Research Support for the Power Industry

A revolution is sweeping the electric power industry. Vertically integrated monopoly suppliers and tight regulation are being replaced with a diversified industry structure and competition in the generation and supply of electricity. Although these changes are often termed “deregulation,” what is actually occurring is not so much a removal of regulation as a substitution of regulated competitive markets for regulated monopolies.

Why is this change occurring? Cheap plentiful gas and new technology, particularly low-cost highly efficient gas turbines and advanced computers that can track and manage thousands of transactions in real time, have clearly contributed. However, as with the earlier deregulation of the natural gas industry, a more important contributor is a fundamental change in regulatory philosophy, based on a growing belief in the benefits of privatization and a reliance on market forces. In the United States, this change has been accelerated by pressure from large electricity consumers in regions of the country where electricity prices are much higher than the cost of power generated with new gas turbines.

Although the role of technology has thus far been modest, new technologies on the horizon are likely to have much more profound effects on the future structure and operation of the industry. How these technologies will evolve is unclear. Some could push the system toward greater centralization, some could lead to dramatic decentralization, and some could result in much greater coupling between the gas and electric networks. The evolution of the networked energy system is likely to be highly path-dependent. That is, system choices we have already made and will make over the next several decades will significantly influence the range of feasible future options. Some of the constituent technologies will be adequately supported by market-driven investments, but many, including some that hold great promise for social and environmental benefits, will not come about unless new ways can be found to expand investment in basic technology research.

New technologies in the wings

Several broad classes of technology hold the potential to dramatically reshape the future of the power system: 1) solid-state power electronics that make it possible to isolate and control the flow of power on individual lines, in subsystems, within the power transmission system, and in end-use devices; 2) advanced sensor, communication, and computation technologies, which in combination can allow much greater flexibility, control, metering, and use efficiency in individual loads and in the system; 3) superconducting technology, which could make possible very-high-capacity underground power transmission (essentially electric power pipe lines), large higher-efficiency generators and motors, and very short-term energy storage (to smooth out brief power surges); 4) fuel cell technology for converting natural gas or hydrogen into electricity; 5) efficient, high-capacity, long-term storage technologies (including both mechanical and electrochemical systems such as fuel cells that can be run backward to convert electricity into easily storable gas) which allow the system to hold energy for periods of many hours; 6) low-cost photovoltaic and other renewable energy technology; and 7) advanced environmental technologies such as low-cost pre- and postcombustion carbon removal for fossil fuels, improved control of other combustion byproducts, and improved methods for life-cycle design and material reuse.

Two of these technologies require brief elaboration. The flow of power through an alternating current (AC) system is determined by the electrical properties of the transmission grid. A power marketer may want to send power from a generator it owns to a distant customer over a directly connected line. However, if that line is interconnected with others, much of the power may flow over alternative routes and get in the way of other transactions, and vice versa. Flexible AC transmission system (FACTS) technology employs solid-state devices that can allow system operators to effectively “dial in” the electrical properties of each line, thus directing power where economics dictates. In addition, existing lines can be operated without the large reserve capacity necessary in conventional systems, which can make it possible to double transmission capacity without building new lines.

Distributed generation, such as through small combustion turbines, fuel cells, and photovoltaics, with capacities of less than a kilowatt to a few tens of megawatts, also holds the potential for revolutionary change. Small gas turbines, similar to the auxiliary power units in the tails of commercial airplanes, are becoming cheap enough to supply electricity and heat in larger apartment and office buildings. As on aircraft, when mechanical problems develop, a supplier can simply swap the unit out for a new one and take the troublesome one back to a central shop. Fuel cells are becoming increasingly attractive for stationary applications such as buildings and transportation applications such as low-pollution vehicles. The power plant for an automobile is larger than the electrical load of most homes. Thus, if fuel cell automobiles become common and the operating life of their cells is long, when the car is at home it could be plugged into a gas supply and used to provide power to the home and surplus power to the grid, effectively turning the electric power distribution system inside out. Finally, the cost of small solar installations continues to fall. The technology is already competitive in niche markets, and if climate change or electric restructuring policies make the use of coal and oil more expensive or restrict it to a percentage of total electric generation, then in a few decades much larger amounts of distributed solar power might become competitive, particularly if it is integrated into building materials.

We have grown accustomed to thinking about electricity and gas as two separate systems. In the future, they may become two coupled elements of a single system. There is already stiff competition between electricity and gas in consumer applications such as space heating and cooling. Gas is also the fuel of choice for much new electric generation. Owners of gas-fired power plants are beginning to make real-time decisions about whether to produce and sell power or sell their gas directly. Such convergence is likely to increase. Unlike electricity, gas can be easily stored. To date, most interest in fuel cells has been in going from gas to electricity. But, especially in association with solar or wind energy, it can be attractive to consider running a fuel cell “backward” so as to make storable hydrogen gas.

These are only a few of the possibilities that new technology may hold for the future of the power industry. Whether that future will see more or less decentralization and whether it will see closer integration of the gas and electricity systems depends critically on policy choices made today, the rate at which different technologies emerge, the relative prices of different fuels, and the nature of the broader institutional and market environment. What does seem clear is that big changes are possible. With them may come further dramatic changes in the structure of the industry and in the control strategies and institutions that would be best for operating the system.

Electric power is not telecommunications

It is tempting to conclude that the changes sweeping electric power are simply the power-sector equivalent of the changes we have been witnessing in telecommunications for more than a decade. But although a change in regulatory and economic philosophy has played an important part in initiating both, the role played by technology and by organizations that perform basic technology research has been and will likely continue to be very different in the two sectors.

New technology played a greater role in driving the early stages of the revolution in the telecommunications industry. Much of the basic technology research that provided the intellectual building blocks for that industry was done through organizations that have no equivalent in the power sector. An obvious example is Bell Telephone Laboratories. For historical and structural reasons, the power industry never developed an analogous institution and for many years invested a dismayingly small percentage of its revenues in research of any kind. Even in recent years, firms in the electric industry have spent as little as 0.2 percent of their net sales on R&D, whereas the pharmaceutical, telecommunications, and computer industries spend between 8 and 10 percent.

The aftermath of the 1966 blackout in the northeast, which brought the threat of congressionally mandated research, finally induced the industry to create the Electric Power Research Institute (EPRI). Today EPRI stands as one of the most successful examples of a collaborative industry research institution. But for a number of reasons, including the historically more limited research tradition of the power industry, pressures from a number of quarters for rapid results, and the dominant role of practically oriented engineers, it has always favored applied research. Nothing like the transistor, radio astronomy, and the stream of other contributions to basic science and technology that flowed from the work of the old Bell Labs has emerged from EPRI. Of course, with the introduction of competition to the telecommunications industry, Bell Labs has been restructured and no longer operates as it once did. But in those years when research could be quietly buried in the rates paid by U.S. telephone customers, Bell Labs laid a technological foundation that played an important role in ultimately undermining monopoly telephone service and fueling the current telecommunications revolution.

Bell Labs was not the only source of important basic technology research related to information technology. Major firms fueled some of the digital revolution through organizations like IBM’s Thomas J. Watson Research Center, but government R&D, much of it supported by the military, was even more important in laying the initial foundations. For example, academic computer science as we know it today was basically created by the Defense Advanced Research Projects Agency (DARPA) through large sustained investments at MIT, Stanford, Carnegie Mellon, and a few other institutions.

Some analogous federal research has benefited the electric power industry. Civilian nuclear power would never have happened without defense-motivated investments in nuclear weapons and ship propulsion as well as investments in civilian nuclear power by the Atomic Energy Commission and the Department of Energy (DOE). Similarly, the combustion turbines that are the technology of choice for much new power generation today are derived from aircraft engines. Although the civilian aircraft industry has played a key role in recent engine developments, here again, government investments in basic technology research produced many of the most important intellectual building blocks. The basic technology underpinnings for FACTS technology, fuel cells, and photovoltaics also did not come from research supported by the power industry. These technologies are the outgrowth of developments in sectors such as the civilian space program, intelligence, and defense.

Although one can point to external contributions of basic technology knowledge that have benefited the electric power sector, their overall impact has been, and is likely to continue to be, more modest than the analogous developments in telecommunications. Nor are external forces driving investments in basic power technology research to the same degree as in telecommunications. The communications industry can count on a continuing flood of new and ever better and cheaper technologies that pour into its design engineers as the result of research activities in other industrial sectors and government R&D programs. At the moment, despite a few hopeful signs such as recent DARPA interest in power electronics, the electric power industry does not enjoy the same situation.

Within the power industry, neither the electric equipment suppliers nor traditional power companies can be expected to support significant investments in basic technology research in the next few years. From 1995 to 1996, the electric and gas industry reduced private R&D funding in absolute terms and cut basic research by two-thirds. Of course, many of these companies may increase their investments in short-term applied research to gain commercial advantages in emerging energy markets. Indeed, from 1995 to 1996, dollars spent by private gas and electric firms on development projects increased in absolute terms. In the face of competitive threats from new power producers, traditional power companies understandably have shortened their time horizons and increased their focus on short-term issues of efficiency and cost control. Similarly, most equipment manufacturers are concerned principally with the enormous current demand to build traditional power systems all over the industrializing world. Future markets offered by changes occurring in developed-world power systems lie too far in the future to command much attention.

Putting all these pieces together, the result is that current investments in basic technology research related to electric power and more generally to networked energy systems are modest. Without policy intervention, they are likely to stay that way.

Need for research

What difference does it make if a future technological revolution in electric power gets postponed a for few decades because we are not making sufficient investments in basic technology research today to fuel such a revolution? We think it matters for at least three reasons.

First, there is opportunity cost. The world is becoming more electrified. Once energy has been converted to electricity, it is clean, easier to control, easier to use efficiently, and in most cases, safer. An important contributor to this process is the growing numbers of products and systems controlled by computers, which require reliable high-quality electricity. A delay in the introduction of technologies that can make the production of electricity cheaper, cleaner, more efficient, and more reliable as well as make its control much easier will cost the United States and the world billions of dollars that might otherwise be invested in other productive activities.

Second, there are environmental externalities. Thanks to traditional environmental regulation, the developed world now produces electricity with far lower levels of sulfur and nitrogen emissions, fewer particulates, and lower levels of other externalities than in the past. But the production of electric power still imposes large environmental burdens, especially in developing countries. The threat of climate change may soon add a need to control emissions of carbon dioxide and other greenhouse gases. Eventually we may have to dramatically reduce the combustion of fossil fuels and make a transition to a sustainable energy system that produces energy with far fewer environmental externalities and uses that energy far more efficiently. This will not happen at reasonable prices and without massive economic dislocations unless we dramatically increase the level of investment in energy-related basic technology research, so that when the time comes to make the change, the market will have the intellectual building blocks needed to do it easily and at a modest cost.

Third, there can be costs from suboptimal path dependencies. Will current and planned capital, institutional, and regulatory structures facilitate or impede the introduction of new technologies? System and policy studies of these questions are not likely to be very expensive. But because there may be strong path-dependent features to the evolution of the networked energy system, without careful long-term assessment and informed public policy, the United States could easily find itself frozen into suboptimal technological and organizational arrangements. This, in turn, could significantly constrain technological options in other electricity-using industries.

Mechanisms for research

The most common traditional policy tool for supporting a public good such as energy-related basic technology and environmental research has been direct government expenditure. But in the case of energy, the system has serious structural problems that are not easily rectified. DOE is the largest government funder of energy research. However, most of DOE’s energy budget is more applied in its orientation than the program we are proposing. The DOE basic research program is modest in scale, and for historical reasons much of it does not address topics that are likely to be on the critical path for the future revolution in energy technology. The National Science Foundation (NSF) supports only a few million dollars per year of basic technology research that is directly relevant to power systems.

DOE’s budget is subject to the usual vagaries of interest group politics, which makes it difficult to provide sustained support for basic technology research programs. Support for research in areas with a long-term focus and a broad distribution of benefits is particularly at risk. Although DOE has emphasized the important and unique role it plays in funding such research, and in some instances has a track record of protecting such work, it must carefully pick and choose what to support among competing areas of basic work. Recent pressures on the discretionary budget have further reduced the agency’s ability to sustain a substantive portfolio of basic research, because such programs compete under a single funding cap with stewardship of the nation’s atomic warheads, cleanup of lands contaminated by the weapons program, and programs in applied energy research and demonstration .

The President’s Council on Science and Technology concluded in its 1997 report that the United States substantially underinvests in R&D, observing that: “Scientific and technological progress, achieved through R&D, is crucial to minimizing current and future difficulties associated with . . . interactions between energy and well-being. . . . If the pace of such progress is not sufficient, the future will be less prosperous economically, more afflicted environmentally, and more burdened with conflict than most people expect. And if the pace of progress is sufficient elsewhere but not in the United States, this country’s position of scientific and technological leadership-and with it much of the basis of our economic competitiveness, our military security, and our leadership in world affairs-will be compromised.”

President Clinton’s FY99 request for energy R&D was approximately 25 percent above the funding levels for FY97 and FY98. However, much of the focus continued to be on applied technology development and demonstration projects incorporating current technological capabilities, with relatively modest investments planned in energy-related basic technology research. Congressional reaction has not been favorable.

Given the difficulty that the United States has had in carrying out a significant investment in basic energy-related and environmental technology research as part of the general federal discretionary budget and the obstacles to realigning agency agendas, we believe that strategies that facilitate collaborative nongovernmental approaches hold greater promise. Properly designed, they may also be able to shape and multiply federally supported R&D.

Several mutually compatible strategies hold promise. The first is a tax credit for basic energy technology and related environmental research. Proposals now being discussed in Congress would modify the tax code to establish a tax credit of at least 20 percent for corporate support of R&D at qualified research consortia such as EPRI and the Gas Research Institute (GRI). These proposals are designed to create an incentive for private firms to voluntarily support collaborative research with broad public benefits where the benefits and costs are shared equitably by members of the nonprofit research consortium, where there is not private capture of these benefits, and where the results of the research must be public.

Although such a change in the tax code will help, it is unlikely to be sufficient to secure the needed research investment. For this reason, we believe that new legal arrangements should be developed that require all players in the networked energy industry to make investments in basic technology research as a standard cost of doing business. Why single out the energy industry? Because, as we argued above, it is critical to the nation’s future well-being and, in contrast with other key sectors, enjoys fewer spillovers from other programs of basic technology research.

A new mandate for investment in federal technology research could be imposed legislatively on all market participants in networked energy industries, including electricity and gas. It should be designed to allocate most of the money through nongovernmental organizations without ever passing through the U.S. Treasury. For example, market participants could satisfy the mandate through participation in nonprofit collaborative research organizations such as EPRI and GRI. Other collaborative research organizations, similar to some of those that have been created by the electronics, computer, and communications industries, might be established for other market participants to fund research at universities and nonprofit laboratories.

The long-term public interest focus of such research would be ensured by requiring programs to meet some very simple criteria for eligibility, set forth in statutes. Industry participants should be given considerable discretion as to where they make their research investments. In most cases, they would probably choose to invest in organizations that already are part of the existing R&D infrastructure. Firms that did not want to be bothered with selecting a research investment portfolio could make their investment through a fund to be allocated to basic technology and environmental research programs at DOE, NSF, and the Environmental Protection Agency (EPA). Because of the long-term precompetitive nature of the mandated research investment, it is unlikely to supplant much if any of firms’ existing research.

To the extent possible, the mandated research investment should be designed to be competitively neutral. The requirement to make such investments should be assigned to suppliers of the commodity product (such as electricity or natural gas) and to providers of delivery services (such as transmission companies and gas transportation companies), so that both sets of players (and through them, the consumers of their products) are involved in funding the national technology research enterprise. Because the required minimum level of investment would be very small relative to the delivered product price [a charge of a 0.033 cents per kilowatt hour (kwh)-less than 0.5 percent of the average delivered price of electricity-would generate about a billion dollars per year], it is not likely to lead to distortions among networked and non-networked energy prices.

A presidentially appointed board of technical experts drawn from a wide cross-section of fields, not just the energy sector, should oversee the program’s implementation, establish criteria for eligibility, and oversee operation. Strategies will have to be developed for modest auditing and other oversight. Some lessons may be drawn from past Internal Revenue Service audit experience, but some new procedures will probably also be required. Membership in the board could be based on recommendations from the secretary of energy, the EPA administrator, the president’s science advisor, the NSF director, and the National Association of Regulatory Utility Commissioners. To avoid the creation of a new federal agency, the board should receive administrative and other staff support from an existing federal R&D agency such as NSF.

Our proposal extends, and we believe improves on, the public interest research part of “wires charge” proposals that are now being actively discussed among players in the public debate about electric industry restructuring. Such a non-bypassable charge, paid by parties who transport electricity over the grid, is typically discussed as a source of support for a variety of public benefit programs, including subsidies for low-income customers, energy efficiency programs, environmental projects, and sometimes also research. A number of states are already implementing such charges or are contemplating implementation.

For example, California’s new electric industry restructuring law has provided for about $62 million to be collected per year for four years through a charge assessed on customers’ electricity consumption. The purpose of this charge is to support public interest R&D. Funds are being spent primarily on R&D projects designed to show short-term results, in part to provide data by the time that the four-year program is reviewed for possible extension. New York is considering a charge to collect $11 million over the next three years to fund renewable R&D. Massachusetts has adopted a mechanism to fund renewable energy development, with a charge based on consumption that will begin at 0.075 cents/kwh in 1998 and grow to 0.125 cents/kwh in 2002. This charge is expected to generate between $26 million and $53 million per year over time for activities to promote renewable energy in the state, including some R&D, as well as to support the commercialization and financing of specific power projects.

In 1997, state regulators passed resolutions urging Congress to consider-and EPRI, GRI, and their constituents to develop-a variety of new mechanisms, including taxes, tax credits, and a broad-based, competitively neutral funding mechanism, to support state and utility public benefits programs in R&D, in addition to energy efficiency, renewable energy technologies, and low-income assistance. Several restructuring proposals in Congress, including the president’s proposed comprehensive electricity competition plan, include a wires charge. The president’s program would create a $3-billion-per-year public benefit program for low-income assistance, energy efficiency programs, consumer education, and development and demonstration of emerging technologies, especially renewable resources. Basic technology research is not mentioned. The president’s plan, which would cap wires charges at one-tenth of a cent per kwh on all electricity transmitted over the grid, would be a matching program for states that also establish a wires charge for public benefit programs.

There are two serious problems with state-level research programs based on wires charges. For political reasons, their focus is likely to be short-term and applied, and they are likely to result in serious balkanization of the research effort. Balkanization will result because most state entities will find themselves under political pressure to invest in programs within the state. This will make it difficult or impossible to support concentrated efforts at a few top-flight organizations. Many of the issues that need to be addressed simply cannot be studied with a large number of small distributed efforts.

New carbon dioxide control instruments, now being considered as a result of growing concerns about possible climate change, offer another opportunity to produce resources for investment in a mandated program of basic energy technology research. Carbon emission taxes or a system of caps and tradable emission permits are the two policy tools most frequently proposed for achieving serious reductions in future carbon dioxide emissions. Over time, both are likely to involve large sums of money. Following the model outlined above, a mandate could require that a small portion of that money be invested in basic technology research. For example, in a cap and trade system, permit holders might be required to make small basic technology research investments in lieu of a “lease” fee in order to hold their permit or keep it from shrinking.

Although the mechanisms we have proposed to support basic technology and environmental research are different, they are all intended to be competitively neutral in the marketplace, national in scope, and large enough to fund a portfolio of basic technology research at a level of at least a billion dollars per year to complement and support other more applied research that can be expected to continue as the industry restructures. With the implementation of such a set of programs, the United States would take a big step toward ensuring that we, our children, and their children will be able to enjoy the benefits of clean, abundant, flexible, low-cost energy throughout the coming century.