ENERGY EFFICIENCY AND RENEWABLE ENERGY
IN THE U.S. ENERGY FUTURE
TESTIMONY OF
JOHN P. HOLDREN
FOR THE
COMMITTEE ON SCIENCE
UNITED STATES HOUSE OF REPRESENTATIVES
HEARINGS ON
"THE NATION'S ENERGY FUTURE:
ROLE OF RENEWABLE ENERGY AND ENERGY EFFICIENCY"
FEBRUARY 28, 2001
MR. CHAIRMAN, MEMBERS, LADIES AND GENTLEMEN: I am John P. Holdren, a professor at Harvard in both the Kennedy School of Government and the Department of Earth and Planetary Sciences. Since 1996 I have directed the Kennedy School's Program on Science, Technology, and Public Policy, and for 23 years before that I co-led the interdisciplinary graduate program in Energy and Resources at the University of California, Berkeley. Also germane to today's topic, I was a member of President Clinton's Committee of Advisors on Science and Technology (PCAST) and served as chairman of the 1997 PCAST study of "Federal Energy Research and Development for the Challenges of the 21st Century" and the 1999 PCAST study of "Powerful Partnerships: The Federal Role in International Cooperation on Energy Research, Development, Demonstration, and Deployment." A more complete biographical sketch is appended to this statement. The opinions I will offer here are my own and not necessarily those of any of the organizations with which I am associated. I very much appreciate the opportunity to testify this morning on this timely and important subject.
Introduction
The comprehensive review of U.S. federal energy research and development that I chaired for the White House in 1997 was carried out by a panel of 21 senior individuals from industry, academia, and public-interest organizations. In addition to members with experience and expertise across the full range of energy options--fossil fuels, nuclear fission and fusion, renewable energy sources, and increased end-use efficiency--it included others of senior research, management, and policy-advising experience outside the energy field (including a former chair of the Council of Economic Advisors and a former CEO of Hewlett-Packard), who held no prior brief for increasing federal energy research. The panel concluded (1, p ES-9) that
Energy-technology improvements, achieved in the United States and then deployed here and elsewhere, could:
- ower the monetary costs of supplying energy;
- lower its effective costs still further by increasing the efficiency of its end uses;
- increase the productivity of U.S. manufacturing;
Holdren testimony for House Committee on Science, 28 February 2001, page 8
- increase U.S. exports of high-technology energy-supply and energy end-use products and know-how;
- reduce over-dependence on oil imports here and in other countries, thus reducing the risk of oil-price shocks and alleviating a potential source of conflict;
- diversify the domestic fuel-supply and electricity-supply portfolios to build resilience against the shocks and surprises that an uncertain future is likely to deliver;
- reduce the emissions of air pollutants hazardous to human health and to ecosystems;
- improve the safety and proliferation-resistance of nuclear-energy operations around the world;
- slow the build-up of heat-trapping gases in the global atmosphere; and
- enhance the prospects for environmentally sustainable and politically stabilizing economic development in many of the world's potential trouble spots.
The panel noted that the public benefits of these outcomes, beyond the private benefits to energy firms that invest in achieving them, warrant public investments in energy-technology innovation supplementing the efforts of the private sector; and it argued that the federal government's investments in this category at the time of the review--FY1997--were "not commensurate in scope and scale with the energy challenges and opportunities that the twenty-first century will present," taking into "account the contributions to energy R&D that can reasonably be expected to be made by the private sector under market conditions similar to today's" (1, p ES-1). The PCAST panel recommended, accordingly, a substantial strengthening of the federal energy R&D portfolio, ramping up DOE budget authority for R&D on end-use efficiency, fission, fossil, fusion, and renewable-energy options from a total of $1.3 billion in FY1997 to $2.1 billion in FY2003 (expressed in constant 1997 dollars). The following table shows the distribution of the proposed increases.
Table 1. PCAST-Recommended DOE Budget Authority for Energy-Technology R&D (millions of constant 1997 dollars)
Efficiency / 373 / 755 / 382 / 48.6%
Fission / 42 / 102 / 60 / 7.6%
Fossil / 365 / 371 / 6 / 0.8%
Fusion / 232 / 281 / 49 / 6.2%
Renewables / 270 / 559 / 289 / 36.8%
TOTAL / 1282 / 2068 / 786 / 100%
These budget recommendations--putting 85% of the real annual increment in FY2003 compared to FY1997 into efficiency and renewables--were unanimous, notwithstanding the diversity of energy (and nonenergy) backgrounds represented on the panel and notwithstanding the history of disagreements among the different energy constituencies about funding priorities. The unanimity on the panel emerged from detailed joint review and discussion of the content of the existing programs, the magnitudes of unaddressed needs and opportunities, the current and likely future role of private industry in each sector, and the size of the public benefits associated with the advances that R&D could bring about. Efficiency and renewables received the bulk of the increment because they scored high on potential public benefits and on R&D needs and opportunities unlikely to be fully addressed by the private sector.1
In what follows here, I will discuss, for both the efficiency and the renewables sectors, the opportunities as seen by the PCAST panel, and I will try to explain, in so doing, some of the reasons that PCAST appears to be more optimistic in its stance on renewables and efficiency than the Energy Information Administration is in the forecasts out to 2020 in the latest Annual Energy Outlook (4). At the end, I will offer some observations on measures, beyond R&D, that I believe would be warranted in pursuit of increased contributions from efficiency and renewables and the public benefits that such contributions would bring.
Efficiency
In the period from 1955 to 1970, the energy intensity of the U.S. economy stayed essentially constant, at about 19 quadrillion Btu per trillion 1996 dollars of GDP. From 1970 to 1980, a period marked by the Arab-OPEC-induced oil-price shocks in 1973-74 and 1979, the energy intensity of the economy fell at an average rate of 1.7% per year; and from 1980 to 1985 (at the beginning of which period the real world oil price was nearly six times its 1972 value) it fell 3.5% per year. In the decade from 1985 to 1995, the rate of decline of energy intensity in the United States slowed to about 1.0% per year. From 1995 to 2000, the rate of decline has been 2.7% per year.2
The improvements since 1970 in the overall energy efficiency of the U.S. economy resulted from a complex and changing mix of increases in the efficiency of energy transport, oil refining, and electricity generation, transmission, and distribution; increases in the technical efficiency of energy end-use in space conditioning, household and commercial appliances, manufacturing, and the transport of passengers and freight; and a transition from a more-energy-intensive to a less-energy-intensive mix of productive activities in the economy.3 From the overall numbers alone, however, it is easy to calculate that if the U.S. economy of the year 2000 had been generated at the energy intensity of economic activity that prevailed in this country in the period from 1955 to 1970, the United States would have used 177 quadrillion Btu of primary energy in 2000 rather than the 98 quadrillion Btu actually used. The rate of reduction of energy intensity of the U.S. economy averaged over the whole 30 years from 1970 to 2000 was 2.0% per year.4 The annual energy savings attributable in the year 2000 to this decline in energy intensity compared to the 1955-1970 value (amounting to 177 - 98 = 79 quadrillion Btu per year) is more than two and a half times larger than the total increase in U.S. energy supply from all sources in the same period (which amounted to 98 - 68 = 30 quadrillion Btu per year).
The fact that reductions in demand due to reduced energy intensity of economic activity were far larger between 1970 and the present than increases in energy supply goes a long way toward explaining the interest of the PCAST energy panel (and most other analysts of the energy situation) in determining and exploiting the potential for continuing improvements in energy efficiency in the decades ahead. The other pillars underpinning this interest are the economic and environmental attractions of energy-efficiency improvements, across a wide range of circumstances, compared with available means of increasing supply. On this point the 1997 PCAST report stated (1, p ES-15)
Increasing energy efficiency has an extraordinary payoff. It simultaneously saves billions of dollars, reduces oil imports and trade deficits, cuts local and regional air pollution, and cuts emissions of carbon dioxide.
The PCAST study set forth specific goals for a bolstered program of energy-efficiency R&D in the years immediately ahead as follows (1, p ES-15):
Buildings. To fund and carry-out research on equipment, materials, electronic and other related technologies and work in partnership with industry, universities, and state and local governments to enable by 2010: (1) the constructing of 1 million zero-net-energy buildings; and (2) the construction of all new buildings with an average 25-percent increase in energy efficiency as compared to a new building in 1996. Additional longer term research in advanced energy systems and components will enable all new construction to average 70 percent reductions and all renovations to average 50 percent reductions in greenhouse-gas emissions by 2030.
Industry. By 2005, develop with industry a more than 40-percent efficient microturbine (40 to 300 kW), and introduce a 50-percent efficient microturbine by 2010. By 2005, develop with industry and commercially introduce advanced materials for combustion systems to reduce emissions of nitrogen oxides by 30 to 50 percent while increasing efficiency 5 to 10 percent. By 2010, achieve a more than one-fourth improvement in energy intensity of the major energy- consuming industries (forest products, steel, aluminum, metal casting, chemicals, petroleum refining, and glass) and by 2020 a 20 percent improvement in energy efficiency and emissions of the next generation of these industries.
Transportation. By 2004, develop with industry an 80-mile-per-gallon production prototype passenger car (existing goal of the Partnership for a New Generation of Vehicles C PNGV). By 2005, introduce a 10-mpg heavy truck (Classes 7 and 8) with ultra low emissions and the ability to use different fuels (existing goal); and achieve 13 mpg by 2010. By 2010, have a production prototype of a 100-mpg passenger car with zero equivalent emissions. By 2010, achieve at least a tripling in the fuel economy of Class 1-2 trucks, and double the fuel economy of Class 3-6 trucks.
The report concluded that these efforts "complemented by sound policy, can help the country increase energy efficiency by a third or more in the next 15 to 20 years."
An increase of one third over a 15-year period would constitute an average rate of improvement of 2.7% per year, equal to what the United States achieved from 1995 to 2000 and considerably better than the 2.0% per year 1970-2000 average. An increase of a third over a 20-year period would correspond exactly to 2.0% per year. For comparison, the "reference" scenario in the Energy Information Administration's 2001 Annual Energy Outlook entails an average rate of reduction of energy intensity between 2000 and 2020 of 1.6% per year, along with real economic growth averaging 3.0% per year (4, p 7).5 It is instructive to consider the difference between the EIA's "reference" value of a 1.6% per year decline in energy intensity compared to the lower of the PCAST figures, 2.0% per year (corresponding also to what was actually achieved between 1970 and 2000). When applied to the period 2000-2020 under the "reference" economic growth assumption of 3.0% (real) per year, the EIA rate of decline in energy intensity of 1.6% per year yields primary energy use of 129 quadrillion Btu in 2020; a 2.0% per year decline in energy intensity over this period yields primary energy use in 2020 of 119 quadrillion Btu, cutting 10 quadrillion Btu off the increase. If a 2.4% per year decline in energy intensity could be achieved in this period (still not as high as the 2.7% per year actually achieved for 1995-2000), primary energy use in 2020 would be 110 quadrillion Btu, 19 quadrillion Btu below the EIA "reference" case.
Why was PCAST more optimistic about energy-efficiency potential than the most recent EIA Annual Energy Outlook appears to be? It is important to understand, first of all, that neither of these studies is making unconditional predictions. Their scenarios depend on assumptions, variously explicit and implicit, about a variety of factors that will influence rates of economic growth, rates of technological innovation, and the rates of application of available energy-efficiency technologies. The EIA reference case assumes that the world oil price in 2020 will be about $22 per barrel C compared to $17 per barrel in 1999 but $27 per barrel in 2000 (all of these prices in 1999 dollars) C under world oil production of 117 million barrels per day (compared to 76 million barrels per day in 1999) and an OPEC share of this production reaching 49% (up from 40% in 1999). The highest world oil price in 2020 in any of the EIA scenarios is $28 per barrel (1999 dollars). As best I can tell, moreover, the EIA scenarios do not account for the possibility of policies much more aggressive than today's for promoting energy end-use efficiency, nor for any "future legislative or regulatory actions that might be taken to reduce carbon dioxide emissions" (5, p 6).
The PCAST study did not develop explicit scenarios about future energy prices and policies, but I believe it fair to say that most if not all of the PCAST panelists would have considered the range of possibilities for the world oil price in 2020 to extend considerably above the figures considered by the EIA. The panel also concluded that "there is a significant possibility that governments will decide, in light of the perceived risks of greenhouse-gas-induced climate change and the perceived benefits of a mixed prevention/adaptation strategy, that emissions of greenhouse gases from energy systems should be reduced substantially and soon" (1, p ES-10). And its assessment of what could be achieved in the way of energy-efficiency improvements over the next two to three decades was conditioned on the full implementation of its recommendations for increased federal R&D in this area, as outlined above. The increases in DOE's energy-end-use-efficiency R&D budgets actually achieved since the publication of the PCAST report have fallen considerably below what was recommended, as will be seen in a moment.
Renewables
What have been the changes in the U.S. energy-supply mix and what has been the role of renewables in this evolving picture? The changes in U.S. primary energy supply from 1970 to 2000 are summarized in Table 2. Table 3 shows, in a similar format, the changes in the electricity- generation sector in the past 10 years. Renewable energy contributed 6.0% of U.S. primary energy
Table 2. Changing Composition of U.S. Primary Energy Supply 1970-2000 (energy contributions in quads = quadrillions of Btu)
coal / 12.3 / 22.0 / +9.7 / +2.0%/yr
domestic oil / 22.0 / 16.9 / -5.1 / -0.9%/yr
imported oil / 7.5 / 21.1 / +13.6 / +3.5%/yr
natural gas / 21.8 / 22.8 / +1.0 / +0.15%/yr
nuclear / 0.24 / 8.1 / +7.9 / +12.5%/yr
hydropower / 2.7 / 3.1 / +0.4 / +0.46%/yr
biomass / 1.4 / 3.6 / +2.2 / +3.2%/yr
geothermal / 0.011 / 0.31 / +0.30 / +11.8%/yr
solar / 0 / 0.08 / +0.08 / NA
wind / 0 / 0.05 / +0.05 / NA
Table 3. Changing Composition of U.S. Net Electricity Generation 1990-2000 (generation figures in TWh = terawatt-hours = billions of kilowatt-hours)
/ net electric generation in 1990 (TWh) / net electric generation in 2000 (TWh) / change 1990-2000 (TWh) / average annual percent changecoal / 1590 / 1950 / +360 / +2.1%/yr
nuclear / 577 / 760 / +183 / +2.8%/yr
natural gas / 378 / 620 / +242 / +5.1%/yr
hydropower / 288 / 270 / -18 / -0.6%/yr
oil / 124 / 102 / -22 / -1.9%/yr
biomass / 43 / 74 / +31 / +5.6%/yr
geothermal / 16 / 14 / -2 / -1.3%/yr
wind / 2.2 / 4.9 / +2.7 / +8.3%/yr
solar / 0.6 / 0.9 / +0.3 / +4.1%/yr
in 1970, rising to 7.3% of a larger total in 2000. The renewable share of U.S. electricity generation was 11.6% in 1970, falling to 9.6% of a larger total in 2000. The contribution of non-hydro renewables to electricity generation was 2.0% in 1990, rising to 2.5% in 2000. Clearly, the interest of PCAST and other groups in the prospects for a large contribution from renewables over the next few decades is based--in contrast to the case of energy efficiency--more on hopes for the future than on the experience of the recent past.
The PCAST study noted that the principal obstacle to more substantial deployment of renewable energy options has been the high costs of the energy delivered by these technologies. It found grounds for optimism in the sharp declines in these costs, for a number of the renewable options, over the preceding two decades;6 and it concluded that continued and expanded investments in public- and private-sector R&D on renewables--together with measures to move these technologies along the learning curve through increased purchases under, e.g., renewables portfolios standards--could allow renewable energy technologies to "become major contributors to U.S. and global energy needs over the next several decades" (1, p ES-22). The focuses and goals of the expanded Federal effort in R&D on renewables recommeded by PCAST were described in its report as follows (1, pp ES-22/23):
Wind. Reduce by 2005 wind electricity costs to half of today's costs, so that wind power can be widely competitive with fossil-fuel-based electricity in a restructured electric industry, through R&D on a variety of advanced wind turbine concepts and manufacturing technologies.
Photovoltaics (PV): Pursue R&D that would lead to PV systems prices falling from the present price of $6,000/kW to $3,000/kW in 5 years, to $1500/kW by 2010, and to $1,000/kW by 2020. R&D activities should include assisting industry in developing manufacturing technologies, giving greater attention to balance of system issues, and expanding fundamental research on advanced materials.
Solar Thermal Electric Systems. Strengthen ongoing R&D for parabolic dish and heliostat/central receiver technology with high temperature thermal storage, and develop high temperature receivers combined with gas-turbine based power cycles; goals should be to make solar-only power (including baseload solar power) widely competitive with fossil fuel power by 2015.
Biopower. Enable commercialization, within ten years, of advanced energy-efficient power-generating technologies that employ gas turbines and fuel cells integrated with biomass gasifiers, building on past and ongoing R&D for coal in such configurations, and exploiting the advantages of biomass over coal as a feedstock for gasification. These technologies could be widely competitive in many developing country markets and in U.S. markets that use biomass residues or use energy crops in systems that derive coproducts from biomass.