|
Article Excerpt
"Research and development is our nation's investment in its own future. America's science and technology base may well stand as our most important renewable resource. The overarching public goal of US R&D policy, of which energy R&D is a major component, must be to assure for future generations that our Nation's capacity to shape the future through scientific research and technological innovation is continually being renewed" (US Department of Energy Report). (1)
"In a nutshell, the government that had created this regulated industry was saying, 'We don't want to regulate you anymore. Here's your business. Good luck.' However, the restructuring process initially generated more questions than answers, as the various players in the market tried to understand how the configuration of this industry might need to change." (CEO, TXU) (2)
INTRODUCTION
The two statements above, one by the Department of Energy taskforce and the other by an electric utility CEO, succinctly capture the central ideas on which this paper is based. First, investment in research and development (R&D) activities is fundamental for economic progress (Schmookler, 1966), and energy-related research is especially important. Second, the restructuring of the US power market has created uncertainties about the future business landscape, and electric utilities are trying to adapt as best as they can. This paper analyzes how R&D has fared in this environment. Data shows that, coincident with the movement towards restructuring the electricity industry in the early to mid-1990s, there was a dramatic decline in research activity in this sector in a relatively short period of time. This decline is worrying because it may "result in slowing technology development, sacrificing future prosperity to meet short-term goals, and failing to meet national energy goals" (GAO Report, 1996) and may thus have implications for the future of U.S. energy security. This paper is a step towards understanding how the transition from a regulated to a market regime influences firm R&D behavior.
Four major entities performed R&D in the U.S. electricity sector--the electrical equipment manufacturers (EEMs), such as General Electric (3); the investor owned utilities; the Department of Energy; and the Electric Power Research Institute (EPRI). This paper focuses on the utilities' R&D expenditure, which fell precipitously between 1993 and 2000. (4)
[FIGURE 1 OMITTED]
From its highest level of $741 million (in 2000 dollars) in 1993, R&D expenditure declined to $193 million in 2000--a drop of nearly 74% (Figure 1). (5) R&D expenditures by the other players declined as well. Collaborative research by EPRI experienced a sharp drop (71 percent) as major utilities (6) slashed contributions. State electricity R&D funding declined by 30 percent between 1993 and 1999, and the Department of Energy's funding decreased by 3 percent during the same period. (7) In addition, for EEMs, total R&D as a percent of sales declined as well. To provide a context, we can benchmark these numbers against other industries. In Appendix Figure 1, we provide graphs for the total (real) R&D trends from 1990-2000 for four industrial sectors: drugs, chemicals, machinery, and autos. (8) We find that three out of the four sectors show a steady increase in R&D expenditures over the time period. (9) This illustrates that the changes in the electric utility industry R&D during this period were not due to any general macroeconomic factor but were rather, peculiar to the industry itself. As such, the sharp drop in utilities' R&D expenditure requires further investigation.
The magnitude and timing of these changes raise questions about the forces behind this decline. Since this drop in R&D was coincident with market restructuring, the obvious factors appear to be the changing institutional and market structures in the power sector. At the outset, however, it is important to ask whether such a decrease is cause for concern. To answer this question, we need to briefly address two related issues. First, what are the technology improvements that have taken place in the electricity industry, and what is the contribution of electric utility research to such progress? Second, how has the composition of R&D spending by utilities changed during restructuring, and should we be concerned?
On the technology front, there have been some major breakthroughs. The first is the rise of distributed generation technologies, such as the combustion turbine, fuel cells, and photovoltaics, which have eliminated the need for large scale power plants. The second is the development of the Integrated Gasification Combined Cycle (IGCC), (10, 11) which is a part of the clean coal technology program and can confer significant environmental and efficiency benefits. The third type of technology relates to renewables, such as wind, geothermal, and biomass energy. The fourth innovation is in the creation of demand side management programs to reduce electricity use by consumers. Although a majority of these technologies are being developed by large electric equipment manufacturers, such as GE, utilities have been playing a large role in testing these technologies and improving them.
As an example, in 1994 Southern California Edison spent approximately $1 million a piece on heat-rate improvement of fossil fuel generation and distributed generation; $8 million on renewable technologies, such as photovoltaics; and another $6 million on demand side management programs, for a total budget of approximately $70 million. (12) In 2000, the company was not involved in any of the above projects, and total R&D expenditure fell to just $1 million. Duke Energy Corporation spent 21 percent of its internal R&D budget on fossil-fuel generation research in 1994, and this declined to 9 percent in 2000. During the same period, their total R&D budget shrank from $18 million to $9 million. Yet another example is the Gulf Power Company, which spent about 40 percent of its internal R&D budget on clean coal technologies in 1994 and reduced this expenditure to zero in 2000. With the vast reduction in these budgets, the equipment manufacturers may have to look elsewhere for testing their technology, and this may adversely impact new technology development.
In addition, utilities spent a sizeable portion of their internal research budgets on public interest environmental R&D projects, such as research on global warming and the effects of electromagnetic fields on health. From the data, we find that for most utilities, the large projects with potential externalities, such as air and environmental quality research, have been replaced by smaller and more targeted research projects, such as hydro basic research projects. (13) Again using Southern California Edison as an example, we find that such R&D expenditures have declined from $11 million to 160,000 dollars, a drop of almost 98 percent. In general, we find that collaborative research and classic "public interest" R&D has led the charge in the overall decline in research spending (Hirsch, 1998; U.S. House of Representatives Testimony, 1998). Since such research does not confer any short-term monetary benefits, one would assume that the equipment manufacturers would not be willing to step in and pick up the slack. The combination of declining federal R&D support and the precipitous drop in utility research expenditures does not bode well for such research. These are important concerns, and analyzing how the restructuring policies have affected utility R&D behavior will help in understanding the relation between market structure and the conduct of R&D.
Considerable theoretical and empirical attention has been focused on the linkages between market structure, innovation, productivity, and R&D (14) for manufacturing firms. However, relatively little attention has been devoted to the behavior of firm R&D when such firms are regulated and have to adapt to a competitive threat. The electricity industry restructuring provides a good opportunity to study this question. The restructuring of the U.S. electricity industry is likely to substantially alter innovation incentives and, consequently, affect the level and composition of R&D. We believe that this study has broader implications for the conduct of R&D in restructured industries and may serve as a good benchmark for other regulated industries in transition. In addition, identifying the factors behind the decline may suggest ways in which such funding decreases can be mitigated.
Our results indicate that restructuring, and consequent changes in market and institutional arrangements, are a significant determinant of the R&D response of utilities. In particular, the expectation of restructuring and the early stages of the restructuring process are associated with a decline in research spending. R&D suffers as the specter of competition draws closer and as the level of competition increases. A part of this decline is reversed when an actual restructuring order has been passed and utilities are fairly certain about the emerging institutional structure. However, this increase is offset by a further decline when states actually implement restructuring legislation. In aggregate, restructuring is responsible for a 78.6 percent decline in electric utility R&D expenditures. In addition, a majority of the research is conducted by large generation companies, especially if they are part of a larger holding company. State preferences also influence the amount of R&D conducted by the utilities, and firm and state characteristics have a different impact on R&D before and after restructuring. The derivation and the implication of these results are analyzed in subsequent sections of the paper.
This paper is organized into five sections. The first section briefly outlines the restructuring process and the nature and organization of R&D activities in the US power industry. The second section discusses earlier literature that investigates both the R&D model and the incentives to conduct R&D under a regulated versus a market framework. Methodological issues, data sources, and the specifics about the variables used are presented in the third section. The fourth section explains the empirical results, and the last section concludes.
1. BACKGROUND
The Energy Policy Act (EPAct) of 1992 gave impetus to wholesale power competition by creating a new class of power producers called the exempt wholesale generators and by creating open-access transmission grids for wholesale transactions. In 1996, FERC Orders 888 & 889 (15) encouraged retail competition for the first time while furthering wholesale competition. (16) These were federal guidelines, and each state interpreted and implemented the Orders in different ways. This led to the emergence of a diverse set of state-specific institutional arrangements that governed utilities and this paved the way for altering the existing regulated monopolies into firms that could successfully navigate the new competitive environment (17) (Blumstein, 1997; Borenstein, & Bushnell, 1997; Bushnell & Stoft, 1995; Hogan, 1997; Joskow, 1989, 1997, 1999; Moyer, 1993; Sloan, 1994; Taylor, 1998; Wolak, 1997).
Despite common federal guidelines, deregulation and restructuring of the industry did not occur as a one-time monolithic change. Instead, the processes involved a complex set of institutional changes that ranged from implementing pricing rules and setting up auction mechanisms for the power market to modifying the policies governing various operational facets of electric utilities. The pace and nature of restructuring differed considerably from state to state. Most states, however, went through three common stages of restructuring, starting with the initiation of investigations into the deregulation process and ending with the passage of some form of retail competition order. In addition, several other policies, including those that affected the mandated date for retail competition, the percentage of market open to such competition, the guidelines governing the divestiture of assets, default provider rules, and stranded cost recovery, were important as well. All these policies changed not only the operating environment of the utility, but also the very nature of the firm, and hence have implications for the conduct of R&D.
We are interested in studying electric utilities because they were the entities most directly affected by the restructuring process. Historically, the EEMs have conducted most of the R&D and generated a majority of the innovations in this industry. (18) Compared to these EEMs, the R&D levels for utilities were low since the latter could not internalize the benefits of such research. They allocated some of the research dollars to in-house projects, such as those focused on global warming, and the rest of the money was given to external collaborative research agencies, such as EPRI. Within the cohort of utilities, R&D was primarily conducted by big, vertically integrated firms that owned generation, transmission, and distribution. Companies that were solely distribution or transmission companies invested very little in research. In the regulated environment, R&D investment was essentially a risk-free venture for the utilities since such investments could be recouped. With restructuring, the incentives for R&D changed, and appropriability and cost concerns became important. The next section briefly discusses the motives for R&D in the regulated and restructured regimes and frames the observed decline in light of the changed incentives.
2. LITERATURE REVIEW
2.1 General Determinants of R&D
R&D is considered a critical engine of economic and productivity growth (Griliches, 1979, 1986; Griliches and Mairesse, 1984; Nadiri, 1979; Van Reenen, 1997), and there has been substantial research on the drivers of R&D (Cumming and Macintosh, 2000). Two broad hypotheses have emerged from such analysis. Pioneered by Schmookler (1966), the demand-pull hypothesis emphasizes demand side factors, such as consumers' demand for new products, and cost-reductions as primary drivers of R&D. The supply-push hypothesis, on the other hand, (Rosenberg, 1974) holds that supply-side factors, such as differences in the technological environment of the firm and industry concentration, lead to variations in R&D expenditures. Empirical evidence suggests that neither hypothesis alone can explain firm R&D behavior and that both demand and supply aspects are important (Scherer, 1982; Pakes and Schankerman, 1984; Jaffe, 1986, 1988). Combining the hypotheses, one can point to a common set of factors that are important drivers of R&D spending.
First, a firm's characteristics and organizational structure (Piga and Vivarelli, 2003) are important in explaining its R&D propensity. Firm size seems to be of critical importance, and large firms are more research intensive than small ones (Grabowski, 1968; Lee, 2003). Profitable firms, and those with excess cash, also conduct more R&D (Reynard, 1979; Connolly and Hirschey, 1984). Unionization is an important determinant of R&D, and greater unionization has adverse impact on such spending (Menezes-Filho et. al, 1998). Second, features of the industry within which the firm operates are equally important. Industry concentration, product market competition (Wahlroos and Backstrom, 1982), and exogenous technology accumulation by other firms (Lee, 2003) also explain a firm's R&D behavior....
|
