P. Blair: Electricity Overview in Figure 2) projects natural gas-fired generation in 2000 to be 17% of total generated electricity while NERC projects only about 9%. Page 3 In the longer term, through 2010 and beyond, the projections of both capacity and electricity generation vary dramatically among the published forecasts. New construction is more important in this time frame and the dominant fuel of choice for such new generation depends on expected demand growth, fuel prices, and regulatory policy, all of which are highly uncertain. Hence, the differences among various long-term forecasts in projected fuel mix result from different assumptions about total electricity demand (and growth in peak versus base-load demand), the performance of powerplants in terms of efficiency and availability, implementation of the 1990 Clean Air Act changes, economic regulatory structure of the electric utility industry itself (perhaps varying by region), the evolution of technology, major policy developments in dealing with reducing greenhouse gas emissions, and, how aggressively natural gas -- the current fuel of choice for new capacity -- can be employed in power generation. The likely fuel mix is particularly dependent on expectations about future energy resource prices and availability. Concerns about fuel availability prompt more diversity in resource plans. Perhaps the largest of these uncertainties is in electricity demand. For example, to illustrate how badly forecasters have done at this projection, Figure 3 shows the collective forecasts by utilities through the 1980s, which overestimated demand consistently for many years to the point where cumulative forecasts represented on the same graph became known as the "NERC fan." To be fair, it wasn't just the utility forecasts reflected in the NERC compilation that were wrong, it was those of virtually all other analysts as well. Figure 4 shows a number of current national forecasts by utilities and others. The primary reasons for the differences in total demand stem from the different assumptions of the forecasting groups about regional and national economic growth, changing population, and demographics. As shown in the figure, the current range of published forecasts is about 1% to about 4% average annual peak demand growth. Simplistically this translates to a range of around a 30 gigawatt (GW1) surplus to a 280 GW shortfall of capacity beyond currently planned additions and retirements by 2010. Even within individual forecasts, the range of uncertainty is typically very high. For example, Figure 5 shows NERC's projection of total electricity demand (summer peak demand) for 1999 with an 80% probability band of 128,000 MW -- amounting to a 100 GW shortfall or a about a 28 GW surplus at the ends of the uncertainty range compared to currently planned additions and retirements. I will discuss the sources and implications of these uncertainties more later, but I believe that this uncertainty drives much of the decision-making in the power industry today. In the old days (1960s for the utility business) of steady demand growth, falling marginal costs (due largely to improving technology) and low interest rates, an excess of new capacity wasn't all that costly, and demand growth would quickly erase the excess. Now, uncertainty dominates -- it is not only greater, but also more important. The tragedy is how long it has taken us to accept any level of uncertainty in forecasting demand and to incorporate it into planning processes. Yet it is now being incorporated in many utilities' planning, as they begin to plan for a range of plausible future scenarios rather than committing to a fixed inflexible plan. When load growth is more sudden than anticipated, as in New England and the Mid-Atlantic Regions in the 1980s, shorter lead-time resources such as demand-side management (DSM) and combustion turbines are called upon. Also, some utilities are performing pre-construction planning and site preparation to reduce the time required to construct new units, in case demand grows rapidly. The uncertainty in 11 Gigawatt equals 1,000 megawatts. P. Blair: Electricity Overview Page 5 load growth provides the opportunity to dramatically expand the role of DSM and smallerscale, shorter lead-time generating technologies (e.g., natural gas-fired combined cycle units) in utility resource plans. The uncertainty in demand, supply performance, and availability of fuels as well as the decisions that must be made in the face of it, accentuate the tradeoff between minimizing electricity costs and providing adequate generating capacity. Just as the costly excess of capacity which affected many regions in the 1980s was neither planned nor desired, it is possible that shortages may occur in some regions in the 1990s. I believe that physical shortages of electricity in the U.S. over the next decade are unlikely. Even as structural changes occur in the industry, seeking a balance between the cost of additional capacity or of investments to reduce demand and the benefits of higher reliability will remain a driving force in utility planning. The cost of avoiding such shortages or excesses of capacity could be quite high, however, unless we can find ways of incorporating efficiency improvements more effectively into electric utility resource plans. In the short term, this is the challenge of Integrated Resource Planning and, without it and more efficient natural gas generating units such as combined cycles, the task of filling unanticipated capacity shortfalls will fall to much less efficient combustion turbines. As a result, the fuel use and costs of preserving reliability could be very high. In the longer term, the role of new technology2, strategies for reducing air pollution and greenhouse gas emissions, and fuel availability will be much more prominent. I will discuss these issues later. 2OTA has examined new electric power technologies in New Electric Power Technologies: Problems and Prospects for the 1990s (1985), Nuclear Power in an Age of Uncertainty (1984), Starpower. The U.S. and International Quest for Fusion Energy (1987), and others listed in attachment A to this testimony. P. Blair: Electricity Overview THE FEDERAL ROLE IN INFLUENCING THE GENERATION MIX Page 6 The second group of questions you posed, Mr. Chairman, addressed the current and possible Federal role in influencing the Nation's power generating mix of technologies and resources. Jurisdiction over activities of electric utilities is shared between Federal and State authority. Since the U.S. power system is highly interconnected, crossing State boundaries, most generators are now subject to both Federal and State rate regulation. In particular the Federal Energy Regulatory Commission (FERC) has authority over prices, terms, and conditions of wholesale power sales involving privately owned power companies and of transmission of electricity at wholesale. The Public Utility Regulatory Policies Act (PURPA) gave FERC expanded responsibilities for the encouragement of cogeneration and small power production using alternative energy technologies. In addition in the early 1980s, the since-repealed provisions of the Powerplant and Industrial Fuel Use Act precluded the use of natural gas in many utility boilers. But much of the Federal influence over the nation's generation mix has come about through environmental regulation such as the Clean Air Act and through the Federal role in commercialization of new power technologies, such as nuclear power, clean coal technologies, and renewables. The Federal government continues to have enormous but diffuse leverage over the nation's generating mix. Federal policy concerns, however, include broader policy interests than those of primary interest to the States. National priorities include security concerns over oil import vulnerability, national environmental, health, and safety regulation, or international competitiveness. It is the sense of urgency about these issues that is likely to dictate the degree of the Federal role in the next decade. We think that the current programs at the State level including competitive bidding mechanisms for new supply and demand resources as well as efforts to incorporate DSM programs and "nonprice factors" such as environmental costs into utility integrated resource plans are P. Blair: Electricity Overview Page 7 crucial. They serve both as experiments for increasing competition in the utility industry as well as tests of our national ability to internalize environmental and energy security costs into market mechanisms. They can be given the chance and encouraged to succeed. If they fail, more aggressive Federal regulation will be likely. The evolving structure of the industry you described in your question to us included all the current trends: independent power, rolling prudence reviews, competitive bidding systems, integrated resource planning, and wheeling experiments. Many of these on-going changes are focussed on increasing competition in the industry. In our assessment report released two years ago, Electric Power Wheeling and Dealing: Technological Considerations for Increasing Competition, we concluded that: "This assessment has not identified any specific reason why competition cannot be made to work well, but insufficient analysis has been done to determine whether benefits outweigh costs overall. It is clear that there are ways of implementing competition that would work very poorly. There are many pitfalls that must be avoided." There are opportunities and risks in any approach, including retaining the status quo. Of particular concern to us in the current debate over industry structural change is the lack of data and analysis. Any proposed change from the existing system naturally raises uncertainty about how well the new system will work. We know that today's power system works, although some believe it to be somewhat inefficient or inequitable. We also know that the system is currently evolving and accommodating increased competition: nonutility generation and competition among suppliers is increasing substantially in many regions of the country; transmission access is also increasing, although to a lesser degree. However, we will not know the actual impact of these changes on the reliability and economy of the power system for years to come. For the most part, the advantages and disadvantages are still speculative. |