural test Laboratory Accreditors and by certifying parties bed facilities have been constructed for full-scale throughout the world. Additionally, NREL has testing of turbines. Tests are performed on developed a wind turbine design evaluation prototypes to validate design assumptions, test quality system to enable design certification by materials, and make corrections. Testing includes international organizations.

fatigue testing, strength static testing, and non-destructive analysis such as photoelastic stress Power Electronics Advances. Power electronics analysis. International efforts have resulted in enable variable speed operation of the Zond Z-750 safety and performance certification standards for turbine, improving electricity generation efficiency wind turbines. In the United States, the Under- and reducing structural loads by allowing a light-writers Laboratories, Incorporated (UL), certifies weight, low-cost configuration. In both the United turbines using international standards issued by States and Europe, improvements in inverter the International Electrotechnical Commission design21 and smart controls and reduction of the (IEC). The NREL National Wind Technology cost of such components has contributed to 21 The inverter converts direct current (DC) to alternating current (AC). This is necessary in some turbine designs because variations in wind speeds can cause variations in the frequency (e.g., 60 cycles per second) of AC power production, which must be tightly controlled in order to be usable. In contrast, DC power conditioning issues are easier to manage. Therefore, wind turbines often convert AC-generated wind power to DC, condition it, and use the inverter to convert it back into AC electricity.

80 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Figure 1. Wind Energy System Schematic Rotor B lade Nacelle with S wept A rea Ge arbox a nd of B lade s Ge nerator Rotor Diameter Hub Height Tower Unde rground E lectrical Fo undation Conn ections (S ide View)

(Fron t View )

Source: Canada Center for Mineral and Energy Technology (Ottowa, Canada, 1999)

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 81

addressing power quality more cost-effectively. required for offshore applications, the cost of such wind Remote access and control of wind systems via plants can be up to 30 percent higher. However, due to modem or satellite has also become common place stronger winds offshore (as well as the waters smoother in most sites. surface than land), the higher production will offset the higher installation costs over the life of the facility. Aside

Smart Aerodynamic Control Devices. Smart, reli- from this, Vestas and Micon still lead the markets in able controls reduce the likelihood that high winds manufacturing advanced, land-based, utility-scale tur-and generator load loss will cause significant dam- bines. In 1999, Micon and Vestas were the number one age to turbines. In addition, such controls enable and number two wind turbine manufacturers world-turbine operation to adapt to natural wind speed wide, sharing about 40 percent of the global market.23 variations, insect-impact accumulations, and minor blade damage, which cause inefficient rotor output. Wind turbine design is dictated by a combination of technology, prevailing wind regime, and economics.

Modeling and Wind Characterization Capa- Wind turbine manufacturers optimize machines to bilities. New computer simulation codes allow a deliver electricity at the lowest possible cost per kilo-wide array of system architectures to be designed watthour (kWh) of energy. Design efforts benefit from for various applications, while simulating results knowledge of the wind speed distribution and wind using local wind regimes for particular sites. Wind energy content corresponding to the different speeds characterization has reached a greater degree of and the comparative costs of different systems to arrive accuracy through the use of sophisticated instru- at the optimal rotor/generator combination. Optimizing mentation and monitoring capabilities. for the lowest overall cost considers design factors such as relative sizes of rotor, generator, and tower height.

Capability to Optimize WECS Design. Currently, Euro- For example, small generators (i.e., a generator with low pean turbine manufacturers supply the majority of the rated power output in kW) require less force to turn world market for utility-scale wind turbines.22 Enron than larger ones. Therefore, fitting a large wind turbine Wind Corporations Zond Energy Systems subsidiary rotor with a small generator will produce electricity was the fifth largest manufacturer worldwide in 1999 during many hours of the year (harvesting energy at with 9 percent of market. Zond is the only U.S. manu- lower wind speeds), but will capture only a small facturer presently manufacturing utility-scale turbines. portion of high-speed wind energy. Conversely, a large Zond's Z-750 turbine is the first U.S. machine in several generator will be efficient at high wind speeds, but years to be installed in large numbers in wind power unable to turn at low wind speeds. For a given turbine plants owned by independent power producers. Enron, rated output (e.g., 750 kW), rotor diameter can be a which purchased Zond Energy Systems in California in design variable, specifying a smaller rotor diameter for 1996 and German manufacturer Tacke in 1997, has plans turbines that will operate at sites with high wind speeds.

to develop a 1 MW next-generation turbine by 2002. In In addition, system design can be optimized further and addition, another U.S. company, The Wind Turbine performance efficiency can be increased with innovative Company, has announced similar plans for a 1 MW tower design, increased tower height to 50-70 meters machine. Both companies are developing their 1 MW- (which increases energy output), and lighter weight scale machines under DOE's Next-Generation Turbine turbines.

Development Program.

In general, most utility-scale wind turbines on the The general trend is toward wind turbines with maxi- market today are three-bladed systems that use asyn-mum power output of 1 MW or more. European firms& chronous generators and sophisticated controls to such as Danish companies Vestas and NEG Micon& monitor and regulate turbine operation in different currently have more than 10 turbine designs in the conditions and the quality of power delivered to the megawatt range with commercial sales. Due to grid. The following synopses provide a general over-decreasing wind development sites with adequate wind view of the current technologies utilized by the three regimes on the landmass, Europe has recently focused major utility-scale wind turbine manufacturers to on developing larger-than-megawatt turbines for off- optimize design. 24 NEG Micon has the simplest design shore wind farms. Because expensive foundations are while Zond the most complex design:

22 BTM Consult ApS, International Wind Energy Development-World Market Update 1999 (Aingkobing, Denmark, March 2000), p. 15.

23 Ibid., p. 15.

24 Information is based on manufacturer literature and on personal communications between Donald M. Hardy, PanAero Corporation (Lakewood, CO) and William R. King, SAIC, 1999.

82 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

NEG Micon. This design approach is the simplest lightning, and salty air in coastal regions. Modifications of the three major manufacturers; the basic design to enable operation in climates that are hotter or colder is about 20 years old. The blades have a fixed pitch than the design temperature operating range, operation and rotate at a constant speed (fixed rpm). Parts in coastal environments with salty air, and enhanced are bolted to the frame in a way that makes it easy lightning protection will add to the cost of the turbine to remove and replace a part. The turbine is con- system. The following discussion covers some of these nected directly to the electricity grid. The power modifications.26 flowing through the grid is used to maintain a constant turbine speed through electromechanical Ability to Operate Over a Range of Wind Speeds.

means. Currently available wind turbine designs enable reliable operation over a range of wind speeds. Rotor diameter

Vestas. This turbine has a variable pitch design; a can be modified from a standard diameter to one computer system controls blade pitch. Like the slightly larger for sites with low wind speeds or one NEG Micon machine, the turbine operates at a slightly smaller for sites with high wind speeds.

constant speed. The Opti-Slip technology incorpor-ated into the design allows slight speed variation Protecting Turbines in High Winds. Wind turbines are to relieve stress on the turbine.25 The Opti-Slip designed to operate up to a certain wind speed. Winds technology acts like a spring, allowing an increase above this speed could damage the turbine, so all in speed to relieve stress, then returning to a rated turbines are designed with a cut-off or shutdown mecha-speed. Like the NEG Micon turbine, the Vestas nism. The following examples discuss such mechanisms machine is connected directly to the grid without for each major manufacturer:

power electronics; speed is controlled electro-mechanically by the grid. NEG Micon. The turbine operates at a fixed

Zond. The Zond turbine has both a variable pitch rotation per minute (rpm). Its blade is shaped so blade design and a variable speed rotor and elec- that the energy conversion efficiency of the turbine tric generator design. Together, these design ele- drops at high speeds and the turbine stalls. The ments enable the turbine to convert wind energy to turbine has two braking systems. The tip of each rated turbine power output over a broader range blade turns 90 degrees at high centrifugal force to of wind speeds than possible with the constant exert drag that stops the blade. A disk brake speed generator design employed in the NEG system exerts hydraulic pressure to release the Micon and Vestas turbines. Because of the variable brake as long as electricity is available.

speed design, electricity from this turbine must flow through power conditioning equipment prior Vestas. Blade pitch control is used to stall the to entering the grid. The power conditioning turbine. Pitch control is achieved by feathering the equipment converts the variable frequency AC blades. Disk brakes also can stop the machine.

from the generator into DC, then (via an inverter) to 60 cycle AC that is also synchronous with the Zond. The blades have variable pitch control to grid. enable feathering at wind speeds above the rated 50 to 60 mph range.

Operational Characteristics Ability to Operate in Hot or Cold Climates. In hot Wind turbine manufacturers have developed basic wind climates, the transmission cooling system is upgraded, turbine designs that can be modified to optimize the and blades are made with epoxy resins that withstand turbine for reliable operation at a specific site. The wind heat and ultraviolet light. In cold climates, a heater is farm developer provides the manufacturer with site added to ensure that generator oil, transmission fluid, characteristics that will have an impact on the turbines and hydraulic systems are maintained at adequate capacity factor and on the reliability of turbine opera- operating temperatures. Black blades are advantageous tion. Factors include annual distribution of wind speed, as a deicing mechanism in cold climates because they annual variation in site temperature, frequency of absorb heat. For example, the NEG Micon turbine 25 For a given design, wind speeds beyond certain levels can damage the turbine. By varying the pitch (angle) of the blade tips at higher wind speeds, the blades will turn slower, reducing stress on the blade.

26 Unless noted otherwise, information in this section is based on manufacturer literature and on personal communications between Donald M. Hardy, PanAero Corporation (Lakewood, CO) and William R. King, SAIC, 1999.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 83

operates optimally in the -20ºC to 35ºC range.27 Below - farms to provide electric power in a form compatible 20ºC, a cold weather package is installed; above 35ºC, a with grid power quality. Different manufacturers have hot weather package is installed. different solutions, as seen in the following examples:

Ability to Operate in Coastal Salty Air. Paint sealants NEG Micon and Vestas. The design does not and nacelle designs that inhibit penetration of salty air require power electronics to maintain power are used to protect the turbine, generator, blades, and quality. The grid electromechanically controls the tower from corrosion. The sealant is baked on at the turbine to keep blade rotation speed at a fixed factory. rotation rate (e.g., rpm). This control solves the power conditioning problem but captures less Lightning Protection. Lightning is attracted to the tallest wind energy than do other solutions.

structure in an area, making wind turbines a prime target. Turbines are designed with a lightning protection Zond. Because the turbine design incorporates a system, and lightning damage may be included in the generator that is variable speed rather than con-warranty. For instance, Vestas offers Total Lightning stant speed, power electronics are required in the Protection in its 600 kW and 1.65 MW turbines, pro- design to maintain power quality. While power viding a route for the lightning to travel through the electronics add to system cost, they enable the tur-turbine to the ground.28 Vestas blades are protected by bine to convert more wind energy into electricity.

a 50 mm2 copper conductor, enabling lightning to travel along the blade without a significant increase in temper- Electronic controllers in modern wind turbines prevent ature. The lightning travels from the blade to the blade damage from power surges by constantly monitoring hub into the nacelle. The rear of the nacelle is protected grid voltage and frequency. For example, disturbances by a lightning conductor. Lightning protection in the in the grid may lead to islanding, which refers to a nacelle protects the wind vane and anemometer. Light- power outage in one part of the grid while the wind-con-ning is carried down the tower to the earthing system nected section of the grid is still supplied with power. If through two parallel copper conductors. The earthing disturbances are large enough to cause islanding, elec-system, which provides grounding for the turbine, con- tronic controllers automatically disconnect the turbines sists of a thick copper ring conductor placed one meter from the grid, and aerodynamic brakes are used to stop below the surface and one meter from the turbines the rotor. As connection to the grid is re-established, concrete foundation. The copper ring is attached to two electronic controllers protect the turbine from power diametrically opposed points on the tower and to two surges.

copper-coated earthing rods on either side of the foun-dation. Additionally, the turbine transformer is also An asynchronous or induction generator, which gener-protected. ates alternating current, is presently used for wind farm applications. These inexpensive generators may be Compatibility with Grid Power Quality. Power described as an electric motor that operates in reverse, quality refers to voltage stability, frequency stability, generating rather than consuming electricity. Wind and absence of various forms of electrical noise (e.g., cranks the rotor, which creates an electromagnetic force flicker or harmonic distortion) on the electrical grid. in the generator. The faster the rotor moves (greater than Power companies deliver three phases of alternating the generator stator's rotating magnetic field), the more current and power, each with a smooth sinusoidal current is induced in the generator and converted to shape, with few jags, breaks, or surges in any phase (less electricity, which is fed into the grid. One of the most than 9 percent harmonic distortion). Once the wind is important properties of an induction motor is that it will strong enough to turn the rotor and generator, the reduce its speed, as increases in wind speed lead to an turbine connects and is synchronized to the grid's phase. increase in torque on the motor, leading to less wear and Lack of synchronization may lead to rotor overspeeding tear on the gearbox. Another beneficial feature is that the and overtaxing of equipment components. The impact generator must be magnetized by power from the grid on the turbine could be costly equipment wear and tear. before it works, facilitating its synchronization with grid power.

Wind turbine designs and balance of system components are available currently that enable grid-connected wind 27 Personal communication between Jesper Michaelsen, Marketing Manager, NEG Micon USA, Inc., and William R. King, SAIC, 1999.

28 Vestas, manufacturer literature, 1999.

84 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Current Federal R&D To Improve WECS support testing of large-scale turbine components Performance and Reliability such as generators, rotors, data acquisition systems, and controls. The ART Test Bed is being The objective of the U.S. Department of Energy (DOE) used in FY 2000 for the Long-Term Inflow and Wind Energy Program is to enable the U.S. wind Structural Testing Program (LIST), which aims to industry to complete the research, testing, and field understand inflow and resulting loads on turbines.

verification needed to fully develop cost-effective and reliable advanced wind technology.29 Activities are Materials, Manufacturing, and Fatigue. This classified under one of three research areas: applied research aims to reduce capital and maintenance research, turbine research (which includes large, utility- costs by improving blade strength and reliability scale turbines), and cooperative research and testing. during the manufacturing process. Activity areas The cooperative research and testing activity offers the include the development of advanced manu-wind industry the facilities to test their turbines and facturing techniques and blade fabrication and turbine components and provides a turbine certification testing.

test program. This activity helps the industry control costs by limiting the extent to which turbine manu- Avian Research. This research uses analyses of facturers in the United States need to invest in and staff bird deaths at current wind turbine sites to such facilities. develop strategies to avoid bird fatalities. Research has addressed impacts of wind turbines on Applied Research.30 The Applied Research Program individual birds protected under legislation such seeks to understand the basic scientific and engineering as the Migratory Bird Treaty Act of the Endan-principles that govern wind technology and underlie the gered Species Act, as well as impacts on specific aerodynamics and mechanical performance of wind species. Research has been conducted to survey turbines. The program also seeks to improve the cost what species use a wind resource area, what part and reliability of different wind turbines by conducting of the site they use, and when they use it. Research applied research in the following areas: also focuses on studies of factors that may affect the impact of wind turbines on birds. Factors

Aerodynamics and Structural Dynamics. The include analyses of the impact of topography, objective is to lower turbine cost and increase weather, habitat fragmentation, urban encroach-turbine life, possibly by developing lighter, more ment, habitat loss, species abundance, distribution, flexible turbines. Such turbines may be made bird behavior, and turbine type and location.

possible through an understanding of complex Preliminary results of survey and factors research wind/wind turbine interactions and using such indicate that wind turbines can be installed with-information to improve design codes. Data for out causing any biologically significant impacts on such analyses come from both highly instrumented bird species.

experimental wind turbines and turbine testing in the NASA Ames Research Center low turbulence Turbine Research.31 The objective of this research is to wind tunnel. The advantage of the low turbulence assist the U.S. wind power industry in developing wind tunnel is that it enables three-dimensional competitive, high-performance, reliable wind turbine testing of the dynamic response of full-scale wind technology for global energy markets. The program turbines to steady wind inflow, as the tunnel funds competitively selected industry partners in their eliminates normal atmospheric turbulence. development of advanced technologies. Wind turbines in various sizes from 10 kW to more than 1 MW are con-

Systems and Components. The objective of this structed and tested.

research is to advance the design of wind turbine components and subsystems beyond the current Currently, some of the research projects include: a Near-generation. The Advanced Research Turbine (ART) Term Research and Testing contract with Zond Energy Test Bed tests innovative approaches to component Systems; Next-Generation Turbine Development con-design. The highly instrumented ART turbines also tracts with the Wind Turbine Company and Zond 29 U.S. Department of Energy and National Renewable Energy Laboratory, Wind Power Today, DOE/GO-102000-0966 (Washington, DC, April 2000), p. 28.

30 Ibid., pp. 29-30.

31 U.S. Department of Energy and National Renewable Energy Laboratory, Wind Power Today, DOE/GO-102000-0966 (Washington, DC, April 2000), p. 31-32.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 85

Energy Systems; Small Wind Turbine Projects with Ber- operating life of the turbine. Generally, maintenance gey Windpower Company, WindLite Corporation, and costs are low for new turbines and increase as the World Power Technology; and a cold weather turbine turbine ages. Failure of wind turbine system components development contract with Northern Power Systems. can be characterized by a relatively higher initial rate of failure followed by a lower failure rate through most of Cooperative Research and Testing. The Federal Gov- the turbine's design life until components begin to wear.

ernment, through the National Wind Technology Center During the initial period, assembly defects are detected at the National Renewable Energy Laboratory, offers and rectified. Commonly, wind turbines are sold with a cooperative research, testing and certification, and 2- to 5-year manufacturer warrantee covering the cost to standards programs to wind turbine manufacturers.32 repair these design-related breakdowns.34 Wind turbine Without these programs, the industry would bear the models are available today for which minimal initial costs, which would be reflected in a higher wind turbine failure rate problems may be expected because the cost. Cooperative research enables turbine manufac- current turbine design is (1) a variation of past designs turers to leverage their R&D efforts with related Federal that have proven successful in the field and (2) manu-efforts and ensures, through commitment of manufac- factured with adequate quality assurance procedures.

turer resources, that R&D worthwhile to them is The reliability of new turbine designs improves over pursued. Wind turbine blade testing includes three types time as field experience enables resolution of technical of tests&ultimate static strength, fatigue, and non- problems. Field experience is particularly important for destructive&to identify and correct problems before more complex designs, including those that deviate going into full-scale production. Modal testing provides more from past design generations.

useful information about the structural dynamic charac-teristics of a wind turbine system. This information is The average annual maintenance cost for newer turbines used to avoid designs that are susceptible to fatigue- is approximately 1.5 percent to 2.0 percent of the cost of related failure and excessive vibrations. Testing of full- the machine.35 Most of the maintenance expenses are scale wind turbine drivetrains on a 2.5 MW Dynamo- associated with the routine service of turbines. Wind meter Test Stand located at NREL was initiated in mid- turbine manufacturers and service contractors certified 1999. The dynamometer can test turbine drivetrains as to perform maintenance on a manufacturers turbines large as 2 MW both to identify weak points in the design can be contracted on an annual basis to perform planned and to measure the lifetime of systems. Receipt of certifi- preventive maintenance. For example, the cost of a cation services enable U.S. manufacturers to show that preventive maintenance contract for a 750 kW turbine their turbines meet international standards; such certifi- ranges from $12,000 to $14,000 per year, per turbine.36 cation is needed for U.S.-made turbines to sell in many Maintenance on a 600 kW or 660 kW turbine can be per-foreign markets. formed for a comparable cost, $12,500 per year, per turbine.37 Comparable maintenance on a 1.65 MW tur-Operation and Maintenance for Wind Farm bine would increase to $18,000 per year, per turbine.38 Some analyses state the cost of preventive maintenance Turbines in terms of dollars per kilowatthour of electricity output.

Modern wind turbines are designed for about 120,000 When expressed in these units, turbines with higher hours of operation over a 20-year lifetime.33 During this annual kilowatthours of electricity output have lower period, planned preventive maintenance and breakdown per-kilowatthour maintenance cost. A turbine with maintenance are performed. Additionally, system com- higher electricity output either has a higher maximum ponents may be replaced as their performance degrades; kilowatt output rating or a higher capacity factor. Such such replacements also are performed to extend the analyses have stated a maintenance cost of around $0.01 32 Ibid., p. 32.

33 Danish Wind Turbine Manufacturers Association, Operation and Maintenance Costs for Wind Turbines. See website http://www.windpower.dk/tour/econ/oandm.htm (October 23, 2000).

34 Personal communication between Donald M. Hardy, PanAero Corporation (Lakewood, CO) and William R. King, SAIC, 1999.

35 Danish Wind Turbine Manufacturers Association, Operation and Maintenance Costs for Wind Turbines. See website http://www.windpower.dk/tour/econ/oandm.htm (October 23, 2000).

36 Personal communication between Donald M. Hardy, PanAero Corporation (Lakewood, CO) and William R. King, SAIC, 1999.

37 Personal communication between Soren Christensen, Project & Sales Coordinator, Vestas-American Wind Technology (North Palm Springs, CA), and William R. King, SAIC, November 1999.

38 Ibid.

86 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

per kWh.39 Larger generation capacity turbines are equipment maintenance is usually contracted to the serviced at the same frequency and cost as smaller ones, utility company.

which results in a lower maintenance cost per installed kW; however, over time stresses and strains inherent in Electrical Safety Maintenance operation of larger capacity turbines cause more wear and tear on system components, leading to accelerated Regular maintenance of the turbine's electrical systems component replacement. and a complete set of replacement parts minimize down-times caused by electrical faults and ensure operational Additionally, wind farms benefit from the economy of efficiency. A maintenance program may consist of scale related to semi-annual maintenance visits, admin- monthly inspection of breakers, security, and battery istration, and inspection. Wind farm operators increase voltages; annual checks of relay settings, oil levels, the life of a turbine by replacing certain components, ground connections, and corrosion; 2-year interval such as rotor blades, generators, and gearboxes, which testing of protection mechanisms, oil quality and levels, are subject to more wear before the end of the turbines and high voltage circuit breakers; and 4-year inspections design life. The price of replacement components is of all the switchgear, the grid transformer, and all usually 15 percent to 20 percent of the price of the wiring.

turbine and can extend the life of the turbine by the same or longer amount.40 In addition, since some components need to be ordered, carrying a comprehensive set of replacement parts may be the difference between minor downtime or shutdown Planned Maintenance of the entire wind farm to await delivery. For this Planned maintenance covers all preventive maintenance, reason, a full set of protection relays, transformer including routine checks, periodic maintenance, periodic windings, bushings, moving contacts, fuses, and gaskets testing, blade cleaning, and high voltage equipment must be stocked on-site.

maintenance. Routine checks are performed monthly for every machine using a checklist that includes inspection Breakdown Maintenance of the gearbox and oil levels, inspection for oil leaks, observation of the running machine for unusual drive The frequency of wind turbine shutdowns or break-train vibrations, brake disc inspection, and inspection of downs is affected by operational factors and design all emergency escape equipment. complexity. More major system faults are generally categorized as human error, acts of God, design faults, Periodic maintenance takes place approximately every or system component wear and tear. Operational factors 6 months and includes checking the security of all that affect breakdown frequency include overspeeding, supports and attachments, high-speed shaft alignment, excessive vibration, low gearbox oil pressure, yaw error, brake adjustment and pad wear, and yaw mechanism pitch error, unprompted braking, synchronization performance; greasing bearings; inspecting cable ter- failure, loss of grid, and loss of batteries. A significant minations; and replacing oil filters. For pitch-regulated portion of wind turbine maintenance events can be machines, the pitch calibration is also checked. In detected by wind turbine system controllers, which can addition, this may be the time to replace components sense problems such as loose connections due to that are known to fail after a few years of operation, vibration or defective sensors.

such as anemometers, wind vanes, and batteries.

Wind turbine designs, evolving with new research and Periodic testing of the overspeed protection system development breakthroughs, have in some cases become should be conducted to ensure proper operation of this more complex. Initially, a turbine that incorporates feature. Blade cleaning should be a maintenance con- several new design concepts may experience an increase sideration when the performance of the turbine is in breakdown frequency when compared to older affected due to dirt buildup; however, because of the proven turbine designs. Breakdowns may be caused by high cost of equipment for accessing the blades, this task the design of a specific part or by problems that arise should be evaluated for cost-effectiveness. High voltage when parts incorporated into the new design do not 39 U.S. Department of Energy (Office of Utility Technologies) and Electric Power Research Institute, Renewable Energy Technology Characterizations, TR-109496 (Washington, DC, December 1997), p. 6-13. Danish Wind Turbine Manufacturers Association, Operation and Maintenance Costs for Wind Turbines. See website http://www.windpower.dk/tour/econ/oandm.htm (October 23, 2000).

40 Danish Wind Turbine Manufacturers Association, Operation and Maintenance Costs for Wind Turbines. See website http://www.windpower.dk/tour/econ/oandm.htm (October 23, 2000).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 87

function together as a system. Field experience enables Improvements in the aerodynamics of wind technical problems to be detected, facilitating their turbine blades, resulting in higher capacity factors resolution through additional development. and an increase in the watts per square meter of swept area performance factor.

Beyond the initial period of resolving technical problems in a new turbine design, more complex machines may Development of variable speed generators to experience higher expenditures on periodic planned improve conversion of wind power to electricity maintenance and higher replacement part costs. over a range of wind speeds.

Expected higher expenditures do not necessarily reflect on the reliability of the turbine; they reflect more on the Development of gearless turbines that reduce the cost of maintaining and replacing complex parts. The on going operating cost of the turbine.

cost-effectiveness of the turbine depends on such costs being covered by the incremental electricity production Development of lighter tower structures. A by-benefit that rationalizes the new design. product of advances in aerodynamics and in generator design is reduction or better distribution In Europe, gradual changes in wind turbine design of the stresses and strains in the wind turbine.

during the past 20 years have been accompanied by Lighter tower structures, which are also less testing and certification and by the hours of field expensive because of material cost savings, may be experience needed to demonstrate wind turbine relia- used because of such advances.

bility. This process of turbine design, testing, cer-tification, and field experience has resulted in the NEG Smart controls and power electronics have enabled Micon and Vestas wind turbines deployed in wind remote operation and monitoring of wind turbines.

farms currently being developed in the United States Some systems enable remote corrective action in and worldwide. In the United States, the U.S. Depart- response to system operational problems. The cost ment of Energy, the National Renewable Energy of such components has decreased. Turbine Laboratory, and Underwriters Laboratories, Inc., have designs where power electronics are needed to worked together to provide comparable turbine testing maintain power quality also have benefitted from and certification for U.S. wind turbine companies.41 a reduction in component costs.

Summary In the United States, the Zond Z-750 series turbine represents a very innovative but less gradual design Research and development throughout the past 20 years change. Enron Wind Corporation wind farms, which use has resulted in a current generation of utility-scale wind the Zond Z-750 technology, address the risk of the turbines, with maximum electricity generating capacity design innovation with performance contracts that often exceeding 500 kW per turbine, designed for about guarantee turbine electricity production, in addition to 120,000 hours0 days 0 hours 0 weeks 0 months of operation over a 20-year lifetime. In the power curve and reliability guarantees normally United States, wind farm development activity in 1999 included in wind turbine performance contracts. The was motivated by the June 1999 expiration of the Federal results of R&D have been incorporated into utility-scale production tax credit, and dominated by installation of wind turbine design more gradually in Europe, followed utility-scale turbines manufactured by NEG Micon and by operation in wind farms to assess reliability over Vestas, both Danish firms, and by Zond Energy Systems, time.

a subsidiary of Enron Wind Corporation, a U.S. firm.

Research and development for utility-scale turbines has Near-term R&D efforts are expected to continue in been directed toward increasing the amount of wind directions that increase the efficiency with which wind energy that a turbine can convert into electricity for the turbines convert wind energy to electricity. For instance, lowest amount of capital investment and the lowest on- researchers report that further optimization of blade going operating cost. Following are examples of the design is possible.42 Taller towers and rotor/generator R&D efforts that have contributed to current utility-scale systems with maximum power ratings exceeding turbine technology: 1 MW will continue to be improved. Other areas of 41 National Renewable Energy Laboratory, Certification Program Opens Markets to U.S. Turbines, DOE/GO-10099-820 (Golden, Colorado, June 1999), p. 16.

42 U.S. Department of Energy and National Renewable Energy Laboratory, Wind Power Today, DOE/GO-102000-0966 (Washington, DC, April 2000), p. 31-32.

88 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

development that affect turbine cost include advanced success. Eligible projects receive a tax credit of 1.5 cents manufacturing methods and use of alternative, more per kilowatthour of electricity produced, adjusted for cost-effective materials for turbine system, and tower inflation, for the first 10 years of the projects life. Even fabrication. when levelized over the full life of a project, this benefit is significant. Immediately prior to the expiration of the The result of turbine R&D has been a reliable utility- production tax credit, a rush of projects came on line in scale wind turbine generator that can be optimized for spring 1999. Since then, development has continued, but operation in a variety of wind farm locations. For at a slower pace. This tax credit was valued at more than example, annual wind farm capacity factors of 28.5 $20 million for 1998, virtually all of which was for percent to 32 percent have been achieved in Denmark wind.44 and the United States, respectively, and capacity factors of 35 percent to 38 percent are projected for wind farm EPACT also created the Renewable Energy Production capacity that was recently installed in Minnesota and Incentive (REPI). This incentive is paid to wind gener-Iowa, respectively (Table 4). ation facilities owned by State and local government entities and not-for-profit electric cooperatives that are tax exempt. Qualifying facilities are eligible for annual The Changing World for Wind Power incentive payments of 1.5 cents per kilowatthour (1993 dollars and indexed for inflation) for the first 10 years of In addition to technological improvements in wind operation subject to the availability of annual appropri-turbines, governmental and private efforts to increase ations in each Federal fiscal year of operation. REPI the Nations consumption of renewable-based electricity payments for fiscal year 1998 production were $4 have grown. Because wind energy is generally the most million, of which wind accounted for about $32,000. The economically competitive, widely available renewable majority of the funds were used for biomass digester electricity source other than hydropower, some of these gas, wood waste, and landfill methane.

efforts have had their greatest impact on wind power.

Another Federal incentive is research and development expenditures and efforts. Applied research and develop-Federal Incentives ment (R&D) activity is considered a support program A wide variety of Government actions can be used to because, when successful, it reduces the capital and/or influence energy markets and achieve Government operating costs of new products or processes. The objectives. These actions, broadly called incentives, mission of the Wind Energy Systems Program is to include taxes, payments, trust funds, insurance, low-cost establish wind energy as a regionally diversified, cost-loans, research and development, and varieties of effective power generation technology, through a coor-regulation. For a more detailed discussion of issues dinated research effort with industry and utilities that surrounding incentives for renewable energy, see the will minimize technical and institutional risks for U.S.

article, Incentives, Mandates, and Government Pro- companies competing in domestic and international grams for Promoting Renewable Energy, contained in markets. In addition to improving existing turbines, this report. DOE and industry are improving particular turbine components. The National Renewable Energy Labora-The most significant Federal incentive for wind power is tories (NREL) and Sandia National Laboratories have the production tax credit established by the Energy worked since 1994 with industry on cost-shared projects Policy Act of 1992 (EPACT). This credit expired in June to develop the cutting-edge wind turbine components 1999, but now has been reinstated and applies to profit needed to create larger, more cost-effective turbines.

wind and closed loop biomass projects in operation by Already since 1980, the cost of wind generation has December 31, 2001.43 This type of incentive (when declined from 35-40 cents per kilowatthour to a pro-compared to an investment tax credit) rewards energy jected 6 cents in 2000.45 The DOE Wind Energy Program production and thus supports project performance/ was funded at around $33 million in fiscal year 2000.46 43 Biomass projects must utilize biomass grown exclusively for energy production.

44 1993-2004: Office of Management and Budget, Analytical Perspectives, 2000 (Washington, DC, 1999).

45 Energy Information Administration, Annual Energy Outlook 2000, DOE/EIA-0383(2000), National Energy Modeling System run AEO2k.d100199A.

46 U.S. Department of Energy, Office of Chief Financial Officer, FY 2001 Budget Request to Congress - Budget Highlights, DOE/CR-0068-8 (Washington, DC, February 2000).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 89

Federal Electric Power Industry Restructuring energy may differ substantially among the States. Net metering is used by a number of States to support rela-Competition in the electric power industry holds tively small facilities, so it is generally more applicable promise for more efficient operations at generating for solar energy than for wind. All of these activities are facilities and a reduction in costs, which should lead to documented in detail for each State in Appendix A.

lower electricity prices. However, concern has arisen that higher cost, but environmentally friendly, energy Other State financial incentives support wind energy:

sources (i.e., renewables) will lose out to less environ-mentally friendly fuels used for producing electricity Net Metering. Provisions vary by State and utility, having a low short-run marginal cost. To protect the but usually apply only to very small generators environment, Federal and many State restructuring that typically use solar or wind energy. This plans include incentives to promote the use of renewable system usually permits a customer operating a energy. Hence, competition and the restructuring of the small generator to purchase extra electricity when electric power industry, when accompanied by environ- needed. Also, any excess power at the end of the mental provisions, could be a push for new renewable month can be sold back to the utility. Pricing energy development. schemes vary by individual utility circumstances.

The administration and members of Congress have pro- Accelerated Depreciation. For example, in Minne-posed a number of plans to restructure the electric sota this incentive is modeled after the Federal power industry. Efforts have been expended to get a income tax Modified Accelerated Cost Recovery consensus legislative package out of Congress, but no Schedule (MACRS) schedule for depreciation of agreement is forthcoming, because so many differences equipment, thus improving the owner/operators still remain.47 The administrations latest electric tax position.50 industry competition plan, as of April 15, 1999, would provide for phasing in retail competition by 2003 and Sales Tax Exemption. This type of incentive may support for renewable energy through regulatory exempt from sales tax all of the cost of wind energy mechanisms, including a renewable portfolio standard equipment and all materials used to construct (RPS), public benefit fund (PBF), and net metering.48 wind energy systems. Alternatively, the sales of wind power itself may be exempt from sales tax.

State Incentives

Property Tax Exemption. This incentive excludes With Federal legislation promoting electric wholesale from property taxation all or part of the value competition in place, 25, or just half the States, have added by wind energy systems.

comprehensive restructuring policies in effect (Table 5).

Many of the States with plans to implement retail Special Grants. These grants may be given for competition also have regulatory mechanisms to support research and development of wind energy renewable energy. As with the Administrations pro- resources or technology.

posed electric competition plan, the most important regulatory mechanisms for support of renewable energy Loans. States may offer low interest loans under are the RPS, PBF, and net metering. Currently, 10 States certain conditions to wind project developers.

(Arizona, Connecticut, Maine, Massachusetts, Nevada, However, frequently these loans are restricted to New Jersey, New Mexico, Pennsylvania, Texas, and small projects, so the benefit is limited.

Wisconsin) have an RPS in place.49 Thirteen States (Cali-fornia, Connecticut, Delaware, Illinois, Massachusetts, Some of these provisions have been in place a number of Montana, New Jersey, New Mexico, New York, Oregon, years, while others have recently been enacted. In the Pennsylvania, Rhode Island, and Wisconsin) use a early years, investment tax credits were popular but system benefits charge (SBC) to support a PBF. The pro- later found flawed as they rewarded development, not visions within a States RPS or SBC to support renewable performance.

47 For an Electric Utility Restructuring Weekly Update see the U.S. Department of Energys website:

http://www.eren.doe.gov/electricity_restructuring/weekly.html (summer 2000).

48 For more details on the administrations proposed Comprehensive Electricity Competition Act, see website http://www.doe.gov/policy/ceca.htm (summer 2000).

49 As of summer 2000.

50 Refers to a 5-year, 200-percent, double declining balance, accounting method.

90 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Table 5. Renewable Incentives and Support Programs by State and Status of Implementing Electric Power Industry Restructuring Renewable System Portfolio Benefits Green States Standard Charge Pricing a With Comprehensive Restructuring Policies:

Arizona . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Arkansas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

California . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x Connecticut . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x Delaware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x District of Columbia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Illinois . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Maine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Maryland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Massachusetts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x Michigan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Montana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x Nevada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x New Hampshire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

New Jersey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x New Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x x New York . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Ohio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Oklahoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Oregon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x Pennsylvania . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x Rhode Island . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Texas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x Virginia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

West Virginia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Remaining States: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Alabama . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Colorado . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Florida . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Georgia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Hawaii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Idaho . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Indiana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Iowa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Kansas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Kentucky . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Louisiana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Minnesota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Mississippi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Missouri . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Nebraska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x North Carolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

North Dakota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x South Carolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

South Dakota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Tennessee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Utah . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Vermont . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Washington . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Wisconsin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x x Wyoming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 13 22 a

Utility programs available to at least some customers in the State. Some programs start in 2000.

Sources: Electricity Restructuring Status: Energy Information Administration, Status of State Electric Industry Restructuring Activity as of May 2000, Website: http://www.eia.doe.gov/cneaf/electricity/chg_str. Renewable Portfolio Standard and System Benefit Charge: Wiser, R., Porter, K. and Bolinger, M., Lawrence Berkeley National Laboratory. Comparing State Portfolio Standards and System-Benefits Charges Under Restructuring, Memorandum (August 23, 2000) to various officials of the U.S. Department of Energy and the National Renewable Energy Laboratory. Green Pricing: U.S. Department of Energy Website: http://www.eren.doe.gov/greenpower (June 2000).

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 91

Other Support Improved wind technology Green pricing/marketing, which lets renewables com- Federal production tax credits pete on a basis of consumer demand, also provides support for development of renewable energy, including Presence of a State law mandating development of wind power. Proponents of this type of support argue renewable and/or wind capacity that as consumer awareness of the benefits of renewable energy is raised, they may choose to consume more Various State incentives examples, of which are tax renewable energy even if it requires paying a small advantages (accelerated depreciation, property and premium to do so. So far, these programs can be sales tax exemptions), low interest loans, grants, characterized as lively, if small in impact. By the end of access laws, net metering, and green pricing. These 1999, 50 utility green pricing programs were in place incentives currently are available in Minnesota across the United States.51 Premiums for wind power and/or Iowa.

range from a low of 1 cent per kilowatthour to upwards of 5 cents per kilowatthour in a handful of cases.52 Texas has several moderately sized projects that According to data compiled by the National Renewable together add up to more than 140 MW of added new Energy Laboratory, green pricing/marketing activities capacity. These projects include McCamey, Texas (75 resulted in the addition of nearly 100 MW of new wind MW), Culberson County, Texas (30 MW), and Big capacity by July 2000.53 Spring, Texas (35 MW). Projects were constructed using the federal production tax credit and in response to the Developments demand from green pricing programs. Since the time commitments to these projects were made, Texas passed What developments have these incentive and electric restructuring (with retail competition to begin in 2002) power industry restructuring policies spawned? and also a renewable portfolio standard, both of which Industry sources estimate that more than 900 MW of will affect the future. Other States, such as California, new or repowered wind capacity was constructed in Colorado, Oregon, and Wisconsin, are in the process of 1999 (Table 1). Where and why did this development developing projects at least in part as a result of green take place? States with new capacity include Alaska, pricing programs.

California, Colorado, Iowa, Kansas, Minnesota, Nebras-ka, New Mexico, Texas, Wisconsin, and Wyoming (See Appendix A.). Capacity additions in these States vary in Conclusions significance. Iowa, Minnesota, and Texas had the most capacity added, States, followed by Colorado, Wisconsin Although the economics of wind energy have improved and the others, including California, which has a over the last decade, wind energy is generally not yet significant repowering program. competitive with traditional fossil fuel technologies.55 Enactment of State electric restructuring legislation that Together, Iowa and Minnesota installed two large wind includes support for renewable energy and the rein-projects in 1999: Storm Lake, Iowa (193 MW), and Lake statement of the federal production tax credit will Benton II, Minnesota (104 MW).54 Neither of these States provide an impetus for wind energy. Until wind energy has yet passed restructuring legislation. Thus, several is competitive, the future for wind energy is likely to be primary factors influenced the projects: in those States providing additional support to renew-able energy. This support may take the form of financial

Availability of good wind resources and land incentives, regulatory programs (such as a renewable 51 R. Wiser, M. Bolinger, E. Holt, Lawrence Berkeley National Laboratory, Customer Choice and Green Power Marketing: A Critical Review and Analysis, in Proceedings of ACEEE 2000 Summer Study on Energy Efficiency in Buildings (Pacific Grove, California, August 2000).

52 For recent or more detailed information, see the U.S. Department of Energys website: http://www.eren.doe.gov/greenpower .

53 Lori Bird and Blair Swezey, National Renewable Energy Laboratory, Estimates of Renewable Energy Developed to Serve Green Power Markets, July 2000 on the Department of Energys green power website:

http://www.eren.doe.gov/greenpower/new_gp_cap.shtml (July 2000).

54 Minnesotas other large wind project was the Lake Benton I facility with 107 MW of capacity, which came on line in 1998.

55 For analysis of issues related to integrating renewable energy and wind power into the U.S. energy supply, see Energy Information Administration, Annual Energy Outlook 2000, DOE/EIA-383(2000) (Washington, DC, December 1999).

92 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

portfolio standard or system benefits charge), or green Minnesota and Texas should continue to support wind pricing, in which wind will be competing for benefits energy. Further advances in technology and per-with other renewable energy sources. Electric retail com- formance are expected to lower costs and improve petition, without the States support of renewable project economics, making wind more competitive with energy, could be a setback to the penetration of wind other energy sources, renewable and nonrenewable.

energy. Commitments such as those evident in Energy Information Administration/ Renewable Energy 2000: Issues and Trends 93

Appendix A. State Wind Profiles: A Compendium This appendix presents assessments of State-level wind Alaska has two small wind facilities in rural areas. The energy programs.56 Each assessment begins with the one in Kotzebue began with 500 kilowatts (kW) of major issue likely to affect wind energy: the status of capacity installed and has plans for future expansion.

electricity restructuring and implementation of retail This project was funded in part by a grant from DOEs competition in each State.57 The assessments follow with Wind Turbine Verification Program. A small 225 kW information about State incentives and support from facility is also located on St. Paul Island. Following the green power programs available for wind power (in success in Kotzebue, other remote communities are addition to possible Federal incentives discussed earlier) proposing to build new wind facilities. Wales, Alaska, and ends with the status of wind power development planned to have a new 100 kW facility on line in 2000.

through 2000. A list of sources of information follows at the end of the appendix. This list can be used to obtain Arizona. Arizona began retail competition for some of more up to date information as needed. its consumers in 1999. This phasing in was to continue until completion in January 2001. In April 2000 the Alabama. Because Alabama is a low-cost State and for Arizona Corporation Commission approved a renew-other reasons, action on restructuring has been slow to able portfolio standard that will require utilities and progress. In February 2000 the Public Services Commis- other electricity providers to derive 1.1 percent of their sion scheduled hearings to address two key issues: energy from renewable sources (including wind) by whether the electric power industry restructuring 2007. In turn, 50 percent of that must come from solar towards competition is in the best interests of consumers energy. Funds from the existing system benefits charge and what the regulatory/jurisdictional role of the Public may be used for renewable portfolio standard com-Services Commission would be in a market-based pliance costs.

system. Alabama has a green pricing program starting in 2000 that could promote wind energy when available. Arizona has other incentives for renewable energy, Alabama has no existing identified wind capacity and no possibly including wind. However, they are generally new wind capacity was planned for 2000. directed towards fairly small operations. Among them is a Qualified Environmental Technology Facilities Alaska. In May 1999, the State Public Utility Commis- Credit. This incentive allows a credit toward the sion received a report which investigated the possibility personal or corporate income taxes in the amount of 10 for deregulation in Alaska. Included in the report was percent of the cost of construction of a qualified consideration of creating retail pilot programs, en- environmental technology manufacturing, producing or couragement of power trading markets, and creation of processing facility.

a central dispatch point and an Independent Systems Operator (ISO). An adjunct effort by the State Senate has A personal income tax provision allows a 25 percent tax reorganized the Public Utility Commission (PUC) into credit on the cost of a solar or wind energy device up to the Regulatory Commission of Alaska and a panel of $1,000. The Revolving Energy Loans for Arizona (RELA) five new commissioners. In April 2000 a Senate bill was Program provides loans up to $500,000 to companies introduced that, if passed, would implement retail that manufacture renewable equipment or acquire it for choice in the rail belt (Anchorage and Fairbanks) by use in their own processes. The Solar and Wind Energy September 2001. Equipment Tax Exemption of up to $5,000 applies to solar and wind energy equipment. Finally, Arizona 56 Note: Some States may have wind turbines that are so small or so dispersed they are not counted in the usual surveys of wind capacity. This could include turbines used for water pumping on ranches or farm land. In this analysis these States are described as having no identifiable wind generating capacity even though they may have a small amount.

57 Information for this appendix was taken from various websites, and is current as of summer 2000.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 95

has net metering provisions depending on the utilitys the separate, private Green-e certified program. To service area. Arizona Public Service Company permits be recognized by the green-e program, a product net metering for facilities under 10 kW, while Tucson must have 50 percent or more renewable content Electric Power Company allows net metering for and meet other requirements.58 Many of these facilities under 100 kW. include wind power explicitly in their renewable generation portfolio. Two municipal utilities, Los To date, Arizona has no identified wind facilities and Angeles Department of Water and Power and the none were planned for 2000. City of Palo Alto, have green pricing programs that promote wind energy.

Arkansas. The status of deregulation is that Senate Bill (SB) 791 will restructure Arkansas electric power Research and Development. The Public Interest industry and allow retail access by January 2002. In Energy Research Program (PIER) supports the December 1999 the Public Service Commission began public interest research development and work on a series of reports to facilitate implementation demonstration that utilities were required to do of retail competition. No incentives for wind power exist before deregulation. It makes $62 million available and there are no existing or planned wind facilities annually through 2001.

identified for 2000.

California has a mature wind industry. At the end of California. The process of restructuring began in 1998, EIA estimates that Californias wind net summer September 1996 when the California State legislature capability stood at 1,487 megawatts (MW).59 A number passed Assembly Bill (AB) 1890 to begin restructuring of new and repowered projects with capacity totaling Californias electric power industry. The retail electricity 290 MW came on line in 1999 and nearly 210 MW more market opened officially for all consumers in California were planned for 2000. For details, see the American on March 31, 1998. The following measures support Wind Energy Associations website: http://www.awea.

renewable energy: org/projects/california.html. Further into the future, the new technologies account of the renewable set aside

Renewable Setaside. AB 1890 also established a program is expected to support development of some system benefits charge of 0.7 percent on all additional new wind capacity.

electricity sold by Californias Investor Owned Utilities. Funds (estimated at total of $540 million) Colorado. Several bills to allow retail competition and would be used to support development of renew- restructure the electric power industry were introduced able energy during a 4-year transition period to in the legislature in 1998. None, however, have passed open competition beginning in 1998. Legislation the State legislature. The Colorado Electricity Advisory extending the setaside for ten years through Panel, created by SB 152, released a final report in January 1, 2012 was signed into law in September November 1999. The majority of the panel opposed 2000. It authorizes collection of $135 million per restructuring and retail competition, because of their year for investment in renewable sources. concern that Colorado already has low electricity rates, and that prices might rise under open competition. In

Net Metering. Solar and wind installations equal addition, it is believed that rate impacts would be to or under 10 kW in capacity are eligible. disproportionately shared among classes of consumers with low-income, fixed income, rural, residential and

Green Power. Any number of green power small consumers seeing the greatest increases. On programs are supported by the customer side another front, the Colorado Public Utilities Commission account portion of the setaside program mentioned adopted rules in January 1999 which requires investor-above. The customer side account provides rebates owned utilities (IOUs) to itemize the fuel sources used of up to 1.5 cents per kilowatthour to customers for generated and purchased electricity; thus, who purchase energy from renewable electric ser- increasing public awareness. Unbundled billing has been vice providers registered with the Energy Commis- implemented and the utilities provide this information sion. Rebates for industrial customers are limited to customers twice a year. Also, Colorado has net to $1,000 per year. Renewable products may be metering for qualified facilities equal to or less than 10 marketed using these rebates and/or as part of kW in capacity.

58 For details, visit the Green-e website: http://www.green-e.org (summer of 2000).

59 Energy Information Administration, Renewable Energy Annual 1999 With Data for 1998, DOE/EIA-0603(99) (Washington, DC, March 2000), p. 96.

96 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Colorado has one investor owned utility with a green In February 1999, the Public Services Commission ruled pricing program. To encourage development of wind that investor-owned utilities must disclose the sources of resources, Public Service Company of Colorado (PSCo) generation and purchased power to consumers. The has opened its green power program, WindSource. As Florida Energy Efficiency and Conservation Act of 1980 customers sign up to buy electricity from wind power, requires the Florida Public Service Commission to PSCo is developing the needed capacity. So far in encourage the use of renewables, including wind.

response to demand, PSCo has put more than 16 MW of Florida has no identified wind facilities and no new wind capacity in operation in Ponnequin, Colorado. In facilities were planned for 2000.

addition, five municipal utilities and three electric cooperatives have green pricing programs to promote Georgia. In early 1998 Georgias PSC issued a report that wind energy. investigated electric industry restructuring and made recommendations. No further action has been taken Connecticut. The State of deregulation is that phasing in since then. Georgia has no incentives for renewable of retail competition began in January 1, 2000. The law energy. It has no identified wind power facilities, but a also includes a 7 percent renewable portfolio standard to small 1.98 MW facility was planned for 2000.

be met by 2009 and a provision for establishing a system benefits charge rising to 0.1 cents per kilowatthour Hawaii. An April 1999 legislative resolution provided (kWh) to support renewable technologies. Fourteen mil- that the PUC submit (prior to the 2000 legislative lion dollars is budgeted for the fund in 2000. Connecticut session) a report on restructuring and competition in has net metering for renewable facilities under 100 kW. electric markets. Hawaii offers an income tax credit Connecticut has no wind facilities and none were allowing individuals and corporations a credit of 20 planned for 2000, although Connecticut entities may percent of the cost of equipment and assembly of a invest in out-of-State wind projects, power from which residential or non-residential wind energy system to be would be eligible for complying with the State RPS. applied in the year the system was purchased and placed in operation. There is no limit on the total Delaware. The status of deregulation is that Delaware amount of credit. At the end of 1998, Hawaii had wind has a law that provides for phasing in retail competition facilities operating with total capacity of 20 MW. Hawaii beginning in October 1999, to be completed by April had three new projects planned to come on-line in 2000.

2001. In September 1999 the Delaware PUC issued final Potentially they would add a total of nearly 40 MW of orders for restructuring. Delaware has a public benefit wind capacity.

fund for renewable energy and efficiency, but no decision has been made as to how the fund is to be Idaho. Electricity deregulation in Idaho is on hold.

spent. The legislature has enacted net metering for Investigations concluded that Idaho is a low-cost State renewable facilities equal to or under 25 kW in capacity. for electricity and should be concerned about prices Delaware has no existing wind facilities and no new rising in a competitive market. Idaho has several wind facilities were planned for 2000. mechanisms that could support potential wind projects.

For example, net metering is available to all technologies District of Colombia. The District of Columbia PSC with facilities equal to or under 100 kW in capacity, not approved Potomac Electric Power Companys (PEPCO) just renewable facilities. Another incentive consists of a restructuring settlement in January 2000. Government personal income tax credit up to $5,000 for 40 percent of and commercial consumers will have retail access, and the cost of a solar, wind, or geothermal device used for a pilot program for residential consumers was to begin heating or electricity generation. Low-interest loans are by January 2001. The District of Colombia has no incen- available to residential and commercial consumers for tives for wind power, no existing wind projects iden- renewable projects to generate electricity for their own tified and no new wind facilities were planned for 2000. use. Projects that intend to sell electricity are excluded.

Loan amounts are limited to $10,000 for residential Florida. Florida has been slow to take action towards consumers and $100,000 for commercial consumers.

electric utility restructuring. In April 1998, House Bill (HB) 1888 died in committee without a hearing. In April Idaho has no identified wind facilities and none were 1999, the legislature adjourned with no further effort planned for 2000.

taken on restructuring. In January 2000 House issued a report on the state of the electric power industry in Illinois. Regarding the status of electricity restructuring Florida. Following that in April 2000 Senate Bill 2020 in Illinois, phasing in of retail competition for industrial was introduced and would require a study of electric and commercial customers was to begin in October 1999 utility deregulation and energy policy in Florida. and be completed by October 1, 2000. Residential Energy Information Administration/ Renewable Energy 2000: Issues and Trends 97

customers will receive a 5 percent rate reduction by 80.2 MW in Alta, Iowa, developed by Enron and October 1, 2001. In addition, as part of a court settle- Northern Alternative Energy (NAE) using Zond ment, ComEd is required to make a one-time allocation equipment to sell power to Alliant/IES of $250 million to an environmental and energy efficiency fund. 42 MW in Clear Lake, Iowa, developed by FPL using NEG-Micon equipment to sell power to Illinois has a system benefits charge in place that Alliant/IES.

supports renewables including potential wind projects.

The charge is a flat rate of $0.50/month for residential Other factors influencing development include the and small commercial customers. Larger customers pay following State provisions:

$37.50/month. The fund is budgeted for $5 million every year for 10 years. Fifty percent of the funds collected go Grants for Energy Efficiency and Renewable toward the Renewable Energy Resources Trust. Effective Energy. Sponsored by the Iowa Energy Center, April 2000, Commonwealth Edison established an these grants include support for a wide variety of experimental net metering program for solar or wind research activities, including among them wind generating systems equal to or less than 40 kW in resource assessment.

capacity. Illinois has no identified wind facilities and none were planned for 2000. Guaranteed Buy Back Rates. Within certain set limits, utilities are obligated to purchase renewable Indiana. In March 1999 a restructuring bill, HB 648, was power at incentive buy back rates which are higher introduced, but failed to move beyond a committee than the utilities avoided cost.

hearing. It was opposed by utilities, organized labor, and consumer and environmental groups. Indiana has Alternative Energy Loan Program. This program several incentives for renewables that can benefit the offers 0 percent interest loans for up to half of the development of wind power. First is the property tax project cost with a maximum of $250,000 for incentive, which exempts from property taxes the entire entities in the residential, commercial, and renewable energy device and affiliated equipment. industrial sectors.

Second is net metering for qualifying facilities generating less than 1,000 kWh per month. To date, this incentive Property Tax Incentive. Any city or county in Iowa has benefitted operators of small wind turbines. The has the option to assess wind energy equipment at third is demand side management programs. The a special valuation for property tax purposes Indiana Utility Regulatory Commissions 1995 ruling on following State guidelines. For the first year, wind demand side management programs allows for the energy conversion equipment is assessed at 0 inclusion of renewable energy systems (including wind percent of the total cost. In the second through the facilities) in such utility programs. Indiana has no wind sixth years the equipment is assessed at an facilities identified and there were no plans to build any additional 5 percent per year. From the seventh in 2000. year onward, the assessment is set at 30 percent of total cost.

Iowa. According to data from the American Wind Energy Association, Iowa had a number of small wind Sales Tax Incentive. This statute exempts from facilities in operation before 1999. Some of these facilities Iowa State sales tax the total cost of wind energy were too small to be included in EIA data and some equipment and all materials used in the were just not yet reporting. They included a 2.25 MW manufacture, installation, or construction of wind project in Algona, Iowa, developed by Cedar Falls systems.

Utilities using Zond designed equipment with support from the DOE/EPRI Turbine Verification Program. In Net Metering. This ruling allows Iowa customers 1999, a 1990 State law, mandating that utilities in Iowa with alternative energy generation systems to sell collectively take an average of 105 MW of electricity electricity back to the utilities on a netted basis.

from renewables, was a factor (although not the only Utilities are obligated to buy excess electricity at one) in the major development of approximately 240 their avoided cost. To date, this program has not MW of new wind capacity. This development includes been particularly popular due to impediments some of the following facilities: imposed by the utilities.

112.5 MW in Alta, Iowa, developed by Enron using Research and Outreach Programs. The Iowa Zond equipment to sell power to MidAmerican Energy Center has been involved in assessing the 98 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

States wind resources and developing a model to that can promote wind energy when available. Kentucky be used for siting wind turbines. It also administers has no incentives for renewable energy, no identified a loan program which offers 0 percent interest wind facilities, and no new wind facilities were planned loans for up to half the project cost up to a for 2000.

maximum of $250,000 and as long as funds allocated for wind portion of the renewable loan Louisiana. In March of 1999 the Public Services program are available. Commission issued an order stating that ...a deliberate and cautious approach is still warranted for restruc-In addition, one municipal utility, Cedar Falls has a turing the electric industry. A schedule was set to study green pricing program to promote wind energy. the issues through August 2000. Louisiana has no incentives for wind energy, no existing wind facilities The status of deregulation in Iowa is that a proposed identified, and no new wind facilities were planned for restructuring bill died at the end of the legislative 2000.

session in Spring 2000. The Iowa Department of Natural Resources proposed adding a renewable portfolio Maine. The Restructuring Act of 1997 allows electric standard with a goal of 4 percent renewable electricity power to be sold directly to retail consumers by largely by 2005 and 10 percent renewable electricity by 2015, but deregulated power providers competing with one the restructuring legislation failed to pass. A 600 kW another beginning March 2000. By the end of 1999 the wind project was proposed for Spirit Lake to come on- Maine PUC had finalized rules necessary to implement line in 2000. restructuring on schedule. Electric bill charges were to be unbundled beginning in 1999. Maine has the highest Kansas. The status of deregulation is that several bills renewable portfolio standard in the United States& some were introduced in the 1999 legislative session to 30 percent. However, counting electricity from hydro-restructure the electric power industry, but no action power, biomass, and gas cogeneration, Maine already was taken before adjournment. There are two existing exceeds this using existing renewable capacity. Maine programs that include incentives for wind power also has a net metering program for small facilities development. under 100 kW in capacity. Recently, Maine revised the net metering program to be consistent with retail access.

Renewable Energy Grant Program. This provides Under the old provisions customers could sell excess support in small amounts of funds (less than power to the utility. According to new provisions

$50,000) for development of renewable energy, customers will accumulate a rolling credit, which will including wind, and excluding research and roll over for 12 months, after which the credit goes development. away. Maine has no currently identified wind facilities, but a 20 MW project on Reddington Mt. was in the

Kansas Electric Utilities Research Program process of being permitted with plans to be on line by (KEURP). is a cooperative venture among seven December 2000.

electric utilities performing applied research to proactively seek and deliver technologies en- Maryland. Restructuring legislation provides for a hancing the value of electric services to its mem- phase-in of retail competition starting in July 2000 and bers, utility customers, and the State of Kansas. In ending July 2002. In January 2000 the Maryland PSC the past this has included a collaborative project approved PEPCOs restructuring plan and PEPCO with DOE to conduct a wind siting study. customers were scheduled to begin retail direct access by July 2000. While Maryland has several incentives for In addition, two investor owned utilities have green solar energy, it has no incentives for wind, no identified pricing programs to promote wind energy exclusively. wind facilities, and no new wind projects were planned So far, Kansas completed one small 1.5-MW wind for 2000.

project in 1999 and has no plans for any new wind facilities in 2000. Massachusetts. Open retail competition began in March 1998. Accompanying restructuring is a renewable Kentucky. The Kentucky Task Force on Electric portfolio standard that includes wind. Retailers are Restructuring, established by HRJ95, completed its final required to take 1 percent of their supply from new report and found that retail prices in Kentucky could renewables in 2003. This requirement increases by 0.5 rise under open competition. Kentucky has one percent per year until 2009, and 1 percent per year municipal utility sponsoring a green pricing program thereafter. To support implementation of the renewable Energy Information Administration/ Renewable Energy 2000: Issues and Trends 99

portfolio standard, Massachusetts also has mandated the Minnesota. So far, electric power restructuring has had disclosure of fuel mixes to end use customers. The State little effect on wind power development. Although re-has also established the Massachusetts Renewable structuring legislation was introduced to both the House Energy Trust Fund, which is supported by a system and Senate, it never passed. Of far greater importance to benefits charge which began collection in 1998. Imple- wind energy development in Minnesota is a unique mentation of the full program is proceeding and quid pro quo law regarding storage of spent nuclear includes potential benefits for wind. Massachusetts also fuel. A law passed in 1994 allows Northern States Power has a net metering program for all qualified facilities (as (NSP) to store nuclear waste in dry caskets near one of defined by PURPA and FERC) at or below 60 kW of its nuclear power plants in exchange for a commitment capacity according to legislation enacted in 1997. Net to develop new wind capacity. According to plan 425 excess generation is purchased at the electric utilities full MW of wind power capacity would come on line by avoided cost. 2002 with 400 more megawatts to follow by 2012.

Massachusetts has various other renewable incentives of This legislation is not the only factor affecting develop-less importance, including the following. The State has ment. Minnesota has a number of State incentives and an alternative energy patent exemption, which offers programs that, when taken in combination, can help both corporate and personal income tax deductions for make wind projects viable. These incentives include:

any income received from the sale of a patent or collection of royalties for patents that benefit develop- Corporate Income Tax Credit. Minnesota has ment of alternative energy for 5 years from the time the accelerated depreciation provisions in the State tax deduction is granted. A corporate income tax credit code that mirror the federal Modified Accelerated permits corporations to deduct solar or wind expendi- Cost Recovery Schedule (MACRS). That is a 5-year, tures for space or water heating from their taxable 200-percent double declining balance accounting income. The State also exempts solar and wind facilities method.

from corporate excise tax for the length of the projects depreciation period. Massachusetts has a special grant Special Grant Program. Minnesota provides a 1.5 program for partnerships with the private sector and cent per kilowatthour grant for 10 years to wind local communities. These grants support development of projects 2 MW or smaller in size on a first come fuel cells, wind, and solar photovoltaics. first served basis up to a statewide total of 100 MW wind power capacity. This program is meant to The States renewable energy systems credit provides for encourage establishment of dispersed wind genera-a 15-percent credit (with a maximum limit of $1,000) tion infrastructure.

against State income tax for the cost of a renewable energy system installed at an individuals primary Agricultural Improvement Loan Program. This residence. The local property tax exemption for solar, program provides low interest loans up to $100,000 wind, and hydro exempts these facilities from local to farmers for improvements or additions to property taxes. Massachusetts also exempts from State permanent facilities. Wind energy conversion sales tax, solar, wind, and heat pump systems operating equipment has qualified since 1995.

in an individuals primary residence.

Value-Added Stock Loan Participation Program.

Massachusetts has only two small wind facilities This program can provide small, low-cost loans to identified&each with capacity under 0.5 MW. One new farmers wishing to buy into wind generation wind project with capacity of 7.5 MW was planned for cooperatives. There has been very little activity for 2000. wind in this program thus far, because the maximum amount of capital available is usually Michigan. Recently enacted electricity restructuring insufficient to finance even a small wind project.

legislation allows all customers retail choice by January 2002. One way Michigan supports wind is with a Property Tax Exemption. This provision excludes program, Green Rate, in which customers pay a monthly from property taxation all or part of the value premium to have all their power sourced to the Traverse added by wind systems. The value is determined City 600-kW wind project. Great Lakes Energy Cooper- on a sliding scale. Some small systems have the ative has a second green pricing program to promote total value exempt, while all systems 12 MW or wind power. There were no other plans to add wind greater in capacity have 25 percent of the value capacity in 2000. taxed.

100 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Sales Tax Incentive. Minnesota exempts from Furthermore, facilities with a total of 30 MW capacity at sales tax the total cost of wind energy devices, 17 dispersed sites were to be developed by Northern including equipment and all materials used to Alternative Energy with plans to be on line by the end of manufacture, install, construct, or repair such 2000.

systems.

All of the projects listed above have power purchase

Easements. Minnesota provides for wind ease- agreements with Northern States Power. Additional ments. An easement that benefits the property wind capacity, being proposed, is expected to be cannot add value to the property for tax purposes. developed in the future to meet Northern States Powers complete long-term commitment under the 1994 law.

Green Pricing. Minnesota has one investor owned Also, a 1.98 MW project for Chandler Hills is in the utility (Minnesota Power), four electric cooper- preliminary stages of planning.

atives, and one municipal utility promoting wind power to customers who wish to pay a premium Mississippi. Pending enactment of authorizing legisla-for clean energy. tion, Mississippis electric power suppliers were set to implement retail competition starting January 2001 and

Net Metering. Minnesota offers net metering to ending December 2004. The City of Oxford, North East wind facilities with 40 kW of capacity or less. Mississippi Electric Power Association, has a green Utilities must purchase any excess power gen- program that started in 2000 that can promote wind erated at the average retail rate. energy when available. Mississippi has no identified wind facilities and no new wind capacity was planned

Public Benefit Fund. In addition to developing for 2000.

wind capacity in exchange for storing nuclear waste, the 1994 law also required Northern States Missouri. Several bills to restructure the electric power Power to contribute $4.5 million to a fund industry and allow retail access were introduced in the beginning in 1999 and equal or greater amounts in legislature in the winter of 1999, but none were passed.

successive years. These payments would continue Missouri has a loan program for renewables and indefinitely until either the law is changed or the potential wind projects. Funds are loaned to schools, casks can be shipped to a national nuclear-waste local governments and small businesses. One investor storage or disposal site. Money in this fund will be owned utility, Missouri Public Service (Utilicorp United) used to help finance projects that produce has a green pricing program to promote wind power electricity from nontraditional sources and also when its available. Missouri has no identified wind benefit local economies in Minnesota. facilities and had no plans to build any in 2000.

With the support of the federal production tax credit, Montana. The status of deregulation in Montana is that the 1994 State law, and various other State incentives, retail competition is being phased in with a targeted end Minnesota brought on line nearly 140 MW of wind date of July 1, 2002, though extensions may be granted generating capacity in 1999. The following facilities are up to July 1, 2006 (depending on the utility and service representative of those that came on line in 1999: area involved). Montana has required since May 1997 that electric bills be unbundled. In terms of renewable

107.25 MW in Lake Benton, Minnesota (Lake Ben- energy support, Montana has a number of incentives ton I), developed by Enron using Zond equipment. that could be applied to wind and these will be detailed here. However, the State has no existing wind facilities

103.5 MW in Pipestone County, Minnesota (Lake identified and had no plans for any capacity additions in Benton II), developed by Enron using Zond 2000.

equipment and now owned by FPL Energy, LLC.

Montana has a system benefits charge that went into

11.25 MW in Lakota Ridge, Minnesota, developed effect July 1, 1999, and will continue 4 years until July 1, by Northern Alternative Energy using NEG Micon 2003. Electricity suppliers will contribute 2.4 percent of equipment their 1995 revenues to the fund. Electric utilities will be responsible for spending the monies. Funds allocated to

11.88 MW in Shaokatan Hills, Minnesota, devel- renewable energy could be spent for wind to conduct oped by Northern Alternative Energy using Vestas research and development (R&D) or to actually build a equipment. facility.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 101

Montanas support programs also include the following. New Hampshire. The State enacted HB1392 in 1996, First is net metering, which can apply to wind gen- requiring the PUC to implement retail choice by July erators with capacity equal to or under 50 kW. There is 1998. However, implementation of restructuring was also an income tax credit that could apply to wind. This delayed due to continuing Federal litigation concerning program allows a 35-percent tax credit for an individual, the PUCs efforts to set stranded costs and rates for partnership, or corporation that makes an investment of Public Service of New Hampshire (PSNH). In June 2000

$5,000 or more in wind electricity generating system or SB472 was signed into law. This legislation is aimed at facilities to manufacture equipment. Another provision lowering PSNHs rates and allowing customers to of Montana law exempts from property taxation the choose an energy supplier. In September 2000 the New value added by a qualified renewable energy source, Hampshire Public Utilities Commission issued orders including wind. Montana is also one of four States that approving PSNHs restructuring settlement agreement provides for the creation of wind easements for the and a schedule for phasing in retail competition will be purpose of protecting and maintaining proper access to set.

sunlight and wind. Finally, one electric cooperative has a green pricing program that can promote wind. New Hampshire has several small-scale support pro-grams which could apply to wind, if facilities were built.

Nebraska. Nebraska has been exploring electricity The first of these includes a net metering provision, restructuring, but this effort is still in the investigative which is currently under revision by the State PUC.

stage. Nebraska has several programs that could benefit Under new rules there would be full net metering and potential wind projects, including a wind easement law. credits would roll over at the end of each month.

This law allows property owners to create binding wind Capacity would be limited to 25 kW. Second, a easements for the purpose of protecting and maintaining demonstration grants program provides grants between proper access to wind energy. Another is a low interest $5,000 and $10,000 for renewable demonstration/educa-loan program that can support development of future tion projects. In a recent year, all the grants were for wind projects. Finally, one municipal utility has a green PVs, although wind is eligible. Third, a local option pricing program promoting wind power. Nebraska has property tax statute allows each city or town to offer an one 1.5 MW wind facility on line in Springview not yet exemption on residential property taxes in the amount included in EIA data (but supported in part by the DOE of the assessed value of the eligible renewable energy Wind Turbine Verification Program), and one 1.32 MW system used on the property.

wind facility operating in Lincoln. No additions were planned for 2000. New Hampshire has no identified wind facilities and had no plans for building any in 2000.

Nevada. In June 1999, Nevada enacted new restruc-turing legislation, which amended a 1997 law. The PUC New Jersey. In February 1999, the State enacted legis-has set a schedule to begin retail competition for the lation to restructure New Jerseys electric power largest commercial customers in November 200. Retail industry, providing for the beginning of retail competi-competition will be open to all customers by the end of tion in August 1999. Since then, one agreement between 2001. the Board of Public Utilities and Connectiv provided for a delay of retail competition until November 1999. New Nevada has a few incentive programs for wind, but Jersey has a number of support programs for renewable none of particular significance. These programs include energy development. First, New Jersey also provides for a renewable portfolio standard requiring utilities to have a 4-percent renewable portfolio standard to be met by 0.2 percent of their electricity from renewables by 2012 using non-hydroelectric sources of renewable January 1, 2001 increasing to 1 percent by 2009. Half of energy. Second, New Jersey has a public benefit fund that is required to be solar. There is also a net metering that will total $265 million for 2000-2008. Wind is an law, but only for facilities of 10 kW capacity or less and eligible technology. However, the New Jersey Board of only for the first 100 customers of each utility. A Public Utilities has yet to issue a final rule on how these property tax incentive provides that any value added by will be administered. In addition, since 1999 New Jersey a qualified renewable energy source shall be subtracted has had net metering for wind and PV generators with from the assessed value of any residential, commercial no limit on generator size. Another incentive for renew-or industrial building for property tax purposes. Nevada ables is the exemption from New Jerseys 6 percent State has no identified wind facilities and none were planned tax. New Jersey has no identified wind facilities and had for 2000. no plans for any in 2000.

102 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

New Mexico. Legislation to restructure New Mexicos assistance. A small project was planned for Wyoming electric power industry was enacted in April 1999. county to come on line in 2000.

According to current plans, consumer choice will begin with residential and other small consumers in the North Carolina. Restructuring is under investigation in beginning of 2002, followed by other larger users at a North Carolina. In March 1999, the Research Triangle In-later date. The restructuring legislation contains a stitute submitted its report with recommendations to the provision for a system benefits charge to be levied on all North Carolina Public Utilities Commission, but no kilowatt-hour sales in New Mexico. These funds will be further action was expected in 1999. In April 2000 the used by the New Mexico Department of Environment to Study Commission, which was established by Senate support activities including development of renewable Bill 38 in 1997, issued its final report. It recommends energy by school districts and the governing entities of opening retail electricity markets to half of consumers by cities towns and villages. New Mexico also has a limited January 2005 and the remainder by January 2006, as well renewable portfolio standard. It provides for up to 5 as, creating a public benefits fund that could benefit percent of electricity to come from renewable resources renewables. It also proposed providing a choice for by 2002 if it can be shown renewable resources are green energy or alternatively a renewable portfolio available in New Mexico and if the cost of standard offer standard.

service does not increase.

Presently, North Carolina has one incentive that could New Mexico also has a net metering program that support wind energy development. The income tax benefits small renewable facilities under 10 kW in credit provides a credit against corporate and personal capacity. The State has one investor owned utility, income taxes in the amount of 10 percent of the cost of Southwestern Public Service, with a green pricing pro- equipment and installation of a wind energy system not gram that can apply to wind energy. New Mexico has to exceed $1,000 for any single installation. North one small wind facility in operation, a 0.66 MW facility Carolina has no wind facilities identified as in operation in Clovis and no new facilities were planned for 2000. and none were planned for 2000.

New York. With regard to electricity industry restruc- North Dakota. In November 1998, the Electric Utilities turing, New York is currently phasing in retail Committee submitted its report to the legislature on competition statewide. Each utility has its own timetable restructuring, but no action has yet been taken. The next of targets. Some utilities have reached full retail access, legislature meets in 2001. North Dakota has several while others expect to by the end of 2001. Although it is incentives that could support wind energy. The personal not entirely clear how the industry will change as income tax credit allows any taxpayer to deduct 5 per-restructuring transpires, New York presently has some cent of the cost of equipment and installation of a support for renewable energy (including wind). In the geothermal, solar or wind energy device for a period of past, a surcharge levied on intrastate sales of gas and 3 years. The property tax incentive exempts from local electricity by investor-owned utilities provided funds property taxes any solar, wind, or geothermal energy for, among other things, research, development and device for the first 5 years of operation. North Dakota commercialization of renewable technology as well as also has a net metering program for renewable gen-financial support to further market penetration of erators equal to or under 100 kW in capacity. In North renewable energy. For the future, the New York Public Dakota Minnakota Power Cooperative has a green Services Commission ordered utilities to provide pricing program to promote wind energy development.

unbundled billing by April 2000, which will identify North Dakota has a few small identified wind facilities electricity provided by green sources. Also, the PSC has too small to be included in EIA survey data. Two are set rules for a new system benefits charge to fund R&D operated by Indian tribes. Together, these facilities for renewable energy. The fund will run through 2001 represent less than 0.5 MW of capacity. No new wind and be administered by the New York State Energy facilities were planned to come on line in 2000.

Research and Development Authority (NYSERDA).

New York has net metering, but it is for solar only and Ohio. In July 1999, Ohio enacted legislation to does not apply to wind energy. restructure the Ohios electric power industry. In October 1999, the PUC issued an initial set of rules for One 11.5 MW facility was planned by PG&E Generating transition to a competitive market. Since that time a for Madison, New York, to be on line in 2000. Some of number of utilities have submitted transition plans for the electricity is intended to be sold to green power PUCOs approval. Retail competition was to be phased providers. NYSERDA will provide $2 million as in beginning January 1, 2001. Ohio has net metering Energy Information Administration/ Renewable Energy 2000: Issues and Trends 103

available for wind facilities with no size limit on the commercial, or industrial) derived from the installation generator. Ohios tax system exempts certain equipment, of a qualifying renewable energy device shall not be including wind generators, from property taxation, the included in the assessment of the propertys value for State sales and use tax, as well as the State franchise tax property tax purposes. The fourth is net metering for where applicable. Ohio has no identified wind facilities wind generators with capacity equal to or under 25 kW.

and none were planned for 2000.

Oregon has four green pricing programs supporting Oklahoma. In April 1997, SB 500 was enacted to provide wind energy development. They are sponsored by two for electricity restructuring. It targeted retail competition investor owned utilities, one electric cooperative, and to begin July 2002. Subsequently, SB 888 was enacted, one municipal utility. One example is Portland General which would bring in retail competition earlier. In Electrics (PGE) green pricing program open to large October 1998, the Joint Electricity Task Force began a industrial and wholesale customers. PGE has contracted series of studies on implementing restructuring. The last to supply this program in part with energy from of these studies was to be completed by October 1999. In Oregons existing wind farm, the 24.9 MW Vansycle late Spring 2000 the State legislature was working on a facility, which started operations in December 1998. No compromise bill to establish rules for implementing new wind facilities were planned for either 1999 or 2000.

electric power industry restructuring. Oklahoma has a provision for net metering that could benefit wind Pennsylvania. In 1999, Pennsylvania began phasing in energy development. Customers can request the utility retail competition in stages. In September 1999, utilities to pay for extra power generated, but the utilities are not were required to mail information packages to all con-required to comply. Oklahoma has no identified wind sumers that had not chosen a competitive supplier with facilities, and none were planned for 2000. the hope of getting them in the new system by January 2000. Disclosure of fuel mix is encouraged. In addition, Oregon. In July 1999, Oregon enacted legislation that Pennsylvania has an RPS, SBC, and net metering, but will deregulate the electric power industry and allow for provisions vary for each utility service territory. Separ-customer choice.60 The law will phase in open ately, the PECO Unicom merger established a fund that competition for industrial and commercial customers, has $12 million budgeted for wind over a 5-year period.

but residential customers will have a portfolio of electricity products from which to choose. Products are Pennsylvania also has green power programs that could provided by the incumbent utility and include a green benefit future wind projects, when they are built. Green power option. Generation companies will be chosen by Mountain Energy opened its program in 1998 and sells the utility through competitive bidding, acting as a three products: electricity with 1-percent, 50-percent, middleman for residential customers. The bill also and 100-percent renewable sources at a modest increase requires disclosure of fuel sources, emissions and price, in cost compared to traditional energy sources. Another and creates a public purpose fund with funds set aside program, Connectiv Energy is the first program in for renewables including wind. Beginning in October Pennsylvania to be certified by the green-e program. It 2001 renewables would receive about 17 percent of the offers Natures Power 50 and Natures Power 100 made fund each year for 10 years. Separately, the governor from 50-percent and 100-percent renewable energy, signed into law a bill to implement net metering for respectively. The Energy Cooperative Association renewable facilities less than 2.5 kW in size. sponsors another green power program. Pennsylvania has one 10 MW wind facility, owned by American Oregon already has some other renewable incentives in National Power, which was dedicated in May 2000 in place. The first is the corporate income tax that permits Somerset County, Pennsylvania. Green Mountain Power a 35-percent investment credit up to $100,000 for markets power from this facility. A new 15.6 MW wind construction of systems that produce energy from facility at Mill Run in Fayette County was planned to go renewable sources, including wind. The second is the on line in 2000.

Small Scale Energy Loan Program (SELP). A 1980 amendment to the Oregon constitution authorizes the Rhode Island. The Rhode Island Utility Restructuring sale of bonds to finance small-scale, local energy Act of 1996 provides for electricity restructuring and projects, potentially including wind. Third, Oregons open retail competition was to be phased in during 1998.

property tax exemption for renewable devices states that By September 1999 only a small number of consumers the added value to any property (whether residential, had chosen alternative electricity providers. Rhode 60 Wind Power Monthly, June 1999, p. 38.

104 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Island has a non-bypassable system benefits charge to This would achieve a standard of about 3 percent support the development of renewable energy and renewable electricity for utilities by January 2009. By the demand side management programs. The charge is set winter of 2000 rules to implement the standard were at $.0023 per kilowatthour for a minimum of 5 years finalized by the PUC.

beginning in 1996. Rhode Island also has a net metering program created in 1985 that benefits a few small wind Prior to this, in October 1998, the Texas PUC adopted a generating facilities equal to or under 25 kW in capacity. renewable energy tariff rule that allows all utilities in Rhode Island had no plans for new wind facilities in Texas to offer customers the opportunity to buy renew-2000. able energy. If a utility chooses to offer a renewable energy tariff, its customers buying renewable energy South Carolina. With regard to deregulation, the South may be charged a premium above their standard energy Carolina legislature discussed a new bill introduced in cost to cover any cost of a renewable resource that the Senate and debated the issues in the Spring of 2000. exceeds the utilitys average system cost, plus marketing The Bill did not pass that session. South Carolina has no costs and possible utility profit. Two utility green pricing incentive programs for wind energy development, and programs are sponsored by the investor owned utilities:

no existing wind facilities identified. No additions were TXU Electric and the Texas-New Mexico Power planned for 2000. Company. Two municipal utilities also have programs.

Texas also has net metering for renewable generators South Dakota. Deregulation in South Dakota has been with capacity equal to or under 50 kW.

under investigation. Findings of these activities assert that restructuring would not be good for South Dakota. By the end of 1999 Texas had three large wind facilities Because the State has some of the lowest rates in the on line. They were (1) Culberson County with 65 MW of Nation, it is expected electricity prices would go up Kenetech and Zond turbines, (2) Big Spring, Texas, with under open retail competition. Existing law permits 35 MW of Vestas Turbines, and (3) McCamey, Texas, retail wheeling for new, large customers. with 75 MW of NEG Micon turbines. In addition, several smaller projects, including the 6 MW facility in Fort South Dakota has a property tax incentive that exempts Davis, Texas, received support from the DOE Wind renewable energy systems on residential and com- Turbine Verification Program. Two new projects were mercial property from local property taxes for 3 years planned for 2000. One was a 21.6 MW facility in King after installation with certain restrictions. The East River Mountain and the other is a 3.5 MW plant in Fort Electric Cooperative has a green pricing program that Stockton.

can promote wind energy planned to start in 2000. South Dakota has no identified wind facilities, but the Rosebud Utah. Deregulation in Utah is under investigation. Utah Sioux tribe had a 750 kW facility planned to come on- has a renewable energy income tax credit. For residential line in 2000. systems, the credit is 25 percent of the cost of installation up to $2,000 per system. For commercial systems, the Tennessee. Because the TVA provides most of credit is 10 percent of the cost of installation up to Tennessees electricity cheaply, little interest exists in $50,000 per system. Utah has no identified wind facilities restructuring the electric industry, although it has been operating, but a 225 KW facility in Camp Williams, investigated. Tennessee has a loan program that offers Riverton, was planned for 2000. Utah Power (Pacificorp) loans up to $100,000 for renewable projects including has a new green power program that could apply to wind. The Tennessee Valley Authority (TVA) has a wind energy when available.

green power program that could apply to wind energy when available. Tennessee has no existing wind projects Vermont. Alternative proposals for restructuring date identified, but TVA proposed a 1.98 MW project for back as early as December 1996, but the issue of Buffalo Mountain in Anderson County to come on line stranded costs has been a stumbling block to enacting in 2000. any legislation. At present, all of the utilities have power purchase contracts with Hydro Quebec and local Texas. Texas enacted legislation to restructure the independent power producers that are above market electric power industry and permit retail competition. price. To provide a path to a solution, the Department of The States electricity industry will begin open com- Public Service has already permitted temporary rate petition by 2002, and by 2009 State utilities will be increases, until contracts can be renegotiated. According required to develop 2,000 MW of new renewable-based to restructuring plans filed with the Public Service Board power. Some of this capacity could use wind energy. in March 1999, Central Vermont Public Service and Energy Information Administration/ Renewable Energy 2000: Issues and Trends 105

Green Mountain Power will divest themselves of their provides for an RPS and PBF. The RPS provision major generating assets and merge into one distribution requires 0.5 percent of retail energy sales to come from company. Other details have yet to be announced. renewable energy sources (excluding electricity from Vermont has net metering for small wind facilities with hydroelectric facilities 60 MW and higher in capacity).

capacity equal to or under 15 kW or for farm system This percentage would be boosted to 2.2 percent in 2011.

generators 100 kW or less in size. A small portion of the public benefits fund would go to encourage the development or use of renewable Vermont has one 6 MW wind facility in operation in applications. Some of these renewable provisions could Searsburg, Vermont, not yet included in EIA data. This benefit wind energy development in the future.

project was supported in part by a grant from the DOE Wind Turbine Verification Program. Vermont also had A number of other incentives for wind energy already plans for new wind facilities in 2000. exist:

Virginia. Early in 1999, the Virginia Electric Utility Solar and Wind Energy Equipment Exemption.

Restructuring Act was signed into law. It provides for This tax incentive exempts taxpayers from any retail competition to be phased in beginning January 1, value added by a qualified renewable energy 2002, through until January 1, 2004. Virginia has recently source for property tax purposes.

enacted net metering for residential wind generators with capacity equal to or under 10 kW and for non- Solar and Wind Access Laws. Wisconsin statutes residential wind generators 25 kW or less in size. allow property owners with wind or solar energy Virginia has no existing wind facilities identified and systems to apply for permits which will guarantee had no plans for new wind facilities in 2000. unobstructed access to solar and wind resources.

Washington. In October 1999, a plan&Reliability 2000& Net Metering. Net metering is available to all to restructure the electric power industry was proposed, customer classes for systems with capacity of but has yet to be passed. Among programs that could 20 kW or less. For electricity from renewable support wind projects, one is an exemption from the energy the utilities pay the retail rate for net excess State corporate excise tax. Another is net metering for generation.

wind generators 25 kW or less in capacity. A third type of support is Washingtons research and outreach pro- Green Pricing. Madison Gas and Electric plans to grams that provide prospective renewable developers offer a green pricing program to support its new technical assistance, education, workshops, and other 11.22 MW wind farm in eastern Wisconsin.

field assistance. Washington has three utility green Customers can choose to purchase 100 kWh blocks pricing programs that can promote wind energy when for a monthly premium of around $5. Wisconsin available. Washington has no existing wind facilities Electrics pilot program, Energy for Tomorrow, identified and none immediately planned for 2000. with 9,000 participants was so successful it is being extended to more customers.

West Virginia. In March 2000 the legislature approved the Electricity Restructuring plan submitted by the A Clean Energy Rebate Program was proposed in State Public Services Commission. It will allow retail choice by Senate Bill 56 introduced in February 1999. Under its January 2001. West Virginia has no existing wind provisions, an individual may receive a rebate of up to facilities identified and none were planned for 2000. $2,000 from the State for installing a wind or solar system. 61 Madison Gas and Electric and the Wisconsin Wisconsin. Wisconsin is one State that has not restruc- Electric Power Company are two investor-owned tured its electric power industry, but it has a renewable utilities with green pricing programs to promote wind portfolio standard and public benefits fund. Early legis- energy; in addition, one electrical cooperative has a lation signed into law in April 1998 mandated utilities to program.

create 50 MW of power from renewable sources by 2000.

Subsequently, Wisconsins Reliability 2000 legislation By the end of 1998, Wisconsin had one 1.2 MW facility went into effect in October 1999. In addition to over- on line in De Pere, Wisconsin, (supported in part by the hauling the States transmission system, the law DOE Wind Turbine Verification Program) not yet 61 Personal communication with John Stolzenberg, Wisconsin Legislative Staff, April 29, 1999.

106 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

included in EIA data. Three facilities followed in 1999. EIAs Status of State Electric Utility Deregulation They were (1) Nagara Escarpment%11.2 MW of Vestas Activity, website:

turbines, (2) Lincoln Township%9.24 MW of Vestas http://www.eia.doe.gov/cneaf/electricity/chg_str turbines, and (3) Byron%1.32 MW of Vestas turbines. /tab5rev.html There were no plans for any new wind facilities immediately in 2000. U.S. Department of Energy, Electric Utility Restruc-turing Weekly Update, website:

Wyoming. The Wyoming Public Service Commission http://www.eren.doe.gov/electricity_restructurin issued a paper analyzing electric industry restructuring g/weekly.html in September 1997. Some follow-up action was taken, but no further activity of significance has taken place Strategic Energy Ltds Electricity Competition since June 1998. Wyoming has only one renewable Update, website:

incentive, a solar/wind access law which provides very http://www.sel.com/retail.html little benefit to wind energy. On the other hand, some of and the wind power being developed in Wyoming is to be used to support diversified programs in other States Electricitychoice.com, website:

such as Colorado. Pacific Power (Pacificorp), an investor http://www.electricitychoice.com owned utility, has a green power program.

Information on State incentives and green pricing was Wyoming has two large projects in Foote Creek Rim. taken from:

The first is a 41.4 MW facility that came on line in mid-North Carolina Solar Centers Database of State 1999. Average wind speeds are 25 miles per hour at the Incentives for Renewable Energy (DSIRE), website:

site, thus promising greater potential for wind genera-http://www.ncsc.ncsu.edu/dsire.htm tion. The project is owned 80 percent by PacifiCorp, an investor-owned utility based in Portland, Oregon, and K. Porter, National Renewable Energy Laboratory 20 percent by Eugene (Oregon) Water and Electric (NREL), and R. Wiser, Lawrence Berkeley National Board, a municipal utility. Sea West and Tomen Cor- Laboratory, A Status Report on the Design and poration built the project using 69 Mitsubishi turbines. Implementation of State Renewable Portfolio The second Foot Creek Rim project was Public Service Standards and System Benefit Charge Policies, Companys (PSCo) 25 MW project nearby. It uses 33 750- presented at Windpower Conference 2000 (Palm kW turbines manufactured for the most part by NEG Springs, California, May 2000). See the NREL Micons new facility in Illinois. Other projects include website: http://www.nrel.gov/analysis/emaa Foot Creek Rim III, a small 1.8 MW facility developed by Seawest and Tomen Power for Bonneville Power U.S. Department of Energy, The Green Power Administration, and a 3.3 MW facility by Fort Collins Network website:

Light and Power (of Colorado) in Medicine Bow. http://www.eren.doe.gov/greenpower An additional 10 MW facility on Simpson Ridge was Wiser, R., Porter, K. and Bolinger, M., Lawrence planned for completion in 2000. In mid-2000 Bonneville Berkeley National Laboratory. Comparing State Power announced another purchase power agreement Portfolio Standards and System-Benefits Charges with Seawest to construct a new wind facility and Under Restructuring, Memorandum (August 23, provide more green power. According to plans the new 2000) to various officials of the U.S. Department of Foot Creek Rim IV project was to have 28 wind turbines Energy and the National Renewable Energy Lab-with a total capacity of 16.8 MW and be operating by the oratory, as well as, from contacts with State Energy end of 2000. A small 1.32 MW project in Medicine Bow Commissions and the Public Utility Commissions.

was planned to be on line during the summer of 2000.

Information on wind capacity in place under con-struction in 1999 or planned for construction in 2000 was taken from:

Sources 62 The American Wind Energy Associations project Information on restructuring the electric power industry database (as updated on July 7, 2000 ) on the website:

was taken from the following websites: http://www.awea.org/projects/index.html 62 Information for this appendix was taken from various websites and is current as of the summer of 2000.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 107

Various articles in Wind Power Monthly and Wind ment of Energy, Wind Energy Program, website:

Energy Weekly. http://www.eren.doe.gov/wind/weu.html.

Information regarding projects in the Wind Turbine Verification Program was obtained from the Depart-108 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Glossary Alternating Current (AC): An electric current that Cast Silicon: Crystalline silicon obtained by pouring reverses its direction at regularly recurring intervals, pure molten silicon into a vertical mold and adjusting usually 50 or 60 times per second. the temperature gradient along the mold volume during cooling to obtain slow, vertically-advancing crystal-Amorphous Silicon: An alloy of silica and hydrogen, lization of the silicon. The polycrystalline ingot thus with a disordered, noncrystalline internal atomic formed is composed of large, relatively parallel, arrangement, that can be deposited in thin-layers (a few interlocking crystals. The cast ingots are sawed into micrometers in thickness) by a number of deposition wafers for further fabrication into photovoltaic cells.

methods to produce thin-film photovoltaic cells on Cast-silicon wafers and ribbon-silicon sheets fabricated glass, metal, or plastic substrates. into cells are usually referred to as polycrystalline photovoltaic cells.

Availability Factor: A percentage representing the number of hours a generating unit is available to Climate Change (Greenhouse Effect): The increasing produce power (regardless of the amount of power) in mean global surface temperature of the Earth caused by a given period, compared to the number of hours in the gases in the atmosphere (including carbon dioxide, period. methane, nitrous oxide, ozone, and chlorofluoro-carbons). The greenhouse effect allows solar radiation to Avoided Costs: The incremental costs of energy and/or penetrate the Earth's atmosphere but absorbs the capacity, except for the purchase from a qualifying infrared radiation returning to space.

facility, a utility would incur itself in the generation of the energy or its purchase from another source. Cogeneration: The production of electrical energy and another form of useful energy (such as heat or steam)

Baseload: The minimum amount of electric power through the sequential use of energy.

delivered or required over a given period of time at a steady state. Demand-Side Management, DSM: The planning, implementation, and monitoring of utility activities Biofuels: Wood, waste, and alcohol fuels produced designed to encourage consumers to modify patterns of from biomass (plant) feedstocks. electricity usage, including the timing and level of electricity demand. It refers only to energy and load-Biomass: Organic nonfossil material of biological origin shape modifying activities that are undertaken in constituting a renewable energy source. response to utility-administered programs.

Capacity Factor: The ratio of the electrical energy Direct Current (DC): An electric current that flows in a produced by a generating unit for the period of time constant direction. The magnitude of the current does considered to the electrical energy that could have been not vary or has a slight variation.

produced at continuous full-power operation during the same period. Electric Utility Restructuring: With some notable exceptions, the electric power industry historically has Capacity, Gross: The full-load continuous rating of a been composed primarily of investor-owned utilities.

generator, prime mover, or other electric equipment These utilities have been predominantly vertically inte-under specified conditions as designated by the manu- grated monopolies (combining electricity generation, facturer. It is usually indicated on a nameplate attached transmission, and distribution), whose prices have been to the equipment. regulated by State and Federal government agencies.

Restructuring the industry entails the introduction of Capital Cost: The cost of field development and plant competition into at least the generation phase of construction and the equipment required for the electricity production, with a corresponding decrease in generation of electricity. regulatory control. Restructuring may also modify or Energy Information Administration/ Renewable Energy 2000: Issues and Trends 109

eliminate other traditional aspects of investor-owned valuation of the nonmarket benefits of renewables.

utilities, including their exclusive franchise to serve a Green pricing programs allow electricity customers to given geographical area, assured rates of return, and express their willingness to pay for renewable energy vertical integration of the production process. development through direct payments on their monthly utility bills.

Emission: The release or discharge of a substance into the environment; generally refers to the release of gases Grid: The layout of an electrical transmission and or particulates into the air. distribution system.

EPACT: The Energy Policy Act of 1992 addresses a wide Incentives: Subsidies and other Government actions variety of energy issues. The legislation creates a new where the Governmentss financial assistance is indirect.

class of power generators, exempt wholesale generators, that are exempt from the provisions of the Public Independent Power Producer (IPP): A wholesale elec-Holding Company Act of 1935 and grants the authority tricity producer (other than a qualifying facility under to the Federal Energy Regulatory Commission to order the Public Utility Regulatory Policies Act of 1978), that and condition access by eligible parties to the inter- is unaffiliated with franchised utilities in the area in connected transmission grid. which the IPP is selling power and that lacks significant marketing power. Unlike traditional utilities, IPPs do not Exempt Wholesale Generator (EWG): A nonutility possess transmission facilities that are essential to their electricity generator that is not a qualifying facility customers and do not sell power in any retail service under the Public Utility Regulatory Policies Act of 1978. territory where they have a franchise.

Externalities: Benefits or costs, generated as a by- Integrated Resource Planning, IRP: In the case of an product of an economic activity, that do not accrue to electric utility, a planning and selection process for new the parties involved in the activity. Environmental energy resources that evaluates the full range of externalities are benefits or costs that manifest them- alternatives, including new generation capacity, power selves through changes in the physical or biological purchases, energy conservation and efficiency, cogen-environment. eration, district heating and cooling applications, and renewable energy resources, in order to provide Firm Power: Power or power-producing capacity adequate and reliable service to electrical customers at intended to be available at all times during the period the lowest system cost. Often used interchangeable with covered by a guaranteed commitment to deliver, even least-cost planning.

under adverse conditions.

Kilowatt (kW): One thousand watts of electricity (See Fuelwood: Wood and wood products, possibly includ- Watt).

ing coppices, scrubs, branches, etc., bought or gathered, and used by direct combustion. Kilowatthour (kWh): One thousand watthours.

Generation (Electricity): The process of producing Levelized Cost: The present value of the total cost of electric energy from other forms of energy; also, the building and operating a generating plant over its amount of electric energy produced, expressed in economic life, converted to equal annual payments.

watthours (Wh). Costs are levelized in real dollars (i.e., adjusted to remove the impact of inflation).

Geothermal Energy: As used at electric utilities, hot water or steam extracted from geothermal reservoirs in Marginal Cost: The change in cost associated with a unit the Earth's crust that is supplied to steam turbines at change in quantity supplied or produced.

electric utilities that drive generators to produce electricity. Megawatt (MW): One million watts of electricity (See Watt).

Giga: One billion.

Merchant Facilities: High-risk, high-profit facilities that Green Marketing/Pricing: In the case of renewable operate, at least partially, at the whims of the market, as electricity, green pricing represents a market solution to opposed to those facilities that are constructed with the various problems associated with regulatory close cooperation of municipalities.

110 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

Methane: The most common gas formed in coal mines; Photovoltaic Cell: An electronic device consisting of a major component of natural gas. layers of semiconductor materials fabricated to form a junction (adjacent layers of materials with different Modular Burner: A relatively small two-chamber electronic characteristics) and electrical contacts and combustion system used to incinerate municipal solid being capable of converting incident light directly into waste without prior processing or sorting; usually electricity (direct current).

fabricated at a factory and delivered to the incineration site. Photovoltaic Module: An integrated assembly of interconnected photovoltaic cells designed to deliver a Net Metering: Arrangement that permits a facility selected level of working voltage and current at its (using a meter that reads inflows and outflows of output terminals, packaged for protection against electricity) to sell any excess power it generates over its environment degradation, and suited for incorporation load requirement back to the electrical grid to offset in photovoltaic power systems.

consumption.

Public Benefits Fund (PBF): program, funded through Net Summer Capability: The steady hourly output that a generation or transmission interconnection fee on generating equipment is expected to supply to system electricity used, to fund various public purpose load, exclusive of auxiliary power, as demonstrated by programs, such as, low-income energy assistance, energy testing at the time of summer peak demand. efficiency, consumer energy education, and renewable energy technologies development and demonstration.

Nonutility Generation: Electric generation by end-users, independent power producers, or small power Public Utility Regulatory Policies Act of 1978 producers under the Public Utility Regulatory Policies (PURPA): One part of the National Energy Act, PURPA Act, to supply electric power for industrial, commercial, contains measures designed to encourage the conserva-and military operations, or sales to electric utilities.

tion of energy, more efficient use of resources, and equitable rates. Principal among these were suggested Nonutility Power Producer: A corporation, person, retail rate reforms and new incentives for production of agency, authority, or other legal entity or instru-electricity by cogenerators and users of renewable mentality that owns electric generating capacity and is resources.

not an electric utility. Nonutility power producers include qualifying cogenerators, qualifying small power Pulpwood: Roundwood, whole-tree chips, or wood producers, and other nonutility generators (including residues.

independent power producers) without a designated, franchised service area that do not file forms listed in the Code of Federal Regulations, Title 18, Part 141. Pyrolysis: The thermal decomposition of biomass at high temperature in the absence of oxygen.

Operations and Maintenance (O&M) Cost: Operating expenses are associated with operating a facility (i.e., Quadrillion Btu: Equivalent to 10 to the 15th power Btu.

supervising and engineering expenses). Maintenance expenses are that portion of expenses consisting of Qualifying Facility (QF): A cogeneration or small power labor, materials, and other direct and indirect expenses production facility that meets certain ownership, incurred for preserving the operating efficiency or operating, and efficiency criteria established by the physical condition of utility plants that are used for Federal Energy Regulatory Commission (FERC) power production, transmission, and distribution of pursuant to the Public Utility Regulatory Policies Act of energy. 1978 (PURPA). (See the Code of Federal Regulations, Title 18, Part 292.)

Peaking Power: Generation used to satisfy demand for electricity during the hours of highest daily, weekly, or Refuse-Derived Fuel (RDF): Fuel processed from seasonal loads (demands). municipal solid waste that can be in shredded, fluff, or densified pellet forms.

Peak Watt: A manufacturer's unit indicating the amount of power a photovoltaic cell or module will produce at Renewable Energy Source: An energy source that is standard test conditions (normally 1,000 watts per regenerative or virtually inexhaustible. Typical examples square meter and 25 degrees Celsius). are wind, geothermal, and water power.

Energy Information Administration/ Renewable Energy 2000: Issues and Trends 111

Renewable Portfolio Standard, RPS: Mandate that Transmission System (Electric): An interconnected ensures that renewable energy constitutes a certain group of electric transmission lines and associated percentage of total energy generation or consumption. equipment for moving or transferring electric energy in bulk between points of supply and points at which it is Ribbon Silicon: Single-crystal silicon derived by means transformed for delivery over the distribution system of fabricating processes that produce sheets or ribbons lines to consumers, or is delivered to other electric of single-crystal silicon. These processes include edge- systems.

defined film-fed growth, dendritic web growth, and ribbon-to-ribbon growth. Turbine: A machine for generating rotary mechanical power from the energy of a stream of fluid (such as Roundwood: Logs, bolts, and other round timber water, steam, or hot gas). Turbines convert the kinetic generated from the harvesting of trees. energy of fluids to mechanical energy through the principles of impulse and reaction, or a mixture of the Silicon: A semiconductor material made from silica, two.

purified for photovoltaic applications.

Watt (Electric): The electrical unit of power. The rate of Single Crystal Silicon (Czochralski): An extremely energy transfer equivalent to 1 ampere of electric current pure form of crystalline silicon produced by the flowing under a pressure of 1 volt at unity power factor.

Czochralski method of dipping a single crystal seed into a pool of molten silicon under high vacuum conditions Watt (Thermal): A unit of power in the metric system, and slowly withdrawing a solidifying single crystal expressed in terms of energy per second, equal to the boule rod of silicon. The boule is sawed into thin wafers work done at a rate of 1 joule per second.

and fabricated into single-crystal photovoltaic cells.

Watthour (Wh): The electrical energy unit of measure Solar Energy: The radiant energy of the sun, which can equal to 1 watt of power supplied to, or taken from, an be converted into other forms of energy, such as heat or electric circuit steadily for 1 hour1.157407e-5 days 2.777778e-4 hours 1.653439e-6 weeks 3.805e-7 months .

electricity.

Wind Power Class: A classification method used to Subsidy: Financial assistance granted by the Govern- describe the usable (for electricity generation) wind ment to firms and individuals. resource at a particular site. A classification of 1 denotes the least amount of energy, while a classification of 7 System Benefits Charge, SBC: A non-bypassable fee on denotes the greatest amount of energy.

transmission interconnection; funds are allocated among public purposes, including the development and Wood Pellets: Fuel manufactured from finely ground demonstration of renewable energy technologies. wood fiber and used in pellet stoves.

Tipping Fee: Price charged to deliver municipal solid waste to a landfill, waste-to-energy facility, or recycling facility.

112 Energy Information Administration/ Renewable Energy 2000: Issues and Trends

ML060940443

Text

Navigation menu

Search