1、Creating Incentives for EnvironmentallyEnhancing Technological Change:Lessons From 30 Years of U.S. EnergyTechnology PolicyVICKI NORBERG-BOHMABSTRACTDue to the externalities associated with energy production and consumption, public policy is necessaryto provide a stimulus for the development and dif
2、fusion of more environmentally sound energy technologies.Based on an in-depth history of technological development for four electric power technologies, this paperdraws lessons for the design of future policies to promote innovation in energy technologies. The technologiesexamined are: wind turbines
3、, solar photovoltaics, gas turbines, and atmospheric fluidized bed combustion. Theanalysis considers both supply-push and demand-pull approaches for stimulating technological change. Itconcludes that government activities to promote environmentally enhancing technological development mustinclude bot
4、h supply-push and demand-pull policies during the period spanning precommercialization, firstcommercial use, and lead adoption. Furthermore, this analysis identifies five industry sector characteristicsthat influence the level of government effort necessary to support commercialization: the size, st
5、rength, andrisk of the private market niche; industry structure; firm financial capability; firm technological capability; andsources of innovation. 2000 Elsevier Science Inc.IntroductionThe production and consumption of energy satisfies private needs and wants, butin doing so creates negative exter
6、nalities, including environmental degradation andsecurity risks. It also contributes positively to economic and social development. Theseexternalities, both negative and positive, provide the rationale for public involvementin the energy sector. As managing climate change is added to the environment
7、al,economic, and security issues that have led to government intervention in the energysector in the past, designing effective policy to promote innovation and diffusion of thenext generation of energy technologies becomes ever more pressing. This is becauseVICKI NORBERG-BOHM is the Director of the
8、Energy Technology Innovation Project at the BelferCenter for Science and International Affairs, Kennedy School Government, Harvard University. Prior tostarting this position, Norberg-Bohm was an Assistant Professor of Environmental Policy and Planning atMIT, where she was co-PI of the Environmental
9、Technology Policy Project and Co-Director of the Programon Environmental Education Research. She holds a Ph.D. in Public Policy from Harvard University and aBS/MS in Mechanical Engineering from Washington University in St. Louis.Technological Forecasting and Social Change 65, 125148 (2000)2000 Elsev
10、ier Science Inc. All rights reserved. 0040-1625/00/$see front matter655 Avenue of the Americas, New York, NY 10010 PII S0040-1625(00)00076-7126 V. NORBERG-BOHMeffective management of climate change requires a radical departure from our currentways of producing and consuming energy.While the rational
11、e for government intervention is strong, the ability of governmentsto effectively promote technology innovation for commercial goods, such as power-generation technology, remains a daunting challenge. By examining the role of the U.S.government in the development of four electricity generating techn
12、ologies, this paperdraws lessons on how to effectively design policies to stimulate energy technologyinnovation, with a focus on the role of policy in the commercialization phase. Two ofthe technologies examined in this paper, gas turbines and coal-fired atmospheric fluidizedbed boilers, are fossil
13、fuel based. These technologies are currently commercially competi-tive for grid-connected power applications. The other two technologies, solar photovolta-ics and wind turbines, are based on renewables. Grid applications of these two renewabletechnologies are currently subsidized, although they are
14、the most cost-effective technol-ogy for some off-grid applications.This paper is organized as follows. It begins with a brief discussion of the role ofgovernment in technology commercialization. The second section describes the issuesdriving energy policy over the last three decades and the specific
15、 policies implementedin the United States to address these concerns. The third section examines the role ofthese policies in the development of each of the four technologies. This section concludeswith a comparison of the similarities and differences that government policy played inthe development o
16、f each of the four technologies. This sets the stage for the concludingsection, which draws policy lessons from these histories that can be applied as we moveforward in designing strategies for the development of the next generation of energytechnologies, technologies that would be viable in a green
17、house-gas constrained world.The Role of Government in Technology CommercializationIn the United States, there is a consensus that the government should play a rolein basic science. More contentious is the proper role for government in technologicaldevelopment. Although no consensus exists, over the
18、past several years there has beena growth in government sponsored R“Valley of Death” from Helena Chum, NREL, and Irvine Barash, VenCom Management, Inc., personalcommunications 2.CREATING INCENTIVES IN U.S. ENERGY TECHNOLOGY 127Energy, while traded in the market as a private good, has huge environmen
19、talexternalities. To put this another way, some public goods, such as clean air, are dependenton how we produce and consume energy. Such public goods are privately produced.As a private commodity good, energy technologies must compete largely based on price.There are some characteristics that are de
20、sired in addition to a low price. These include:security of fuel supply, fuel flexibility, modularity, speed of start-up and shut-down, andenvironmental performance. These characteristics can be helpful in capturing marketniches. For example, as will be discussed in the next section, gas turbines we
21、re able tocapture the niche for peaking power long before they were cost competitive with base-load coal plants. But as long as environmental requirements are met, the bottom-linecost of electricitymost often rules, making it difficult for new generating technol-ogies to capture a market niche based
22、 on other qualities. This closes off one importantpath to commercialization, quality improvements, that allow a good to charge higherprices to the lead adopters. Under these circumstances, if we want to capture the publicgoods associated with energy use, government action will have to go beyond inve
23、stingin technological inventions (basic R industry structure; firm financial capability; firmtechnological capability; and sources of innovation.Drivers of Power Sector Development in the Unites States: 1970 to 2000Over the past 30 years, three factors have driven technological change in the U.S.pow
24、er sector. The first, security, is a concern that was prevalent during the oil crisesof the 1970s, and that continues to raise its head, as exemplified by the Persian GulfWar in 1992. The second driver is a set of environmental concerns that started in the1970s with the local and regional problems o
25、f urban smog and acid rain, and nowincludes concern about climate change. The final driver is the end of the era of decreasingelectricity prices, brought on in part by the first two drivers.2Together, these driverscreated windows for new power technologies. It is important to note that they alsocrea
26、ted new approaches for providing energy services (i.e., demand-side management).Demand-side management involves both organizational and technological innovation,and is worth examining in the context of policy as a driver of technological innovation, butis outside of the scope of this paper.The polic
27、y responses to this interlocking set of drivers were multifaceted. Policiesincluded both supply-push (stimulating technological innovations), and demand-pull(creating a market for emerging energy technologies). For supply-push, the federalgovernment funded a variety of R as a fraction of GDP, federa
28、l energy R this may signal the beginning of a growth trend inenergy R the report recommended that 32% of this increase occur inFY1999. PCAST made the following recommendations for Rgas turbines and fluidized bed coal boilers are now cost competitive in private marketswhile the two renewable technolo
29、gies still require subsidies to be cost competitive forgrid-generated electric power. Because we wanted to draw lessons about the governmentrole in the commercialization process, this was an important criteria. This criteria ledus to study atmospheric fluidized bed combustion (AFBC) rather than inte
30、grated gasifac-tion combined cycle (IGCC) technology, even though IGCC is believed to be a bettertechnology for the future use of coal, as it will allow more readily for separation andsequestration of carbon 2. It is also worth noting that we did not include nuclearpower in our study because of its
31、unique characteristics that include security concernsand lack of public acceptance in the United States.One limitation of this study is its focus almost exclusively on the United States.Because the U.S. policies have been directed largely toward U.S. firms, the lessonsdrawn from this work are valuab
32、le. Nonetheless, given the increasingly global structureof the power sector supply industry combined with the fact that governments aroundthe world invest in energy R (2)utilities would only be interested in large wind turbines, as they were accustomed topower plants in the 100- to 1000-MW range; an
33、d (3) the large turbines were viewed astechnological umbrella; if they were successful, smaller turbines could be based onthe same design 31. With NASA at the helm, an aerospace philosophy dominated,emphasizing lightweight designs. In hindsight, the rationales proved incorrect, and thedesign focus p
34、roved ineffective. No commercial turbines of the 3- to 5-MW size havebeen built. There were some useful, albeit expensive, achievements through the ModProgram. It was instrumental in demonstrating that medium-scale wind turbines (100kW) could successfully deliver high-quality electricity to the util
35、ity power grid. It alsodemonstrated the ability to operate at variable speed while connected to a utility gridand gathered much experimental data.9This picture is changing. More recent government efforts have led to the development of wind farmsin other parts of the country 2729.132 V. NORBERG-BOHMI
36、n contrast to the NASA-led Mod Program, DOE sponsored innovation in smallerturbines and small turbine components. The Solar Energy Research Institute, laterrenamed the National Renewable Energy Laboratory, administered this program. Ofthe 12 key innovations in wind turbine components that we identif
37、ied in our research,seven relied on partial or total public funding, and three were developed in the privatesector for other industries and transferred for use in wind turbines. We were unable toidentify the funding source for the remaining two 17.As these statistics indicate, the federally funded s
38、upply-push efforts were supple-mented by innovation in the private sector. During the early and mid-1980s, whenincentives to invest in wind energy were highest, a large amount of innovative activityoccurred, and a great many companies began producing wind turbines 33. By 1985,28 manufactures had ins
39、talled turbines in California, half of them foreign firms 35.Without the incentives of the market there would have been only laboratory inventionand no technological change in commercially available wind turbines 34, 36.Because we were unable to find detailed data on the level of diffusion of each o
40、fthese component innovations, we relied on two proxies as indicators of whether windturbines were adopting component innovations: (1) specific yield that measures thepower generated by a turbine per swept rotor area, and (2) decreases in the cost ofwind generated electricity. Both measures indicate
41、that turbine performance improvedthroughout the California “wind rush” of the 1980s, suggesting the diffusion of compo-nent innovation, as there were no breakthroughs in overall turbine design. This conclu-sion is augmented by interviews with companies that were producing turbines duringthe 1980s. T
42、hose with longevity improved their turbine design over time by improvingspecific components.NREL is continuing its work, both by sponsoring advanced wind turbine designsand innovation for specific components 37. These programs both require cost sharingby the private sector. NREL is now also providin
43、g funds for demonstrations to utilitiesthat install small test plants using the latest technology. In recent years, with the fallingoff of the utility market for wind turbines, as will be described below, this has beenthe only way to get the new technologies tested under actual operating conditions.
44、Electricity restructuring is changing this picture. Some states are creating renewableportfolio standards and systems benefits changes to support the adoption of wind turbinesand other renewable energy technologies.10Throughout the 1980s, wind-generated electricity remained considerably more ex-pens
45、ive than fossil generated electricity. Three public policy interventions on the de-mand-side together created the market for this technology: PURPA, tax subsidies, andthe characterization of the wind resource. The wind turbine market grew in California,not because it was the only state with a viable
46、 wind resource, but because of the statesapproach to implementation of PURPA and tax incentives. Californias avoided costcalculations were generous, although they were not the highest in the country. It wasnot until California instituted long-term contracts in 1983 that installations soared. Theselo
47、ng-term contracts made it easier for developers to borrow capital for an emergingtechnology. The gap between Californias avoided cost and the price of electricitygenerated from wind was filled by federal and state tax credits totaling 25%. The years1984 and 1985 were the peak of wind turbine install
48、ation in California. In 1985, thefederal tax credits expired, the California tax credits were reduced, and the long-term10For updated information on efforts to support renewbales in electricity restructuring see http:/www.eia.doe.gov/cneaf/electricity/chg_str/pbp.html.CREATING INCENTIVES IN U.S. ENE
49、RGY TECHNOLOGY 133Fig. 2. Economic Policies and Wind Turbine Annual Installations in California. 1811contracts were eliminated. In 1987, the California tax credits expired. The wind marketdiminished considerably at this point, as shown in Figure 2.In sum, the big science efforts were not effective, but the efforts in small turbinecomponents were. Because the component research was driven by the needs of themarket, the government-created market was important for both public and privatesector innovation. The demand-pull policies were too inconsistent to create a last
