By Frank A. Scott and Mark C. Berger
From Foresight, Vol. 5, No. 2
published 1998
The market for electricity in Kentucky and the United States is undergoing significant change, and it is important that the citizens of Kentucky understand the changes that are likely to take place in the next few years. In order to contribute to the understanding of this issue, last October we completed a study on the effects of deregulation of the market for electricity in Kentucky (Scott, Berger, and Thompson, 1997). In it we conclude that most Kentuckians are likely to see lower prices for electricity. After our study was completed another study was conducted by David Freshwater and his colleagues that examined the effects of deregulation in four rural counties (Freshwater, Goetz, Samson, Stone, Johansson and Greer, 1997). Unfortunately, this study and a later summary of the results published in the last issue of Foresight (Freshwater, Goetz and Samson, 1998) do not promote a greater understanding of the effects of deregulation. If anything, the Freshwater study promotes confusion because of poor scholarship and failure to understand the way prices are determined in a deregulated market.
The most critical erroneous conclusion of Freshwater and his colleagues is that the price of electricity will rise for most Kentuckians under deregulation. They proclaimed this startling finding more than two months after we had completed our study. They chose to ignore our study and did not defend their conflicting results in light of our earlier findings. Because the findings in our two studies were so different, we initiated a joint seminar with them on the campus of the University of Kentucky where we pointed out the weaknesses in their study. Even after this seminar, they still are unwilling to reconcile their results with ours, as evidenced by their recent article in the last issue of Foresight, in which they again fail to mention the fact that there are different results obtained in other studies.
Because they have not done so, we feel compelled to address the opposite conclusions reached in the two studies. In this article, we 1) provide a summary of our study, 2) compare the Freshwater results with ours and in the process show many of the errors in their analysis, and 3) discuss the likely effects of electricity deregulation in rural areas. We conclude by providing answers to the set of questions that Freshwater, Goetz and Samson pose at the end of their recent Foresight article but are unable to answer with their own analysis.
The electric utility industry is undergoing a transformation. There is general agreement among experts that by the end of the next decade the electric utility industry will be characterized by full consumer choice at the retail level, very much like the market for long-distance telephone service today. Access to the power network will be provided by a local "wire service" monopoly, similar to the role currently played by local telephone service companies. The local utility will own and maintain the distribution network and likely handle billing and many other customer service functions. Transmission lines will continue to be owned and maintained by electric utility companies, but a regional independent system operator will handle operation and access to the transmission grid. Utilities, power marketers, independent power producers, and other entities will compete for the business of supplying electrical energy over the transmission and distribution grid to residential, commercial, and industrial customers.
How we get to this end result is another matter. What happens in the transition period will depend very much on what state and federal legislators and regulators do to facilitate or impede change. Here we analyze the developing competitive market for electricity and discuss what can be anticipated to happen in the short run if Kentucky adopts a stance of facilitating change. Then we discuss what the world will look like after the transition when a new long-run equilibrium state is reached.
To understand what competition will look like in the electric utility industry one must analyze the supply of and the demand for electricity. Electricity is currently supplied to customers in Kentucky by a combination of investor-owned utilities, rural electric cooperatives (co-ops), municipal utilities, and the Federal Government. Since electric utilities buy and sell power across a wide territory, the appropriate geographic market area to analyze is much broader than the boundaries of any one state. For our study we have chosen to analyze a 20-state region centered around Kentucky. This 20-state region is shown in Figure 1. These states were selected because they are in the same electricity reliability council as Kentucky or are in adjoining electricity reliability councils.
In 1994 there were 729 utility-owned power plants operating in these 20 states that had a capacity rating of 25 megawatts or greater. As of January 1, 1996, the nameplate rating of existing capacity owned and operated by electric utilities in this region was 378,214 megawatts. Kentucky’s share of that total capacity was 4.6 percent. Electric power generators in the 20-state region produced 1,728.3 billion kWh of electricity in 1995, or 51.3 percent of the nation’s total output. Of that, 1,627.1 billion kWh (94.1 percent) came from utilities while only 101.2 billion kWh (5.9 percent) came from nonutility sources. Kentucky utilities generated 86.2 billion kWh in 1995, and only a negligible portion of that came from nonutility sources. That was 5.0 percent of the regional output, and 2.6 percent of the nation’s output. By way of contrast, Kentucky has only 1.5 percent of the nation’s population.
The buyer side of the electric utility industry in Kentucky is substantial and diverse. Total sales of electricity to final users in 1995 were 65.6 billion kWh. Industrial customers make up the biggest block of customers of electric power in Kentucky. Industrial customers in 1995 bought 48 percent of electric output, residential customers bought 31 percent of output, commercial customers bought 16 percent of output, and the remaining 5 percent of output went to other uses, such as lighting for highways. Just as with supply, it is appropriate to evaluate demand over a wider region than just Kentucky. Unsurprisingly, a large volume of sales occur within this wider region. Total annual sales in 1995 of electricity to final customers in the wider region were 1,524.7 billion kWh.
In a competitive market the short-run price of the product is determined by the interaction of market demand and supply. The demand curve represents the consumption behavior of buyers of the product, while the supply curve represents the production behavior of sellers of the product. The equilibrium price is determined by the intersection of supply and demand, because consumers will want to buy exactly as much electricity as producers want to sell at that price.
The short-run demand for electricity is heavily influenced by equipment choices that households and firms have made in the past. The result is that short-run demand for electricity tends to be inelastic. The short-run supply for electricity is determined largely by utilities’ historical investments in generating units and will resemble a stair-step function. As the price at which electricity can be sold increases, more and more generating units can profitably produce power. The prices at which utilities will be willing to supply power in a competitive market will depend on the marginal cost of producing power, not the historical costs of investments in plant and equipment. If the price of electricity is very low, only a handful of very efficient plants will be turned on and their power offered for sale. If the price of electricity is high enough, even very old and very inefficient plants will be taken out of mothballs and brought on line.
In the near term, depending on how quickly competitive forces are allowed to come into play, the price of electricity will be set by the interaction of short-run demand and short-run supply. Using information on current consumption behavior by residential, commercial, and industrial customers in the 20-state region, we have simulated the short-run market demand for electricity.(1) Using information on variable production costs of the electric generating units in the same 20-state region, we have simulated the short-run market supply for electricity as well.(2) Our simulations refer to the market for electric power, which is one portion of the bill that consumers pay. This is the portion that is likely to be subject to competitive forces in the near future. The typical electric bill also contains charges for distribution, transmission, administrative costs, taxes and other charges. In the current debate no one is arguing that competitive forces should determine these charges.
Figure 2 illustrates the projection of short-run supply and demand for electric power. The projected average short-run price of electric power in the regional market is 2.1 cents per kWh.(3) This is an estimate of the average price across all utilities and customer classes in the region over the course of an entire year, but its validity is supported by the fact that current transactions in the regional wholesale power market occur at prices in this range or lower. In a competitive market, the variability of the customer’s electricity usage will affect the costs that the customer imposes on the generating system. Customers with constant power usage thus may pay less than 2.1 cents per kWh for their electric energy while customers with highly variable power usage may pay more. To obtain an estimate of the price that end users of electricity will likely pay, one must add transmission, distribution, and any relevant administrative costs, plus taxes and any other charges that regulators might tack on to the consumer’s bill. Even with these add-ons, it is very likely that the typical electricity customer in Kentucky will see their electric bill fall in the short run. This is because currently there is significant excess capacity to generate electricity in the 20-state region.
Figure 2: Short-run Price for Electricity
Based on the short-run elasticity for all customers and an average power charge for electricity of 2.1 cents per kWh, the short-run increase in consumption can be estimated for the 20-state region. For such an estimate, we also require information on the average cost per kWh for transmission, distribution and taxes. We have two different sources for such information. From a 1997 report by the Energy Information Administration, we estimate an average charge of 2.0 cents per kWh charge for transmission, distribution and taxes.(4) This figure is based on a nationally representative sample and will not necessarily apply to any particular utility or customer class in Kentucky or other states in the region. Using another source, White (1996), we estimate the national average transmission, distribution, and tax costs per kWh for investor-owned utilities to be 1.1 cents.(5)
The level toward which price will tend to move in the future is known as the long-run equilibrium price. While short-run disruptions in supply or demand may cause the price to deviate temporarily from the long-run equilibrium level, the expected level of the price in a competitive industry is determined by the minimum long-run average total cost. In the long run, the demand for electricity can be expected to increase in response to lower prices. Even though consumers will demand more electricity in response to lower rates as more and more time passes, the expected long-run price of electricity will still be determined primarily by cost factors. That is because producers, given sufficient time to adjust, will build whatever generating capacity it takes to satisfy consumer demand.
Figure 3 illustrates the expected long-run outcome in the electric utility industry. The long-run demand curve incorporates greater responsiveness to changes in price than is possible in the short run.(6) The period of eight years is roughly the time it would take to reach the long-run demand response. The long-run supply curve indicates that the cost of generating electricity using the most up-to-date technology available is the determining influence on market price. The expected long-run price of electricity will be determined by the intersection of long-run supply and demand, as is shown in Figure 3.
Figure 3: Long-run Price for Electricity
The estimated long-run equilibrium price of electric power is three cents per kWh. Again, this is an estimated average price across all utilities and customer classes in the region over the course of an entire year. This long-run equilibrium is reached because at the same time that demand is increasing, the upward sloping portion of the industry supply curve will be decreasing (or shifting to the left). That will happen because each year some existing generating units become uneconomical to operate and are retired. How long it takes for long-run demand to intersect with long-run supply in the horizontal portion of the long-run supply curve thus depends on the decisions that utilities will make about renovating or retiring existing generating units in a deregulated environment.
Freshwater et al. (1998) conclude that "in Kentucky, where average rates are now well below the national average, the general direction of price changes will be up." They reach this conclusion by reasoning that a movement toward competition will increase prices in low-rate areas and decrease prices in high-rate areas. However, it is not the case that a movement toward competition will necessarily result in a convergence to some average of high- and low-rate areas. This reasoning ignores the fact that today’s regulated rates include large components for historical costs. In a deregulated electric power market, rates in both high- and low-cost areas are likely to fall because the marginal costs of producing power in equilibrium will be far below historical regulated average costs.
Freshwater et al. (1997, 1998) examine the effects of deregulation on Hancock, Pike, Pulaski and Trigg counties in Kentucky. They calculate the current residential price for electricity in each of these counties and compare it to the national price that they assert will exist under deregulation. They do not construct a model and estimate a national price. Instead, they use one of the national average prices published by the Energy Information Administration (EIA) (1997) of 6.5 cents per kWh. However the EIA does not predict that there will be a single nationwide price for electricity. This is the assertion of Freshwater et al. In fact, the EIA (1997) estimates that there will be significant variation in electricity prices across the country even after deregulation. In Kentucky’s region, the estimated price is 5.17 cents per kWh, which is far below the 6.5 cents per kWh asserted by Freshwater et al. (1997, 1998).
When one appropriately uses the EIA data, one obtains a far different picture than the one obtained by Freshwater et al. (1997, 1998) as is shown in Table 1. The first row of Table 1 shows the current residential electricity rates in each of the four counties considered by Freshwater et al. In the second row, we show the national price of electricity of 6.5 cents per kWh that they assert will occur under deregulation. Using this price, two of the four counties (Pike and Pulaski) experience price increases, one has price staying the same (Trigg), and one experiences a price decrease (Hancock). In the last column of Table 1 we summarize the results of the comparison. When the EIA estimate of the price in the region that contains Kentucky is used (5.17 cents), three of the four counties (Hancock, Pulaski, Trigg) experience price decreases, and only Pike experiences a price increase. However, even the result for Pike County is suspect for a couple of reasons. First, Pike is served by an investor-owned utility that will be competing for customers. It is unlikely that it will raise prices above current regulated levels and risk losing customers to competitors in a deregulated world. Second, the EIA regional price is likely to be an overestimate of the price that Kentuckians will face under deregulation because Kentucky’s region in the EIA report includes higher cost states such as Michigan, Indiana and Ohio. In fact, using the price predicted by our 20-state supply and demand model, all four of the selected counties would experience lower electricity rates after deregulation!
Table 1: Comparison of Residential Electric Rates in Four Kentucky Counties
Professors Freshwater, Goetz and Samson have ventured into the field of economics to predict what will happen to rural Kentuckians if the electric utility industry is deregulated. As we have carefully explained, the basic premise of their study is wrong. Even worse, by trumpeting their mistaken assertion that prices will rise for rural residential electric consumers, they may impede changes that hold much promise for rural and urban Kentuckians.
Let us start with some facts about electricity supply. Deregulation of the electric utility industry will affect the generation and the transmission of electric power, but it will not change in any fundamental way the distribution of electricity. Electricity will be distributed by a local utility that has a monopoly service area. Rural electric co-ops will continue to play a vital role in a deregulated market.
The distribution cost component of residential customers’ electric bills will continue to reflect the costs incurred by their local distribution monopoly. Per customer distribution costs will still be lower for densely populated areas with gentle terrain than for sparsely populated areas with rugged terrain. As they do now, some utilities will impose an average distribution charge for all of their customers. Other utilities may choose to eliminate cross subsidies and impose distribution charges that reflect the costs of serving each specific customer. That is a decision that the members of each rural electric co-op currently make and will continue to make for themselves.
The bottom line is that deregulation will neither increase nor decrease that portion of their monthly bill customers now pay to cover distribution costs. Rural customers typically pay more now for distribution costs than urban customers, and that will be true in the future. The promise that deregulation holds for rural consumers, however, lies not with distribution but with the generation of electricity. Competition in the market for generation is predicted to reduce the electric power component of consumers’ monthly bills, regardless of where they live.
It appears likely to us that rural residential customers have the most to gain from deregulation. We have listed in Table 2 the average electric rates paid by various customer groups in Kentucky. The left side of the table compares rates for residential and industrial customers of investor-owned, rural co-op, and municipal utilities throughout Kentucky. Industrial customers of rural electric co-ops pay the lowest and residential customers of co-ops pay the highest rates of any customer class in the state. It is hard to see how rural residential customers could be worse off if the world were to change.
Table 2: Average Electric Rates in Kentucky
The potential gain to low-income rural households becomes even clearer when one examines the rates charged by utilities in the four rural counties chosen by Freshwater et al. (1997, 1998). Pike County is served by Kentucky Power, a subsidiary of American Electric Power, which is an investor-owned utility. It has by far the lowest residential rates, 4.3 cents per kWh, of the four counties, in a region that contains some of the most rugged terrain in the state.
Hancock County is served by Green River Electric Corporation, a co-op, which purchases wholesale power from Big Rivers Electric Corporation. Almost 90 percent of Green River’s sales are to industrial customers, who pay a very competitive 2.9 cents per kWh. Green River’s residential customers pay more than double that rate, 6.8 cents per kWh. Green River’s major industrial customers, which include several aluminum smelters, were induced to locate in western Kentucky by the offer of cheap electric power. They have already realized the benefits of competition. Green River’s residential customers currently have only one choice when it comes to electricity. If Green River/Big Rivers had to compete with other utilities to generate the electric power that is distributed to Hancock County residents, these residential customers might receive better treatment.
It is also informative to look at prices rural electric distribution co-ops pay for wholesale power. These are shown in Table 3, along with comparable rates charged by investor-owned utilities. Green River pays on average 3.0 cents per kWh to Big Rivers for wholesale power. Pennyrile RECC pays TVA 4.5 cents per kWh on average. South Kentucky RECC pays East Kentucky Power Coop on average 3.4 cents per kWh. These rates can be contrasted with Kentucky Utilities’ wholesale rates for supplying power to various municipal utilities around the state. KU charges between 3.0 and 3.1 cents per kWh to cities such as Frankfort, Nicholasville, and Madisonville. KU’s rate is matched only by Big Rivers’ rate to Green River, a co-op that sells almost 90 percent of its power to large industrial customers. Pennyrile would be better off signing up with Kentucky Power as a residential customer and paying 4.3 cents per kWh than buying wholesale from TVA!
Table 3: Comparison of Wholesale Power Rates
Freshwater, Goetz and Samson (1998) pose five questions that policymakers need to understand but which they are unable to answer. Having thoroughly analyzed deregulation and the electric utility industry, we are in a position to help. We will close this article by providing answers to Freshwater, Goetz, and Samson’s questions.
To what extent will competition occur in rural areas and to what extent will it lead to a convergence of electricity rates among regions to a narrower range of values?
Competition among companies that generate electric power will lead them to seek customers in rural as well as urban locations. Competition among suppliers will cause the large disparities in the rates for the electric power portions of customers bills across regions to disappear.
How will rural areas served by cooperatives fare relative to areas served by investor-owned or municipal systems?
Rural areas served by cooperatives will benefit from deregulation, perhaps even more than urban areas, at least in Kentucky.
What is the magnitude of the average price increase that Kentucky will face as more power is moved from region to region? How high will charges for shifting power be and will transmission constraints limit the flow of power? How will policies on lost pass-through costs in other regions affect the demand for power in Kentucky?
Kentuckians will not see an increase in electricity prices as a result of deregulation, and that is the most important error in Freshwater, Goetz, and Samson’s analysis to recognize.
To what extent will rate differentials be preserved to allow manufacturers in Kentucky to retain access to low-cost power? If they narrow, how will Kentucky’s economic development strategy and prospects be affected?
In a deregulated market differences in rates charged to different classes of customers will reflect the different costs of serving these customers. Because rates are so high in other regions, Kentucky industrial customers do not stand to gain as much as out-of-state manufacturers from deregulation, which will obviously lessen Kentucky’s relative attractiveness as a business location.
To what extent will the changes in electricity regulation benefit the national economy and will improvements in the national economy offset the negative effects for Kentucky?
Deregulation of the electric utility industry will significantly benefit the national economy. We see very few negative effects for Kentucky, and if anything we expect rural residential customers to be some of the biggest winners from deregulation.
Energy Information Administration. Electricity Prices in a Competitive Environment: Marginal Cost Pricing of Generating Services and Financial Status of Electric Utilities, A Preliminary Analysis through 2015. Washington, DC: DOE/EIA-0614, August 1997.
Freshwater, David, Stephan Goetz, Scott Samson, Jeffrey Stone, Tulin Ozdemir Johansson and Monica Greer. The Consequences of Changing Electricity Regulations for Rural Communities in Kentucky, College of Agriculture, University of Kentucky, December 1997.
Freshwater, David, Stephan Goetz and Scott Samson, "Benefits of Deregulating Electricity May Bypass Rural Kentucky," Foresight 5:1 (1998): 1-5.
Scott, Frank A., Mark C. Berger and Eric C. Thompson. Competition and Choice in Electric Power: An Analysis of Regulatory Reform in the Kentucky Utility Industry. Center for Business and Economic Research, University of Kentucky, October 1997.
White, Matthew W., "Power Struggles: Explaining Deregulatory Reforms in Electricity Markets," paper presented at the Brookings Institution Microeconomics Conference, Washington, DC, July 1996.
The short-run demand curve is based on current consumption levels at the existing level of rates in the 20-state region. The curve itself is then derived using estimates of short-run price elasticity for residential, commercial, and industrial customers. Return to text.
The short-run supply curve is based upon existing production capacity in the 20-state region. It incorporates the following assumptions. Each fossil-fueled or nuclear generating plant is assumed to be economically viable at a level of average variable cost equal to the lowest level attained by that particular plant in either 1994, 1995, or 1996. Fossil-fueled plants are assumed to be capable of operating at 80 percent of nameplate capacity and nuclear plants are assumed to be capable of operating at 90 percent of nameplate capacity. The amounts of power generated by hydroelectric units and by nonutility generators in 1996 are assumed to remain constant. The original source for the data used in this exercise is the FERC Form #1. Return to text.
If we assume that each fossil-fueled plant will operate at 70 percent of nameplate capacity and each nuclear plant will operate at 80 percent of capacity, then the estimated competitive price rises to approximately 2.25 cents per kWh. Return to text.
This estimate of transmission, distribution, and tax costs is based on Energy Information Administration (1997). The transmission, distribution, and tax cost is assumed to be 1.5 cents per kWh for industrial customers and 2.3 cents per kWh for residential and commercial customers. Industrial customer costs are assumed to be lower because industrial customers in a deregulated environment will not bear the costs of distributing electricity to residential and commercial customers. Based on Table 2 of White (1996), distribution costs are assumed to be 62 percent of total transmission and distribution costs. To arrive at this percentage, one quarter of the "other operating expenses" in White’s Table 2 were assigned to transmission costs and one quarter were assigned to distribution costs. Return to text.
This figure is based on Table 2 of White (1996), with one quarter of "Other Operating Expenses" assigned to transmission costs and one quarter assigned to distribution costs. This implies that actual prices paid by residential and commercial consumers will be 1.3 cents per kWh above the power charge. Since many large industrial customers will be able to tap directly into the transmission grid and avoid distribution charges, it is assumed that 0.8 cents per kWh should be added to the power charge to get the price paid by industrial customers. The overall average of 1.1 cents per kWh thus is a weighted average, where the weighting factor is the industrial customer’s share of final regional sales which is reported in Table 7 of our earlier report. Return to text.
The long-run demand curve also reflects an estimated 2.2 percent annual growth in demand over time due to growing incomes, commercial sales, and industrial output. The annual demand growth rate of 2.2 percent is used because 2.2 percent was the average annual growth in electric utility electricity final sales in the United States from 1990 to 1996. Return to text.
Frank Scott and Mark Berger are professors of economics at the University of Kentucky. Their earlier research was sponsored by the four investor-owned utilities in the state. This article represents their own views and was not sponsored by any group.