In the October 2006 issue of PUF, I read two articles related to two RTOs—ERCOT and PJM. In their article, “Don’t Mess with Texas,” Hind Farag and Gary L. Hunt, based on Figure 6, conclude that there are some winners and losers with LMP. Could they point out any winner whose power costs have decreased after the implementation of LMP? I can bet they won’t find even one single (real) entity.
In his article, “The Most Effective Way,” Thomas L. Welch states that “a uniform clearing-price market encourages generators to offer electricity at their margins.” Setting aside the theory, could Mr. Welch also provide any actual instances when a generator bid into PJM at its marginal cost? Again, based on my actual experience with PJM, he won't be able to provide many examples, if any, of such instances.
Finally, you referred to the Energy-Only Market in MISO (interview with T. Graham Edwards, “The Nation’s Grid Chiefs: On The Future of Markets”). The protesters to such a market forget that the original basis for LMP was the Energy-Only Market, which was first implemented in PJM. I was one of the most active participants in the discussions in the stakeholder process, which led to the formation of PJM, and the Guru of LMP, Professor William Hogan, stated very clearly and loudly that there should be an Energy-Only Market based on LMP, and LMP would attract new generation. Later on, in every filing with FERC (e.g., FERC Docket No. PA03-12-000), prior to proposing RPM, PJM stated that the LMP-based market design was working well and was able to attract a lot of new generation capacity into PJM.
I am glad that MISO is sticking to the original basis of a supposedly competitive market.
Jay Kumar, President, Economic & Technical Consultants Inc.
“The Fallacy of High Prices,” by Howard J. Axelrod, David W. DeRamus, and Collin Cain (November 2006) purports to show that restructuring of wholesale power markets has resulted in significant benefits. However, the analysis it offers in support of this proposition—particularly in regard to the supposedly shrinking rate gap between restructured states and traditional, cost- of-service states—is not credible.
The article compares rate levels between 1998 and 2005, as reported by the Energy Information Administration (EIA), for a sample of Southeastern states and a sample of PJM states. The article finds that average rates in the five Southeastern states in its sample rose by 23.7 percent from 1998 to 2005, while rates in the four PJM states rose by only 7.8 percent. The article notes that the PJM states include rate caps for some customers, but cites New Jersey’s 9.6 percent rate increase as an example of a relatively small increase in a state that no longer has rate caps.
These comparisons reflect nothing more than a carefully selected choice of restructured vs. cost-of-service states. Other Southern states achieved much lower rate increases than the 23.7 percent average for the five states chosen by Axelrod: For example, EIA data shows1998-2005 average rate increases of 9 percent for Arkansas, 12.3 percent for Tennessee, and 11.5 percent for North Carolina. Similarly, numerous restructured states experienced much higher rate increases than the 7.8 percent average for the four states chosen by Axelrod: 1998-2005 average rate increases were 27 percent for Massachusetts, 24.9 percent for Rhode Island, 30.3 percent for New York, and 23.9 percent for the District of Columbia.
In addition, the authors performed an econometric analysis using 1980 through 2004 average rates for all states east of the Mississippi to estimate the effects of wholesale competition and state restructuring on the retail cost of electricity. The analysis found a positive benefit from the introduction of wholesale competition, but the few details provided raise serious questions as to whether the results are at all meaningful. The very time frame analyzed is of concern, as few would peg 1980 as the beginning of restructuring, and prices in restructured states in 2004 were to a great degree still subject to rate freezes.
Thus, it appears that the analysis contains several of the serious errors found in similar studies reviewed by John Kwoka in his recent paper, Restructuring the U.S. Electric Power Sector: A Review of Recent Studies. These include oversimplifying what is meant by restructuring and thus failing to isolate its effects; failing to take account of the fact that post-restructuring prices in many states are the result of rate reductions and freezes, stranded cost adjustments, and excess capacity; and failing to control for other factors that affect prices.
The Axelrod article also refers to a separate study, Electricity and Underlying Fuel Costs, by the Analysis Group as evidence of large cost increases in Louisiana and Oklahoma. However, the numbers that the Axelrod article cites as cost increases reported in the Analysis Group’s study—47 percent in Louisiana and 38 percent in Oklahoma—do not reflect cost increases; rather, they represent the percentage of power generated by natural gas in each state. This carelessness in reading the Analysis Group’s study makes it clear that the authors neither examined the Analysis Group’s study closely nor understood the results well enough to discover the study’s significant methodological problems.
Finally, the article references other analyses that were conducted by Howard Axelrod, but does not provide sufficient detail to allow the reader to assess whether the results are at all meaningful. Given the lack of precision and selective use of data shown in the article, it would be unwise to rely on these analyses without further clarification and detail.
Diane Moody, Director, Statistical Analysis, American Public Power Association
• Diane Moody’s critique focused on my co-authors’ results, for which they will be best equipped to respond. My analysis investigated a sample set of investor-owned electric utilities representing both structured and restructured states. Using FERC Form 1 data for years 1996-2005, I found that the standard deviation of total production costs in competitive markets, (i.e., a portfolio of self-owned generation, competitively bid bilateral contracts, and real-time transactions, as well as financial instruments to hedge price uncertainty), declined by 30 percent over the pre- versus post-restructuring period. I used 2000 as the transition year. While others may have been focusing on the volatility of ISO day-ahead and real-time prices, I found that a well-managed portfolio produced tangible benefits to utilities within restructured markets.
Furthermore, my comparison of total average costs found that while utilities in traditionally structured states, principally in the South, continued to be lower, there was a narrow convergence. Given the far greater dependence on natural gas in the restructured markets, I thought that was pretty significant.
As a final point, the states I used for my sample set included Massachusetts, Connecticut, New York, Delaware, and New Jersey, representing the NY/NE/PJM competitive wholesale markets, and North and South Carolina, Georgia, and Florida for traditional markets. I would hope that would satisfy Moody’s other concern.–Howard J. Axelrod
• Diane Moody’s critique is erroneous in several ways.
Regarding the comparison of PJM states to the five Southeastern states, these comparison groups were hardly cherry-picked. Our intention was to examine groups of states that often are contrasted as “high-cost” restructured states and “low-cost” unrestructured ones. The Southeastern states were chosen because they are contiguous and span the multiple utilities of two companies, Southern Co. and Entergy, that promote the supposed advantages of integrated utilities, and resist restructuring in their territories. PJM, meanwhile, long has been used as a model and benchmark for centralized electricity markets, and the constituent “PJM Classic” states were at the forefront of restructuring efforts and have been targets recently of anti-restructuring rhetoric. These are, thus, sensible and representative comparison groups for our purpose.
It is not clear to us why Moody, the director of statistical analysis at APPA, takes issue with the time period of 1980 through 2004 we used for our econometric analysis. To draw any statistically meaningful conclusions regarding the impact of restructuring over time, it is important to include a period in which none of the sample states had restructured in order to reliably estimate the parameters for other cost drivers. This approach clearly and unequivocally is supported by econometric theory and practice. In fact, our analysis adopts an approach applied by Professor Paul Joskow, which was reviewed in the APPA-sponsored paper referenced by Moody. John Kwoka’s assessment is that “Joskow’s study in many ways represents a good effort at evaluating the price impacts of electricity restructuring.” Our conclusions are consistent with those of Joskow.
We note that Moody does not take issue with our analysis of declining PJM price trends once an appropriate adjustment for fuel costs has been made, yet we believe this is a critical part of the restructuring story. Those who advocate a return to utility owned and operated generation in states that previously restructured seem to believe that restructuring and the benefits of competitive wholesale markets are separable. We do not. New investment in competitive generation has been driven substantially by restructuring and has contributed to robust wholesale power markets. Those benefits are not isolated in restructured states, but are enjoyed in unrestructured states as well, through the availability of wheeled competitive power and through the general promotion of independent generation. Since our analyses have not controlled for this “spill-over” effect, our results should be interpreted not as overstating, but rather understating the benefits of restructuring.–David DeRamus and Collin Cain