National Academies Press: OpenBook
« Previous: Biomedical Research and Development: Measuring the Returns on Investment
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 215
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 216
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 217
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 218
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 219
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 220
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 221
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 222
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 223
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 224
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 225
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 226
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 227
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 228
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 229
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 230
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 231
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 232
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 233
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 234
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 235
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 236
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 237
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 238
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 239
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 240
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 241
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 242
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 243
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 244
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 245
Suggested Citation:"Energy Research and Development." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 246

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

ENERGY RESEARCH AND DEVELOPMENT J ohn ~ . Ahearne* Resources for the Future To address the economic value of federal investment in energy research and development (R&D), this paper will cover three general points: First, what are the stages in progressing from a glimmer of an idea to a commercial product? Second, why should the federal go~renment fund any of these s sages ? Third, what can we learn from past efforts at federal funding? Several stages can be included in Research and de~relopusent. n For example, the Department of Defense (DOD) includes five category es: basic research, exploratory development, advanced development, engineering development, and production. Sometimes a subcategory of test and evaluation is included, but it is really a part of engineering development. As another example, ache Congressional Budget Office has described the stages of energy R&D as research, development, demonstration, and commercialization. These stages are analogous to the last four DOD categories; thus, the Congressional Budget Office excludes basic research. Mere are no clearly defined boundaries between each stage of R&D, overlap is common, and the stages are closer to being part of a continue rancher than being distinct, well-defined categories. However, each stage does have certain characteristics. Specific application is not ~ necessary justification at ache basic research end, whereas it should be considered as a program moves toward demonstration. The funds scenic at each R&D stage increase substantially: Total basic energy science funding is less than 10 percent of the total allocated Deco energy research, development, and demonstration. For a given proj ect, ache funds required as ache proj ect mores from concept to 2 commercial plant increase by a factor of about ten at each stage. Economic analysis is .~s~cifled increasingly as a pro gram progresses through the spectrum of research, development, demonstration, and commercialization, although often ~ ~ is not applied. Final ly, because pro] ect size increases significantly at the demonstration end, political pressures - - to locate the pro] ect in a specific region, to maintain or add funding, or to reduce or end funding- - also increase . * Witch grateful acknowledgement for significant assistance from Joel Darmstadrer; valuable advice from John Deutch, Norman Metzger, Joel Snow, and Karl Willenbrocic; and data collection by Caroline Bouhdili. Responsibility for the conclusions is the author' s alone . - 211 -

BACKGROUND Although warning signs had appeared earlier, the 1973 oil embargo was a shock to the economic systems of many countries. In ache United S tates, energy became a mad or top ic, driven ini tially by long gas lines and later by the political process. The executive branch and Congress focused on energy research and development . Pres ident Nixon announced a $10 billion? five-year funding programs wish Ache goal of achieving energy independence by 1980 . Senator Henry ~ "Scoop" ~ Jackson cour~terproposed $20 billion over ton years. Many sources suggested ways to spend those monies. Some observers questioned whether rational analysis had been applied to the amounts proposed or to the processes by which funds would be distributed. In particular, questions were raised about whether economic analysis had Been sassed in choosing what would be appropriate energy ~ programs. A decade has passed ~ The energy crops is has abated, oil prices have fallen, and energy supplies (of gas and oil in particular) appear p Fanciful ~ at least for the near aced. In the United S tates, the nuclear industry is staggering or dying, the Synthetic Fuels Corporation is being closed down by Congress, and many ir~i~ciatives for encouraging small generators of electricity have beer cut back. But the United S tates will remain an energy- consuming country, and innovative energy technologies could malce substantial improvements in the use of energy and in the national economy. A technology-dri~ren United S tares must rely for future growth on developing new ideas and transferring those ideas into commercial operation successfully. Usually, this rote is ascribed to industry, and, in particular, eo entrepreneurs. Nevertheless, ache federal government has hat substantial involvement in energy-related programs, for example, ehe fission and fusion programs of the national laboratories, and has supported pro grams in energy conser~ra~cion and de~relopmen~c of new techniques for hous ing insulation. The federal government has funded development of windmills, tow-head hydrodynamic generation systems, and other energy systems. Unfortunately, a retrospective look finds several instances in which large funding has led to little obvious benefit. Beween fiscal years 19~l and 19 82, the federal governmen~c spent $34. 7 billion (1985 dollars ~ in research and development to improve energy supply~or reduce energy use (for example, through energy conservation) . ~ This is less than the Mount calculated in some summaries of energy-related items in the federal budget, because the $34. 7 billion does not include nuclear weapons activities of the Department of Energy (DOE) and its predecessors, the Energy Research and Development Administration (ERDA) ant the Atomic Energy Commission (AEC); energy R&D funding through the National Science Foundation (NSF); or funding for basic energy science. ~ Federal funding for energy R&O has matched, approximately, that of nonfederal funding since the early 1970' s. ~ 12 -

In ache years 1971 to 1982, the nuclear research and development programs (nuclear fission, fusion, uranium enrichment, and those of the Nuclear Regulatory Commission) received $14. 7 billion (1985 dollars ), or 42 percent of the total . What has ache nation achieved through these funds ? . A review of the past ten years shows such proj ects as the Clinch River Breeder Reactor (CRBR), canceled after $10 5 billion were expended; ache gas centrifuge uranium enrichment plan~c (GCEP), canceled After $2.6 billion were expended and before operation was achieved; and improvemen~cs in gaseous diffusion uranium plants, one of which is to be closed, for which - about $1. 5 billion were spent. Ano ther maj or rec ip tent o f federal funding has been coal R&D . Fossil fuel projects, with c081 projects predominating, received S6.1 billion ( 1985 dollars ~ from 1971 to 1982, or 18 percent of the total energy R&D funding. These programs funded some work that should have led to commercial use. The mat or federal effort at commercialization of novel fossil energy use has been the Synthetic Fuels Corporation-- also not very successful. For example, the Great Plains synthetic fuel plant has been abandoned by its developers, apparently leaving the federal government with at least ~ $1 billion loss. The government funded an unsuccessful magne~ohydrodyn~ic (~D) proj ect for over eight years at a cost of $450 million. Careful review of past history would uncover additional proj ects that were canceled after being partially under way or closed after completion because they were ineffective or uneconomical. Some federal funding has been fruitful. Proving a hypothesis wrong or a promis ing avenue to be a blind alley is no t a failure in teas ic research- -both advance knowledge . Failure is funding poor experimental design, mediocre work, publication for the sake of publication. Depending on the stage of R&D, finding a technology uneconomical or infeasible may or may not represent ~ misuse of federal funds. The stages of research and early engineering development are for exploration of new concepts and for preliminary determination of economic and ~cechnological feasibility. If these stages are done well, some proposals will fail. lathe failures represent appropriate uses of federal funds, as will the successes. Imprudent use of federal funds occurs when the early stages are not implemented carefully or examined rigorously, and projects advance that should not. BASIC RESEARCH Usually labeled as basic energy sciences within Ache DOE budget, basic research develops fundamental knowledge, and (usually) does not have a specific application as an explicit goal. For example, DOE and NSr help support the application of high- speed computers to quantum mechanical calculations for predicting the behavior of chemical systems. Writing about one-such application, William Goddard - 213 —

described the theoretical prediction and experimental verification of the use of one surface of molybdenum trioxide for the selective catalysis of methanol with oxygen Deco produce formaldehyde, a see? 9 that may lead eventually to the design of new catalytic processes. Basic research need not be a small proj ect. lathe Department of Energy funds the nation's largest accelerator, at Fermilab. that facility has produced many significant results and is, by any scientific standards, ~ successful example of "big science. n Fermilab' s success stems in large part from the ability of its two directors' R. R. Wilson, known widely for his magical ability to get accelerators rumoring, and his successor, Leon Lederman, who before caking over at Fermilab had built an international reputation for large, well. - executed, and innovative experiments . Basic research can be analyzed, criticized, and critiqued. Monies can be allocated wisely or poorly, but how much, and for what, are riot subj ects for economic analysis O This position is similar Deco that of a former president of E. I. du Pont de Nemours (DuPont) who wrote, "We defined fund~en~ca1 research as inquiry into the :0 fundamentals of nature without specific commercial obj entice n He went on to comment on the value of research, but also wrote, "Research is not, of course, a cash-and-carry activity in which a given expenditure can guarantee a given result. ~ Similarly, in a book devoted to exploring the go~rernment's role in funding R&D, ~ commentator described "basic research, . . . research whose strategy is governed by the logic of science rather than potential apply inability, . . . is thus of such a nature that2the area of ultimate payoff cannot be more than vaguely foreseen." This lack of tight links to produces often has concerned critics of research. Several attempts have been made to examine the evider~ce, ti any, that supports direct links. Projece Hindsighr, a report prepared for the Department of Defense in the 1960' s, attempted to trace 20 military products to their origins. Some conclusions were that basic research pays off more than 20 years later; applied research results from hundreds of sources were incorporated in the final products; and no s imple relationship could be fond between the cost of research and the value of the results linked to that research. There is some :4idence that industrial basic research can be linked to productivity. However, the author of this paper knows of no studies protruding similar analytic support for federal funding of basic research, although Harvey Brooks has provided many examples of valuable science performed with federal s~ppor~c, science that can be collected, at least plausibly, to important commercial applications. Basic research usually has been supported across ~ spectrum of political philosophies: — ~ 1 - —

Prudence would also dictate that the federal basic research budget should grow each year at a rate that matches inflation- - and, where poss ible, allows for a small real increase. This proposal logically follows from an agreement that has Deere loped over several decades - - and has been s trongly reaffirmed by the Ford, Carter, and Reagan Adm~nistra~cions-~that the primate sector will not support basic research at an adequate leve 1 0 Support is not unanimous, and Milton Friedman is a notable opponent: What ethical justification do you have for extracting tax money from people for purposes char do not yield them some greater benefit? I challenge you to find a single study j ustifying ache amount of~,oney now being spent on government support of research . A major problem in trying to provide quantitative support for basic research is the time lag between such research and commercial application of the resulting knowledge. Typically, this ~ ag is more than 20 years Has concluded by Projece Rindstghr), when both congressional memory and explict~c links connecting the basic research to subsequent applications have become blurred. line director of Cor research noted recently that 20- 30 years are required from the start of basic research on a new technology to reasonable market penetration. For example, research on glassy, or Amorphous, metals first was funded 2S years ago, leading aven~cually ~co new methods for forming metallic glasses. Glassy metals now are used in recording and playback heads in tape recorders and in the starting circuits of fluorescent lights, and new forming techniques may lead to more substantial uses. Basic research cannot be valued directly. Perhaps like some aspects of environmen~cal quality, such as clear skies in ache Rockies or virgin forests in the Pacific Northwest, it is only through secondary measures that one can measure the value of basic research. Such techniques as contingent valuation ~ for example, willingness to pay) are used for economic analysis of en~riros~mental quality. Bas ic research might be measured economically by such measures . But whereas most citizens can understand, and may appreciate' clear skies, few can understand the areas of basic research. Superstring theory is of bubbling interest today, and debates are strong within ache science community on funding the superconducting super collider--but it is doubtful ~cha~c the public ever will have the level of knowledge deco endear into such debates. Fun~eental.ly, support for basic research is sub; ective, based on ache conclusion echoic basic research leads to important public benefits. - in ideas generated, people trained, and intellectually stimulating environments provided. Critics of basic research find support from a public skeptical of the value of federal funds being used for "research," often because - 215 -

of weil-public~zed failures of costly programs in the later stages off cue R&D cycle. Consequently, the next sections will focus attention on tho s e s sages . WHY FED F:~RALLY ANDES) Rho? Thy should ache federal government be involved in nones ic research and development? Although specific obj ectives of national administrations obviously differ substantially, the fundamental goals for research and development have not changed much over the las ~ Ban years. A 1984 description states, The Federal Government funds R&D activities to serve two broad purposes: · To meet specific Federal Go~rerT'ment needs--where ache principal user of the R&D is ache government itself- - for example, to insure a strong national defense; · To meet broad national needs--where the Federal Government supports R&D that the private s ec tar lacks incentive to invest in adequately, in the national inters- - to help insure the strengy2 of the economy and the quality of life for all people. This statement is not substantially different from that of the Carter Admir~is~ra~ion in 1979: Research, development and demonstration programs figure prominently in fulfilling energy supply The Federal Government subsidizes research and de~relopmen~ when circumstances such as the uncertainty or the inability :~ capture future benefits inhibit priorate industry efforts. Many writers have described and discussed reasons for the federal goverr~ent to fund research asked development that could lead to national use . Table 1 outlines reasons given by fly ve sources, two from the early 1970' s and three from the 1980' s . Both the s imilarities and the differences are important . Tilton24 is an economist; the Research and Development Act is a political creation; Deutch is a former director of energy research and undersecretary of DOE, and a chemist; the members of the Energy Research Advisory Board2 are technologist, primarily, but writing within political constraints; and Brooks is a technologist writing on public policy. Four of the five sources agree that one appropriate role for government funding of R&D is where there is no pri~ra~ce developer (or — 7 i6 —

l - ~ ~ At so z ¢ ~ z 4~ ~ ~ z - - to - ~ Lo JO ~ ~ - c) At ~ ~3 ~ - At ~ - ~ o - ~ - o - - o c - - - ~ o ~ ~ - - ~ - ~ At - ~ - ~ L O _ ~ ~ _ ~ O — C ~ _ ~ Ye _ ~ :-— _ C ~ O o — C C ~ E ~ Q _ ~ ~ ~

developers as a group within the antitrus ~ laws28 ~ that can foresee sufficient benefits to ~chat developer to justify investment in the research. Tilton gives examples.across a broad spectrum Of industries within and outs ide the energy sector, such as earths moving, catalysis, combustion, and construction technologies. Brooks gives as a class ic example the aeronautical research conducted by the National Advisory Committee on Aeronautics (NACA) after World War ~ and con~cinued by the National Aeronautics and Space Administration (NASA) (also cited by Mowery jO These examples are at ache research end of the research, developmen~c, demonstration, and commercialization continuum. Such proj ects tend to cake a long time to fruition, but also are not particularly expensive (although construction of some of NASA's wind tunnels was) . Bell laboratories performed this type of work for decades, in fields focused on communications, but including such diverse efforts as measuring the cosmic 3-degree-Kelvin background noise, which is related to the "Big Bang" theory of cosmic creation. (this research began witch the goal of eliminating noise in commercial long distance circuits. The researchers were led to examine a persistent noise source, which they found come From outs ide the earth. Bell Labs allowed them to follow this research, which led eventually to the Nobel Prize. It is not clear that an unregulated incus try, in which the Bell Laboratories parent now lives, can afford such public services. ~ Four of the five sources agree that this generic area, in which, in ache economist' s terms, there is incomplete appropriability of benefits, is appropr~ ate for governmen~c funding All five sources agree that federal funding is appropriate for research on national problems, including Rep) programs for goods that cannot be quan~citati~reLy economically justified. Deutch describes this as due deco a Security premium. n 1 Examples of national benefi~cs include a cleaner environment, a more stable economy, and the availability of orphan drugs. Another benefit may be maintaining industries or technologies that are described as being essential for some public purpose, usually national security. Linking some of these benefits directly to research projects is difficult. For example, how best to improve U.S. energy security and how to stabilize the economy are matters that have been debated extensively for at least ache last 12 years with no substantial consensus among either the public or the "exper~cs. ~ Four of the five sources also agree that it is appropriate for the government deco fund research ant developmen~c when risk is high. The economist, Tilton, believes ~chat this case has not been proven. He argues that there would be only a smaller number of high-risk pro; ects qualifying for public support, because the more risky pro] acts are concentrated at the basic research end of the research and development spectrum, and the more expensive pro] ects are at the development end. n He also writes, "Whether firms on average actually have higher discount rates for risks than society as a whole, however, is uncertain. Indeed, it is not entire ly clear why they should. n3 Deutch counters with several points ~chat, although made - 2 18 -

in the context of synthetic fuels, apply to the full range of energy R&D . He notes, "The most persuasive reason for federal ac lion is the evidence of inaction b ,, the private sector.... The evidence is quite unambiguous that the private sector will not undertake the massive initial investments without federal assistance. The time horizon for incus try almost always is much closer than a societal benefit analysis would support ~ forestry is an exception) . Consequently, even though low present-value benefits are calculated, the government should give a higher weight Deco future benefits than industry does--a form of Deutch' s security premix. In addition, research and development, by its very nature, whether for energy or other fields, has a pace set primarily by the ability of the research~rse This has been described as the pace of science or the pace of technology, as opposed to the pace of economics . I t is trite that additional funding can solve problems more rapidly--NASA has demonstra~ced this in some of the early space programs. On the other hand, a substantial infusion of funds does not buy an equivalently increased Petrel of complacence. leers are a finite number of good researchers and a limited number of good research institutions. Increased funding may bring in only lower quality people and groups, and not add significantly to the rate at which useful work is done. Other observers have given arguments similar to those in Table in support of federal funding of R&D. The Office of Technology Assessment proposed: "Or~e approach to accelerating; development would be to increase or concentrate Federal R&D efforts on "technologies... where cost and performance are of greatest concern. "3 An earlier report by the Energy Research Advisory Board concluded that federal research support is required "where there is reasonable certainty that timely and adequate3~esponse by industry and commerce is an unrealistic assumption." These conclusions are not restricted to the United S rates . A recent report from the Intel ~~a~cional Energy Agency noted that market forces may be inadequate to develop appropriate solutions in a timely manner when national interests - - such as energy and economic security, health and safety research and regulation, environmental concerns, development of technology infrastructure for future industry, employment, industrial and regional development, and defense- - are concerned. Over the past IS years, each maj or energy- related company, such as Westinghouse and General Electric, has invested su~bs~cantial funds in research and development. As was noted by a participant in a 1974 conference, Public Policy and Energy R&D, hosted by Resources for the Future, it is likely that industry-rela~ced R&D has more direc~c applicability than goverslmen~c-related R&D. Me participant commented That a review of defense and NASA funding showed that the number of patents resulting per dollar of governmen~c3~&}) was only 10 percent of Chose resulting from priorate industry R&D. Of course, this comparison may not be completely fair, s ince federal agency funding does not have getting a patent as an obj active.

A major development was the recognition by the electric and gas industries that Red) can be a substantial aid if not an absolute necess icy . This led to the founding of the Electric Power Research Institute (EPRI) and the Gas Research Institute (GRI ~ . However, as shown in Table 2, government funding domi Dates that of both EPRI arid GRI. This is due partially to the fact that both industrial organizations are funded by regulated uttli~cies. The Energy Research Advisory Board has pointed out that "regulators are often unwilling to authorize risky endures. Furthermore, the benefits of successful risk taking are unlike ly to accrue to the util ity, while the penal~cies of unsuccessful risk taking have often been borne by its stockholders. n There will be some Red) that ache United States will need that indus~cry will not fund because sufficient benefits will not accrue to an industrial developer or because She cast will be too large for anyone but the federal government. 3 A second area of gradually incseas ing importance is where government regulation or responsibility is involved. Traditionally, nati anal defense R&D was seen to be an appropriate, if not necessary, area for federal funds. With increased technological demands on social regulatory agencies (the Nuclear Regulatory Commission and the Environmental Protect' on Agency the foremost), federal R&D also seems to be necessary where government action could lead to increased costs for Aldus try, for example, research on acid rain, pollutant hazards, and nuclear safely. Finally, federal funding may be necessary where successful R&D could undermine significantly the profits o~ existing industries, for example, research on cogeneration t~ch:~olog~es and end-use efficiencies . APPLIED RESEARCH Applied research retains some of the basic research flavor of advancing knowledge, but applied research projects should be connected to specific identifiable applications. The Department of Energy and its predecessors have funded a variety of applied research program. "More applied research is carried out ire Federal. laboratories than anywhere else. n For example, recognizing that imp roved transmission of electricity can be a major factor in deesreloptng the U.S. electrical power network, ache government has funded a superconducting cable pro: ect at the Broo~aven National I^boratory, research at the Los Alamos National Laboratory on superconducting magnetic techniques ~co imp rode transmission efficiency, and test programs to evaluate the effects -on animals of the high-voltage electrical fields around transmission lines . In the L970's, as use of solar energy became a national goal, DOE developed programs on photovoltaic properties of thin films, concentrators, and advanced materials. lathe eventual payoff from — 290 —

i.~L~ ~ Floral and Industry Funding for Selected Energy Research arm. Bee e iopment 19 8 2 ( S .Yil 1 ion) R&D Funding Area Department Electric Power Gas Research of Energy Research Institute Institute Elec~cr~c-related 2, 189 supp ly 141 Liquid- and gas-rel ated 447 39 36 SUMP tY Conservation and 392 64 38 improved end use utilization . Source: The Federal Role in Research and DeveJopmenc, A Report of - the Energy Research Advisory Board co the United Stares Department of Energy. DOE/S - 0016 . Washington , {)C: Department 0 f Energy, February 19 8 3 . _ 9? 1

~ hose programs could be in economically useful solar systems. Since some o f the programs have been funded for nearly ten years, perhaps it is time for a careful survey of what results have been transferred to demonstration or commercialization. Much applied research consists of small, unglamorous proj ects, which may, however, have a significant effect on the national economy or qua City of life. A good area to study is conserva~cion, in which DOE has funded many proj ects, including the following: a method for trapping and then burning paint solvent fees in metal coating processes; a method for using foam rather than water in textile finishing; a study of heat transfer between buildings and the ground; studies of insulating materials; development of ceramic materials for use in combustion en,g~nes; and development of electric storage batteries . 'whether any of these programs has led to usab le results should be part of the evaluation. Positive DOE d~script~or~s should be accepted as the views of program advocates and treated with. neutral skepticism. Applied research can have influence beyond that planned originally . For example, the Atomic Energy Commiss ion and. itch regulatory successor, the Nuclear Rogue atory Commission, funded research in basic geoscience and in seismic design and analysis. The purpose was to develop seismic criteria for nuclear power plant location and design. One recipient of this funding was Nathar: Newmaric, professor of civil engineering at the University of Illinois. ~ Professor Newmark ~crained scores of graduate students in seismic design, and chose students went on deco design buildings around the world, including many :]j the buildings that did not fall in the recent Mexican earthquake. As another example, ache use of radio isotopes in diagnostic and therapeutic medicine has spread throughout the world, based upon techniques and products developed through earlier AEC and ERDA funding. lathe Coccal funding for those proj acts was small, but the rasul~cs were substantial . Basic and applied research are nibbling processes--they are not done in great leaps ~ eas fly seen and costed ~ whose effects on the products and services in the economy can be predicted and measured. The president of Dupont wrote, "The average research man pays his way by the 'bit-by-bi~c' research, the day- to-day effort that produces results which over a short period of time seem inconsequential, but which over the long run are extraordinarily important. n4 Nevertheless, criticism of federal research funding will not disappear, nor should it. Therefore, the government would be wise to invest some effort in studying at least selected areas of applied research to determine what, if any, national value has been added. Solar energy and conservation programs would be appropriate areas to examine for the effectiveness of funding over the last decade, and seismic programs for ache last two decades. All three are areas in which large commercial firms. ~ such as General Electric, AT&T, IBM) are not dominant, so the threads from federal funding may be more apparent . - 222 —

BORG ~ PROS ECTS The large proj ects that begin to appear at ache end of applied research dominate federal R&D funding and political vis ibility . Whether the proj ecus are called R&~) or demonstration often depends on political decisions. Nevertheless, they are included in the federal R&D budget. Analysis of federal R&D funding cannot avoid addressing why such projects are supported, by what criteria their c~ntribu~cions should be judged, and what can be learned from chose already funded. Reiss has given an elegant description of ache current state of economic analys is in this general area and concludes, ~ Current economic measures of returns deco federal R&D at most provide crude historical statements about the contributions of federal R&D.... [C]urrent methodologies simply do not allow us to say that because the returns to energy R&D in the late 1970's were miserab61e, me should expect these to be miserable in the future. n45 Mowery4 and Hertzfeld demonstrate the accuracy of Reiss' s first statement. Examination of the large energy proj ects may lead to an understanding of why they failed and, perhaps, enable a change to be made in Ache underlying features of the U. S . approach to energy proj ects, so that Ache future will not be like the past. As federal proj ects get large, political pressures increase, 48 particularly pressures by those who do not understand the technology i;~red. Susceptibility deco such pressures is increased by the relatively small size of the proj ects in relation to the total federal budget. Although $50 or $100 million in a year can be overwhelming deco those in the fields affected, amid the debate on a $l trillion budget, a $100 million iced may go unnoticed. Therefore, pressures often are applied outside the notice of news media and other political observers. As Deu~cch notes, "Understanding the balance sheet of who gains and wh:9pays is important to understanding the political fate of a program. n As discussed earlier, nuclear programs have dominated federal energy R&D funding, and several of the programs cited most frequently as problems are in this category. However, federal nuclear R&D also can be linked to national benefits. Nuclear power now provides approximately 15 percent of all electric generation in ache United States and is approaching 100 gigawatts (Go) of capacity. During the period 1971 to 1982, in which some $15 billion (in 19BS dollars) were expended on nuclear research, many nuclear power plan~cs were under construction or beginning operation. In the period 1973 to 1984 (a two-year lag has been taken to allow some diffusion of results ), 58 nuclear power plants received operating licenses, with a combined capacity of 50.7 GU. O (three Mile Island-2 has been excluded. ~ An underestimate of the construction cost per plant in 1985 dollars would be $2,000 per kilowatt. (Palo tlerde, a successful plant completed recently, is estimated to have a cost of $2, 245 per kilowatt. ~ Using $2, 000 _ ;~3 _

per kilowatt for :0.7 GO gives approximately $101 billion in capit~l inves cment . A crude comparison to the resul tics expected in incus try from R&D can be made using the fact ~cha~c from 1921 to 195S, for each $1 spent on2research, Dupont spent $ 3 for new plants, products, and processes S Of course, the nuclear Aldus try also spent a subs tantial amount in develop ing products during the period j ust described. Taking one half of the $100 billion, or SSO billion, to relate to government R&D, ache Dupont 3 - to- I, applica~cion- to- research ratio gives approximately $17 billion as being appropriate, compared Deco the S IS b illion that was spent . Of course, much o f the R&D on which the numb ear planes are based occurred in ache 1950' s and 1960 ' s . However, the nuclear safety research was directly applicable, and the attempt here is co provide ~ compare son based on research as a percent of capital investment . This comparison does not ~ ustify the spending, but does indicate ~cha~c the amount of dollars spent in nuclear energy research has not been out of line with the related industry' s capital investment. However, several features of that spending deserve examination.. The first is represented by the criticisms addressed to several of the large proj acts . In is not clear that the federal government has a process deco identify and manage large energy R&D projectse The radioisotope technology development and most, if not all, of ache research that led to improvements in nuclear power were small proj aces, strati at least by the many hundreds of millions or billion dot lar scale Of the large projects criticized above. Second, then appears deco be an inherent b ias agains ~ some types of research . This is reflected not only In nuclear versus non-nuclear arguments, since for many years there has been criticism of the United States' choice of the light water reactor, and then the liquid metal breeder reactor, instead of gas reactors. In an understatement, Deutch notes, "There is evidence to suggest that the federal government, partially as a result of poor judgment and partially as a insult of political pressure, '~ill not always adopt sound projects. n There also is growing criticism from supporters of solar energy--both political and technical--who argue that the government has subsidized nuclear energy unfairly. Although nuclear support has been productive, the large dollar amounts spent on such projects as the Clinch River Breeder Reactor could have been spent much better elsewhere. As far back as 1955, Greenewalt wrote, "I wonder what our position would have been today had amounts of money and effort equivalent deco those expended g: atomic energy been devoted deco ache utiliza~cion of solar energy. ~ That same s tat amens could have been made in 1965 and in 1975, and i~c can be made Acidly. A principal difficulty facing any large government pro] ect is main~caining support over many years. Instability of political support and shifting world conditions weaken continuity. In reviewing al~e~ative energy programs, Teich has noticed, frost were never given a chance to succeed or fail on technical grounds. As in so many ocher instances of federal civilian R&D,5ibe programs were cut off before a real assessment could be made. _ 924 -

~ Border director of the EPRI fossil fuel and advanced systems division has said chat the time required to get the first commercial plant on line is about fin years from the time the decision is made to design and build it. Industry also has difficulty investing in R&D programs that take ten years to complete. Tilton noted that in the 1960's, manufacturing firms expected about 90 percent of their R&D to pay off in less than five years.: Industry can invest in projects that take a long time to complete, but these usually are at the commercial stage. For example, utilities have done this consistently on large base-load plants, although most utilities, at least in the past, have been guaranteed a ra turn on such investments. In recent years, as utility commissions have raised the prudence argument, many utilities have stopped investing in expansions that will take a decade to complete. This regulatory questioning is similar co some congressional criticism of energy projects. The federal government attempts to identify projects in which the private incentive to fund the research or the development is insufficient to give reasonable confidence that the project will be developed. These projects should be aimed at meeting a foreseeable national need. In the energy area, projects may relate to increasing the supply of energy, to increasing the absolute amount available or its distribution, or to decreasing the demand for energy through load management, conservation, and improved efficiency. In some cases, the risk is tco great for tudus try to see a project as a worthwhile investment. In ether cases, the retune on the investment is likely to be spread across many companies and, therefore, not able to be captured for profitable return by a single company interested in inves tiny in the research. Finally, in some cases, the need is seen to be so far in the future that company decisionmalcers are unable to we ight ache need great enough to support current investment . Brooks notes that the government, as it moves into the demonstration or engineering development phases5gf a project, often neglects to analyze future markets objectively. In the past, the government frequently has attempted to use project sharing with industry as a mechanism to ensure sound market analysis. The underlying theory is that industry will not put up its own funds for projects for which it does not see a long-term market need. There are hazards associated with this approach, particularly in estimating the amount of industry contribution necessary to ensure that the project has undergone industrial scrutiny for potential markets. (Of course, there is always the problem that merely having the market analysis conducted by industry does not guarantee that it is sound. Many segments of American industry today are suffering from the lack of sound forecasts.) It was this approach--that industry was heavily involved--that the government used for many years to justify continuing the Clinch River Breeder Reactor. However, of the $258 million pledged, 24 utility companies accounted for 53 percent of the funds by pledging - 225 -

an average of $~.7 million each. The other 699 utilities pledged an average of $174, 000 each. 39 Furthermore, the u~cilities' original commitment was fixed. As the cost of ache proj ect rose from approximately $SOO million to estimates of close to $3 billion, the utilities' contribution remained at $258 million. Both liberal and conservative administrations have seen a need for federal government investment in energy research and development. A review of the failure rate of federal R&D proj ects does not differentiate noticeably between politics' phi' osophies of administrations. Me failures, however, do not deny ache validity of a need for nationally funded R&D proj ects . In general, the time horizon for private industry in~restmen~c is too short to support ache high-cost R&D needed deco develop maj or new energy technologies for the late 1990' s or early 2000' s . There is sufficiently high probability chat the United S tates will need such technologies to warrant an examination of Ache reasons that previous efforts have failed and, in particular, what new approaches can be taken to improve the probability of success. REASONS FOR FAILIJ~E The failures have had many causes. These include: . Ideologically chosen pro, eats, or continued suppo.- based on ideological grounds far past the point at which pruden~c management would have cancated ache project (CRBR is an example ); · Political pressure based primarily on geographical interests ~ for example , the AD proj ect ); A poor understanding of the current state of technology on the part of government managers (for example, ISABEl7-E, the ill- fated Broken accelerator proj ect); An inadequate obj score examina~cion of future possibilities for both energy supply and demand ~ for example , GCEP); · Imprecise or erroneous cri~ceria for success: promis tng economic feasibility, which depends on world economic condi~ciorm at the time of pro] ect completion, ins~cead of technological. feasibility, which should be based on precarious R&D programs; Control of funds by program advocates, with limi~ced offers ight by any review board; Instability of funding; and Ins~cabili~cy of research direction, related to frequent changing of principal policymalcers. - 226 —

I ~ Ah no ~ addressing failed proj ects, Tilton, in 1974, noted several ?rob leas wich federal R&D funding that are related to those l is bed above: . D is torsion ~ n motivation, · Inflexib ili~cy, . Central teat ion o f dec is ionmaking, · Bias in proj ect selection, Instability of government R&D funding, and · Reduction of private R&D. 60 prob hems: More than ten years Tater, Brooks identified several s imilar . A tendency for government to hang on too long, or to dis tore the commercial j udgmen~c of the private sector by o~rerpro~notion; lathe more attractive ache social benefits, ache greater the likelihood that the pri~rate~sector will see a commercial op?or~cuni-Y, but, also, the more political pressure there wi~- be for the government to ensure the real iza~cion o f the benefits in -he shortest possible time; and Inadequate analysis of the market. 6t Deutch has identified another problem:- incompetent management. He notes that the conventional government practice is direct federal R&I) contract support, which "always requires a good deal of governmental involvement in proj egg management which at best obscures and sometimes ruins performance." The proj ects listed previously as failures suffered from several o f these faults . lihe Clinch River Breeder Reactor was uneconomical when first proposed. James R. Schlesinger, then Assistant to the President for Energy, testified: At the time ~ was Chairman of ache AEC I told the staff deco go away and to bring me a cost/benefit study on ache demo plant by itself, and one could not emerge from such a study with a positive benefit/cost ratio, simply looking at the demo plant in isolation as an R&D experiment. It had to be embedded in an entire program of commercialization. So the Clinch River plant Scorned out deco be in~cagra: to the program of con~rnerciatiza~cion. In order, in that study, to get favorable cost/benefit ratios, one had to assume perfect foresight [for example, a very high rate of growth of electric power demand and the breeder being the only technology — 927 —

option for the flu cure j . Perfect foresigh~c is not readily availed He to mankind, but it was embodied in that particular study . World events and U. S . reactions made ache prod ect even less economical, but, by the mid 1970's, it had become a symbol, as bell as a large proj ect. Thus, inflexible and distorted motivation fac .ors prevented address ing the economic analys is . A member of the Council of Economic Advisers, who had developed a detailed familiarity with C8BR economics as a member of ache National Academy of Sciences Committee on Nuclear and Alternative Energy Systems (CONAES) study, wrote in 1977 about "a growing body of evidence which supports that tbe case for the LIFER ~ liquid metal fast breeder reactor] program cannot be supported on economic grounds. n ~ (Emphasis in original. ~ I-est anyone now argue that the Clinch River Breeder Reactor was not viewed as an R&D program, it should be noted that in 1977 the Comptroller General of ache Uni Cad States reaffirmed that "ache I^~BR program should be clearly identified and recognized for what it is: a research and development progr~m.~ At the same t~ me, ache Comptroller General, perhaps unconsciously, identified the symbolic character of the program as he wrote: The President' s decision to defer indefinitely ache Clinch Differ Breeder Reactor. . . does not coincide with those pos~tio-.s we have taken in the past. In our view, this country, should not now abandon the nuclear f ission option nor6'hould it abandon the I-MFI3R research and development effort. ~ The causes for ache other failures vary. The magr~ecohydrodyn~mic facility received its maj or impetus from its location- -Montanan -and was seen by many backers as support both for Western coal interests and for an extremely well-respected senator. The diffusion plant improvement proj eats were in support of government co~ercia' operations . But they did fit Brooks' description: In the United S tates itself there are many examples of go~rerr~ent technical initiatives in housing, transportation, [and] energy. . . that have faked in the market, largely because they were primarily motivated by the recognition of the technological opportunity without adequate analysis of the market. In ache 1970' s, the go~?ernmes~t supported three enrichment improvement programs simultaneously. While funding ache hi. 5 billion cascade improvement and operating programs, the governmen~c started the gas centrifuge plant ~ canceled after expenditures of over $2 billion), which was a combination demonstration project and government commercial venture, and also funded three advanced isotope separation proj ects at about $50 million per year. None of the programs was subj ect to an economic analysis to determine what would be the rest market f or uranium enrichment services . Local politics ~ in Tennessee and Ohio ), international issues, nonproliferation — ~ so

-~._erns, and the power of the nuclear option block in Congress kept all enricnmer.^c programs al ive. The influence of Congress can be seen in Table 3, which shows the funding for breeder and coal research and de~relopmen~c proj ecus . During the Carter years, coal was favored and the breeder disdained. Congress supported increasing the coal budget, but resisted strongly the Carter Administration' s attempt to cut back the breeder program. In ache Reagan years, the roles have reversed. The White House has attempted to reduce coal funding sharply. Congress has gone along reluctantly, but has kept funding substantially above ache Reagan Administration' s request. On the breeder, however, the opposite is true. Finally, Congress concluded that breeder funding should be cut back drastically, but the Reagan Administra~cion tried to maintain it. Table 3 al so ~liustrates Tilton's point that federal program funding has high instability. For R&D programs, which by their trery nature have a time to fruition of five deco ten years, the substantial changes indicated by Table 3 can be devas~ca~cing. Even when funding is not cut as drastically as the administration or Congress would like, staff in the program offices that face funding cutbacks realize clearly that they cannot malce long- teem commitments wisely, and, frequently, they leave for ocher opportuni~cies. S ice may be an additional reason for the cited failures . The proj ects were simply too large for government control. Although often thought of as an overwhelming force, the federal government s more like ~ large collection of single-ceiled creatures - - it moves almost blindly, in slow, disorganized motion. The government succeeds in small proj ects - -but so does priorate industry. We pres iden~c of Control Data Corporation reported that small companies produce 24 times more innovations per dollar invested gad 2- I/2 times more inno~ra~cions per emp loyee than do large companies . This is consisten~c with Mansfield' s findings Chat although the largest firms do most of the research, "they generally seem Deco carry out a disproportionately small Share of the R&D abed at entirely new products and processes. n6 When government funding gets large, political pressures or incompetence may lead deco failure, use Ily in attemp~cs to develop commercial products. Brooks has concluded that much of the influence of government on the development of commercial technology has been irtad~rertent,6 and even a report from the Energy Research Advisory Board has observed, "Overall, few benefits of federa)Otechnology transfer programs have been noted in the past.... ~ Given all scheme problems, to what extent can economic analysis be brought to bear on future decisionmaking in energy research and development, recognizing Chat energy R&D may not be much different from any other R&~? If one accepts the argument presented earlier that economic analysis cannot be applied effectively to basic research, what standards can be used for the later stages of Rag? If the pas ~ approach has done poorly, are there options that can do better in the future? — 229 —

l ~ l ~ e ,= C | ~ ~ ~ ~ ' ' ' ~ ~ u c~ (~ ~ l - c.~ ~ 1 , !~ ~ =- ~ ~ ~ ~ ~ , o ~ , ~ ~ ~ , ~ ~ ~ ~ , , 5- ~ g r~ ~ I, 1 ~ I ~ LO ~ — C =' 1 ~ ~c C == 1 £= mC 1 ' 1 ~= '' ~ - — . . ~ c . .` ~n ~ | , j ~ E , C ~ ~ I !°~ o-o~;C° , ~ o o 0 · c C; , ~ ~ ~ C) ~ C C E ~ ~ | ~ ~ trl = ~ ~ ,~~ 0 53 ~~ ~ c O ~_ ce ;= 53 C —~r' I I ~ :, ,~, 0 g =, c, . . . C _= I ' (e_ ,~,,,,,-~,,—c' c o~ e ~: I ~ . 1 3, ~ _ _ c ~ c ~ _ ~ ~ ,~~ ~ ~ 4) _ C~ ~ _ ~ .~ ,} ~—— = ~ _ _ _ ~ ~ ' ~ ~ ', o. C' C ~Cl ~ k~ ~ ~ Z I ~ ~ ~ C~ {~~ ~ C O _ ~ z,Z ~ O O ~ , ~ ,-= _--~` , . S=S C: I ~ | O 0 8 ~ —' '—C—_ _ L. 1 , 1—_ c o 0 = c: = c c _ _ 1 ' 1 ~ _ - o 0 — =_—_ 2I Z Zo | ~ I ~ _ ~ ~ ~ <l,~ _ _ _ _ ~ —e I _ ' I e ~ Og I u, = ~ - - t' ~ ~ ~, ~ | --= ~°` :~ -°= = ~=c I , | = a = - ~ ~ - ~. C I ~ ~ ~ L ~ ^; ~ ~ ; . . _ i = ~ ~ ~ ~ ~ ~ ' ~ ! ~ _ ~ ~ ~ c~ EE ~ ~ c' o ° 1 1 . 1_, - ~ ~ ~ - o' - ~ - tC ~ ~ ~ ~ ~ O ~ ° O O O e o I ~ ! o ~ C ~ ~ ~ ~ ~ S ~ ~ ~ E · E E ~ ~d S ~ 1 "~ ~ to 0 ~ = ~ to ~ ~ ~ e — ~ ~ ~ ._ ~ ~ C ~ ~ Y ~ i ~ ~ ~ . . . .= . . . r~ ~ ~ ~~~= t~ ~ ~ , ~, E ~ ~ J ~ , ~ _ ~ r~ ~ ~ ~ _ ~ C~ ~ ~ ~ ~ _ ~ '_ ~ ·_ ~ Co ~ ~ CO 53 ~ 0 _ E_ — I — — — — — ~ — — — — — >~! O — 730 —

ALT-=RNATIVE APPROACHES The three stages suitable for government involvement in a proj ect are exploration of concepts (basic and applied research), proof of concepts (development), and test of industrial application ~ demonstration) . The fourth stage, co~ercializa~cion, is inappropriate for ache U. S . government. (If government is the final user, as it is of military products, then a different set of national goods is involved, with different, or an least additional, criteria. ~ In the three suitable stages, government funding can range from little to all, and federal involvemen~c can vary from hands-off to direct proj ect management. Four questions should be addressed by the executive branch before entering each pro; ect stage (and congressional oversight shout d check chat these questions have been asked): ~ t) Blat are the potential benefits from ache program for ache United S traces ? If federal funds are involved, national benefits should be the goal. For example, national benefits might be energy security, an improved economy, or meeting national defense needs. (2) That are the likely costs? Ellis . simple question is ache bete noire of program advocates. ~ Uncertainties should be considered in answering both (~) and (2), and there should be a real' cation hat the future may be similar to the last 60 years, chat is, a series of "surprises. n ~ 3 ~ lies the assessmen~c screened out political factors, symbolism, and prior-commit:men~c bias? If slot, are scheme acknowledged explicitly, so their influence can be understood? (4) Chat criteria will be used deco judge success, and on what are those criteria based? As pro] ecus progress into the later stages of development, and before they enter into demonstration, the following also should be answered: (5) What are Ache performance characteristics and market condi~cions at prod ect completion that are necessary for economic fees ability? (6) What Petrel; of indust:ry participation should be required? ~ Is a large amount needed to guarantee commercial usefulness of Ache pro ~ ect? ~ ~ 7 ~ What Should be the re latitude share of the risk for government and incus try? 7 For examp le. should it be equal, so that if the proj ect costs rise, the industry contribution also rises? If - 231 -

industry has the principal rot e at the demonstration stage, particularly if the go~rernmen~c is only a funder wi Oh no management role, should industry' s share increase if ache project overruns? ~ 8 ~ Is the proposed management capable of handling the prod ect? Normal managers can handle "human sizes pro] eats . Extraordinary managers are required for billion-dollar proj ects . (Wilson and Lederman of Fermilab, Low and Gl~nnan of Apoll o, and Groves and Oppenheimer of ache Manhattan Proj ect fit the Batter description. ~ These questions are simp] e; in fact, they are obvious . However, to paraphrase Chesterton, it is not that they have been tried and found wanting, it is rather that they have been found do f--cult and left unacted. It is easy to ask questions. To answer them usefully requires objective analysis by know' edgeable people--a rare commodity. To use this list effectively will require unusual objectivity by the executive branch and rare restraint by Congress. Answering the questions also would address a research need suggested by Brooks: More effective and systematic ways of estima~cir~g the i' kely social returns . . . and comparing,. . ~ the appropriabie private return " He 8iSO suggested research on ~ institutional arrangements capable of blending market oriented and socially orie,2~t decisions about strategy and technological development. ~ Currently., federal energy research and deve topment is funded almost ent' rely by the Department of Energy. Al~cernatives have been suggested, including the following: ~ ~ ~ ~ elimination of most, if not all, federal R&D funding for large proj ects . This would take a stop Coward reducing federal expenditures at a time of great budget pressure, and tt would recognize the failures of the past and assess the future as being not likely to be different from the past. It would retain some small pro; acts, which both supporters and critics of federal R&D see as federal successes . Accepting the high- risk argument of Table I, this approach would not bring new energy technology into being when needed . (2) An expanded federal research and development agency, which would attempt: to develop ant re~cain or deco bring in more competent program Angers. However, such an agency would be even more subject to congressional pressure and targeted projects, and inherently would have the problems associated with centralized management, which often include an unwillingness to be ob; ective . (3) Art expanded national laboratory system, in which energy research and development would be conducted essentially by full time government ( or government contract) employees. This arrangement could provide greater program stabili~cy, but it would suffer from the. same conservatism and bias Coward continuation of existing programs - 232 —

chat affl ice the national laocra~cories c.urrencly . As a recent review of federal laboratories s rated, "There are Bogy opportunities for low quality research in pedestrian subj acts .... n . ~ 4 ~ An independent bus. federally funded board, which would review pro; acts proposed to it for possible, partial, or complete funding. The board could be similar deco one set rep recently to study the health Effects of diesel emissions fiche Health Effects Institute ) .14 However , examination of the Synthetic Fuels Corporation ~ SFC) experience should provide insight into the pitfalls of such a board. ( S ) A semi - autonomous agency, b lending COMSAT and a scaled- down SFC, with- - if it could be done- - a carefully designated R&D missions Deutch has described several advantages inherent in a nongovernment entity, particularly ache ability to involve the private sector through the use o55such indirect financial mechanisms as loan and price guaran~ceese However, it is difficult to see this type of agency as any~ching floors than another ae' empt at ~ Department of Energy. Thus, not one of these alternatives has obvious success build in. However, there are features of several of them that would be beneficial *co incorporate into the current DOE approach. CONCHES ACES This paper has not examined in detail most of the research and development projects funded by the government in the pass 20 years. Hence, scheme conclusions are tentative. However, based on what has been reviewed here, and on an understanding of botch current U.S. energy needs and the status of energy technologies, the following conclusions are offered: (~) Knowledge developed in basic research can take 20 to 30 years to become visible in successful commercial products or deco have significant national impact. (2) Basic research has strong anecdotal support; however, it is not amenable to standard quantitative economic analysis. (3) Although the boundary between basic research and applied research is-blurred, the latter in general has a more specific potential connection to application. Applied research also can be labeled exploratory or even advanced development. (4) Many relatively small projects have been funded in applied research. However, little research has barn done either to support or to criticize this generic category of R&D. - 233 -

( 5 ) Although basic research usually is inexpensive, applied research and advanced development can become expensive A technical success can be an economic failure. ~ 6 ~ The largest dollar amounts per program go to engineering development proj ects, which also may be called demons~cration projects. These have the highest public visibili~cy and hay-e been studied more than any other energy R&O category. ~ 7 ~ Large federal energy proj ects have a tow success rate . (~) Federal funding for research and development should be tested against several crite_ ia: (a) lithe proj ect should be designed to meet national needs . A reasonable case music be made for how the proj ece will help the nation. For examp le. proj ects supporting programs of federal responsibility such as environmental regulation are an acceptable class . (b ~ Industry cannot capture enough of the benefits from the project, but a security premium makes it worthwhile for the nation. ~ c ~ The risk involved is too high for primate funders because the proj act is very costly, and there are large technical, demand, or regulatory uncertainties. (9) There are several questions that should be addressed as a proj epic moves through research, development, and demonstration. ~10~ In assessing ache answers to these quests one, an independent panel of knowledgeable people would be useful as an objective buffer agains t b ias and pressure . Individuals wi tech incus trial experience should be included. The Health Effects Institute model would be useful here. (1~) Demonstration programs mat be subjec~c to economic analysis Deco estimate possible markets ob] actively and Talc the harsh light of reality on favorite concepts. (12) Industry participation should be required for demonstration projects, with the percentage of participation increasing with proj ect duration. (13) Federal management should be restricted to the research and development stages, witch industry management for the demonstration proj eats lathe government is much better at establishing policy- -what should be done--than in saying how it should be done, or doing it. Thus, the gover:=en~c should begin deco step out during development, and cer~cainly should not be in~rot~ed in demonstration. The private sector is more likely to get together people who can handle large proj ecus (not always, as the Edsel proj ect and several nuclear plants - 234 —

show--but failure there may be a pitfall of a regulated industry). If Congress (or an administration) deems ache absence of government into lvement during deve lopment and demons t. ation no ~ Deco be prudent, then the government should at leas t contract out the management and pay an appropriate salary. For example, if ache government plans a several billion dollar project, then go~rerr~ment should be willing to pay the manager $SOO, 000 a year, five second-le~rei managers $3SO, 000 a year, and 20 third- Petrel managers $200, 000 a year. In constant dollars, given a ten-year proj ect life for a successful large those salaries will cost S62. 5 million- -highly cost proj ect, those salaries will cost $6Z ~ ~ million- -highly cost effective if a multibill~on dollar pro; epic succeeds . (variations might be included; for example, some percentage of the salary ~ such as 30 percent) could be held in a deferred account, payable upon successful completion of the project. ~ (14) For the demonstration stage, governmen.t should explore further such indirect funding mechanisms as tax incentives, re gula~cory exemp tions, and anti trus t waivers . (15) Finally, the government should ingress in understanding the problems of the past, to learn how better to address ache future. As the Energy Research Advisory Board has no ted, "The technology base programs do not include any research in economics . . . which could be of considerable value in determining how and why decisions are made. maj ority of the Board recommends that the Department ~ of Energy] initiate such research.... n ~ ~ . - 235 - . . .A

NOTES AND REP ERE^NCES Energy Research, Developmen A, Demons era Lion, and Commercialize Scion . Washington , tC: Congress iona] Budge t Office, January 1977. 2. Using slightly different terminology for the stages°-scient~fic feasibility, engineering development, engineering demonstration, and commercial plant- - the costs have been estimated Deco be about $i million, $10 million, $~10 mill ion, arid $i bill ion for a new process. Se-e George R. Hill. "U. S. Fossil Fuel Indus~cry The Situation in the Year 2000: Coal. ~ In Technology and che Feature of U.S. Indusery. Edited by James H. Gardner from a lecture series at ache University of Utah. Indianapolis: 'White River Press, Inc . 1985, pp . 72 - 76 . 3. John E. Tilton. U.S. Energy R&0 Policy: The Role of Economics. Washington, 1)C: Resources for ache Future, Inc., September t974, pp . 3- 7 0 This paper uses the GNP deflator, as recommended by the National Science Foundation. This factor has been criticized by Mansfield. See Edwin Mansfield. "~&D and Innovation: Some Empirical Findings. ~ In Rho, Parents, and Producrivi~~. Edited by Zvi Griliches. Chicago: University of Chicago Press, 1984, pp. 129-130. Pedera] ^~ Fading for Energy: Fiscal Years 2972-84. NSF 3 - 30' . ~ashing~con, DC: National Science Foundation, February 1983, Table 3 (modified), p. 3. 6. Ibid, p. 2, ant Science and Technology Dara Book. NSF 84- 331. Washington, 1)C: National Science Foundation, October 1984, Figure 2. Colin Norman. unradical Surgery in Uranium Enrichment, Science, (June 2l, 19853, p. 1407. 8. Stuart lions. Synthetic Fuel Plant Scuttled, n New York Times, (August 2, 19SS), p. DI. . William A. Goddard IIT . Theoretical Chemistry Comes Alive, Engineering and Science, (September 1985), pp. 3-7. 10. Crawford H. Greenewalt. "The Slow Steady Way of Progress. In The Fabulous Future. New York: E:. P. Button, Inc., 1955, P. 108. 11. Ibid, p. 102. 12. Tilton, op. cit., p. 96. — 236 —

1 3 . Raymond S . I s ens on . Prod ec t Hinds igh ~c, Final Repor' . Washington, DC: Office of the Director of Defense Research and Engineering, October 19 6 9 . 14. Mansfield, op. cit., pp. 123-129 (see Reference 4 above). 15. Harvey Brooks. "Returns on Federal Investments: Me Physical Sciences," this conference. 6. Claude E. Barfield. Science Policy from Ford co Reagan: Change and Cont~nui ty. Washington, DC: American Enterprise Institute, 1982, p . ll3 . 17 . Nicholas Wade. "shy Government Should Not Fund Science, " Science , Sol . 210 ~1980), p . 33 . 18 . The Ro] e of the Deparemen ~ of Energy in Energy Research . Unpublished speech, Cornell University Symposium for the Centennial of Electrical Engineering Education, Washington, DC, April 25, 1985, p. 3. . DOE Annual Report to Congress. DOE/S-0010/84. Washington, OC: U. S . Department of Energy, September 1984, p. 81. 20. Italuing Environmental Goods: A State of the Art Assessment of the Con cingent ~Jaluar~on Method. Edited by Ronald G. Castings. New York: Rowman and Allanheld, forthcoming. 24. For a well articulated defense of basic research funding, as well as some ~choughes on how deco develop an economic rationale to support such, see Leon H. Lederman. "The Value of Fundamental Science, ~ Scientific American, Sol. 251(Novamber 1984), pp. 40 -47 . Budget of che Uni ted States Government, Fiscal Year 1984, Special Ana]ys~s A. Washington, OC: U.S. Office of Management arid Budget, 1983 . 23. The United Scaces Budget ~ Brief, Fiscal Year 1979. Washington, DC: U. S . Office of Management and Budget, 197S, p. 36. . 24. Tilton, op. cit. 25. John M. Deutch. Prospects for Synthetic Fuels in eke United States. Boulder, CO: Colorado Associated University Press, 1982. 26. The Federal Role in Energy Research en c! Development, A Repor~c of the Energy Research Advisory Board to the United States Deparemen t of Energy. DOE/S - 0016 . Washington , DC: Department 0 f Energy, February 19 8 3 . - 237 -

27. Harvey Brooks. -~ Science Policy and Commercial Innovation, " The Bridge, Vot. XJ(1985), pp. 7-13. 28. The U. S. antitrust laws often have been described as prohibiting collections of companies from j oining together to fund research proj ects . Several years ago, i9 U. S . companies in ache computer and semiconductor fields formed the- Microelectronics and Computer Technology Corporation deco work on computers of the future, specifically in competition with Japan. The National Cooperative Research Act of i984 was writ~cen deco ease the antitrust barriers to such prod eats . See William C. Norris . "Cooperative R&D: A New Striated, ~ Issues in Science arid Technology, (Winter 1985), p. 95. 29. Silicon, op. cit., p. 29. 3 0 . Madrid C . Mowery . ~ Federal Funding o f Research and D eve lopment in Transportation: lithe Case of Avia~cion, n thiS conference. 31. In the context of synthetic fuel programs, the security premium n is the difference between the market price for oil (which is the teas is for pri~ra~ce sector decis ions ~ and the value to this nation for reducing ail imports. n See Deutch, op. civic., p. 12. 32. Briton, op. cit., p. 126. 33. Deutch, op. cit., pp. 12-13. 34. New Electric Technologies. Washington, DC: Office of Technology Assessment, 1935, p . 3 . . Federal Energy R&D Priorities, A Report of che Energy Research Advisory Board to the United States Department of ~Energy. Washington, DC: Department of Energy, December 1981, p. 7. 36. En erg' Technology Policy. ISBN 92-64-1268880. Paris: In~cernational Energy Agency, 1985, p. 13. 37. Icon, op cit., p. Il7. 38. The Federal Role in Energy Research and Development, op. cit., p. 3. 39 . AJ-l large pro; ects need not be government funded. In the 1~960 ' s, IBM is repor~cad deco have invested US billion to produce its System 360 family of computers, including over $ 500 million on R&D programs See T. A. Rise. "I.B.M. 's $5,000,000,000 Gamble," Fortune, ~ September 1966 ), pp . tI6, 120 . 40. Federal Funds for Research and Development, Fiscal Years 1980, 29.~2, and 2982, Volume XX~. NSF 82 - 321. Washington, DC: National Science Foundation, p. 4. — 238 -

. Energy Conservation and the Federal Governmerrt: Research, I:evelopmene, and Management. DOE/S-0017. Washington, DC: Energy Research Advisory Board, January 1983. 42. David Okrent. "On the His tow of the Evolution of Light Water Reactor Safety in the United States . ~ U. S . Nuclear Regulatory Commission, Washington, DC, pp . 5 -19 - 5- 21; published as Nuclear Reactor Safety: The discos of tbe Regulatory Process. Madison: University of Wisconsin Press, l98i. See also N. M. Newmaric. A Study of ~Jertica] and Horizontal Earthquake Spec era WASl1- 1255 . Washington, DC: U. S . A-comic Energy Commission, April 1973. 43. Personal communication from Leopoldo Sol's, Deputy Director, Banco de Mexico, S . A. 44. Greenewalt, op. cite, p. 101. 45. Peter C. Reiss, "Economic Measures of the Returns to Federal Research and Development, n this conference, p. 21. 46. Mowery, op. cit. 47. Henry R. Hertzfeld. "Measuring the Economic Impact of Federal Research and Devel opment Investments in Civilian Space Activities, ~ this conference. 48 . ~ or a nega~ci~?e view of the political factors, see Roger G . Noil and Linda R. Cohen. "Economics, Politics, and Government Research and Development, n this conference. 49 . Deutch, op . cit., p . 13 . 50. S',~=ry Information Report. NUREG 0871. Washington, DC: U. S . Nuclear Regula~cor~r Commission, June 1984, pp e 2.10-2 ~ 23 51. ~Palo Verde Cited as Relatively Cheap Despite Pump Problems, Nucieonics Week, ( September 15, 19 83 ), p . 5 . 52e Greenewalt, op. civic., p. 100. 53e Deutch, op. cit., p. 9. 54~ Greenewal~, op. civic., p. 106. 5S. Albert H. Teich. federal Support of Applied Research: A Review of the United Stances Experience, n this conference, p. 15. 56. George Hill, op. cit., p. 75. 57. Til.ton, op. civic., p. 127 - 239 -

58. Brooks, "Science Policy. and Commercial Innovation, n op Cit., p. 8. 59. "Breeder Reactor Corporation Member Systems. ~ Memo from Eric S. Becicj ord, Director, Division of Reactor Development and Demonstration, Energy Research and Development Administration, June 6, 1977. 60. pistons op. Cit., p. 45. 61. Brooks, "Science Policy and Commercial Ismova~cion, ~ op. cit., pp. 7-13. 6 2 . Deu~cch, op . cit ., p . 14 . 63. Testimony of Assistant to the President James R. Schlesinger before the House Commi~ctee on Science and Technology, June 7, 19?7. 64. Letter to Senator Dale Bumpers from 'William 13. Nor&aus, July 7, 1977. 65. Letter to Honorable Olin E. Teague, Chairman, [louse Co~~t~e on Science and Technology, from Elmer B. Stoats, Comptroller Genera' of ache United S cates, June i4, 1977 . 66. Brooks, "Science Policy and Commercial Innovation, n op. Cit. p. 3. 67. Norris, op. ci t., p. 96 (see Reference 23 above) . 68. Mansfield, op. cit., p. 129 (see Reference 4 above). 69. Brooks, "Science Policy and Commercial Innovation, ~ op. civic., p. 9. 70. Industria] Energy Research snd Development, A Report to ache Energy Research Advisory Board. DOE/S-0029. Washington, DC: Department of Energy, May 1983, p. 5. 71. A ~ 976 report for the Department of Commerce noted the difference between cost- sharing and Lisle- sharing, and strongly recommended both for gover~ent-ind~try projec~cs. See Walter S. Baer, Leland L. Johnson, and Edward W. Mersow. Analysis of Federally Funded Demonstration Projects: Final Report. Rand R-1926-DOC. Santa Monica, CA: Rand Corporation, April. 1976, pp. AS, 56, 76. Brooks, "Science Policy and Commercial Innovation, n op. Cit ., p. 13. — 240 —

7 3 . Report or the Phi be House Sc: ence Council, Fed eral Laboratory Review Panel. Washington, DO: Office of Science and Technology Policy, May 1983, p. 9. The Health Effects Institute is funded by automobile manufacturers, fuel companies, and the Environmental Protection Agency. Research funds. are allocated by a committee that decides what knowledge is needed and who can best do the necessary research. The committee issues requests for proposals (REP ' s ~ and awards research con~cracts . When the research is comp Deiced, it is evaluated by a second committee, independent of ache first. Me evaluation committee decides if the work was state of the art, and what follow-on work is needed. 15 . Deutch, op . cit ., pp . i3 -IS . See also Fed eral Financial Support for If gh-Tecnnology Industries." Washington, OC: Congressional Budget Office, June 1985, pp . 57 - 80 . 76. The Federal Role in Energy Research and Development, op. cit., p. 2. _ 94 1 -

r

Next: Impact of Defense Research and Development Investments on the Civilian Sector »
Papers Commissioned for a Workshop on the Federal Role in Research and Development Get This Book
×
Buy Paperback | $95.00
MyNAP members save 10% online.
Login or Register to save!
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!