Four years ago we documented that retail rates in both restructured and non-restructured states increased by approximately 31 percent since restructuring started in 1997.1 While 2006 rates in restructured states were significantly higher than rates in non-restructured states, this was already the case in the mid-1990s prior to restructuring.
This article updates these rate comparisons through 2010, a four-year time span that includes significant initial increases in power prices, substantial economy-wide adjustments, and striking changes in electricity markets triggered by a sharp decline in natural gas prices and continued increases in the cost of coal. In addition to updating the analyses, we explore three questions: 1) how much have rates in restructured and non-restructured states adjusted to the change in economy-wide and energy market conditions?; 2) do higher electricity rates in restructured states also correspond to higher residential electricity bills?; and 3) what portion of household income is spent on these electricity bills?
We find that electricity rates in fully restructured states have increased by slightly more than rates in non-restructured states through 2008, but this increase has reversed itself in the last two years with rate reductions in restructured states and continued increases in non-restructured states. As a result, the percentage increase in restructured states’ retail rates since restructuring in the mid 1990s has been slightly less than the rate increases in non-restructured states or states that suspended restructuring.
The level of electricity rates (cents per kilowatt hour) in fully restructured states continues to exceed rate levels in non-restructured states by 37 percent—slightly less than the 42 percent premium that existed prior to restructuring. However, despite these higher rates, average residential electricity bills (dollars per month) in fully restructured states are only 5 percent above average bills in non-restructured states due to lower electricity usage. Considering that median household income in restructured states exceeds the income in non-restructured states, this means that electricity bills account for a lower share of household income for the average residential customers in restructured states.
This unexpected combination of higher average rates but similar residential bills that are more affordable is even more pronounced in high-rate states like California, where the combination of energy conservation, smaller residences, milder climate, and higher household income means that a significantly lower proportion of total household income is spent on electricity bills than in either restructured or non-restructured states on average.
We have updated the 2007 study to reflect the latest state-level data on electricity rates.2 Since 2006, the last year covered in the previous study, restructuring related rate freezes have expired in states such as Illinois and Pennsylvania, while other states, such as Virginia and Michigan, have suspended restructuring activities and retail access. These new data allow us to compare rates in non-restructured states3 with rates in fully restructured (full retail access) states4 and states with suspended restructuring activities.5 As a case study to illustrate the relationship between average rates, average bills, and electricity spending as a percent of total household income, we also report results separately for California, a large state with suspended restructuring that would dwarf the other states in the suspended restructuring subcategory. Finally, we report trends in average residential bills per customer and the percentage of median household income spent on electricity bills.
Figure 1 shows average electricity rates (across all customer classes) in non-restructured states, full retail access states, California and other states with suspended restructuring. The figure shows that rate levels in restructured states continue to exceed those in non-restructured rates: 2010 rates in full retail access states were 37 percent above rates in non-restructured states, compared to 42 percent in 1997; California rate levels were 62 percent above those of the non-restructured states; and rates in other suspended restructuring states were 7 percent higher. As Figure 1 also shows, these rate differentials already existed in the 1990s, prior to restructuring.
Figure 2 shows the increases in average retail rates within each of these groups of states since restructuring was implemented in 1997. This update through 2010 shows that average retail rates in full retail access states are now 42 percent above their 1997 level, compared to increases of 45 percent in California, 46 percent in other suspended restructuring states, and 48 percent in non-restructured states. Figure 2 also shows that rates in fully restructured states have decreased 5 percent since 2008, while rates in non-restructured and suspended restructuring states, including California, have continued to increase. These comparisons show that retail rates in fully restructured states more closely follow prices in wholesale power markets, which sharply increased between 2004 and 2008, but have since declined.
These types of rate comparisons can be subject to misinterpretation. First, while it is correct that rates in restructured states are significantly higher than in non-restructured states, this difference already existed prior to restructuring. While the study results show that rate increases have been similar in restructured and non-restructured states since 1997, the gap in rate levels hasn’t been reduced. As already noted in the earlier paper, it’s clear that restructuring has failed to produce the massive hoped-for benefits, the basis on which restructuring was sold politically in the 1990s. However, it’s also clear that rates in restructured states haven’t increased any faster than rates in non-restructured states.
Second, simple rate comparisons don’t control for the many reasons why rate differentials exist with or without restructuring. These reasons include differences in fuel mix and fuel costs, differences in the costs of labor and other local inputs, differences in load shapes and weather, and differences in demand response and energy efficiency investments. This analysis only documents rate trends and how consumers in restructured states have fared relative to those in non-restructured states—without attempting to estimate the perhaps larger rate increases that restructured states might have experienced absent the restructuring effort.
So what do these differences in rate levels and rate increases between restructured and non-restructured states mean for the average residential consumer? As Figures 1 and 2 show for the average across all customer classes, the price of electricity in restructured states remains above prices in non-restructured states. The same pattern holds true for average residential rates: those rate levels in fully restructured states are 36 percent above the residential rates in non-restructured states.
Figure 3 sheds light on this “rates versus bill” question, showing average monthly electricity bills for residential customers (i.e., dollars per customer per month).6 These differences in bills are much smaller than the differences in rates. For example, the 2010 average monthly residential bills in full retail access states are only 5 percent higher than average residential bills in non-restructured states. Figure 3 also shows that monthly residential bills in suspended restructuring states are 7 percent lower than in non-restructured states. This means that the impact of higher rates in fully restructured states is mostly offset by lower consumption, yielding fairly similar monthly residential bills.
The effect of lower usage is particularly striking for California, where despite the fact that residential retail rates are 50 percent above those in non-restructured states, California residential bills are 25 percent lower. This surprising result likely is driven by a variety of factors: smaller average size of residences in many urban areas, milder climates in the coastal metropolitan areas, stricter building codes, increased spending on energy efficient appliances, substantial utility-sponsored energy efficiency efforts (which tend to increase rates but reduce consumption), and the basic effect of demand elasticity, which reduces consumption in response to higher prices.
How much of household income is spent on electricity bills and how does that compare across restructured and non-restructured states? Figure 4 addresses this question by expressing residential bills as a percentage of median household income as reported by the Bureau of Census.7 This shows that electricity bills amount to less than 3 percent of household income. This is consistent with national compilations of expenditures, showing that spending on electricity accounted for only 2 percent of after-tax personal income, as of 2009.8 Consumers spent more of their after-tax income on other classes of expenditures such as: housing (28 percent), transportation (13 percent), food (10 percent), and health care (5 percent).
As Figure 4 shows, despite the high average rates, electricity bills account for a smaller share of household income in full retail access states than in non-restructured states. While electricity bills account for almost 3 percent of household income in non-restructured states, they were only 2.7 percent of household income in full retail access states and 2.5 percent in suspended restructuring states. In California, only 1.8 percent of average household income is spent on electricity bills. The data also show that while spending on electricity has increased over the last decade, electricity’s share of household income has been fairly stable over time, with current shares in restructured and non-restructured states roughly equivalent to the shares during the 1990s.
This update of retail rate trends in restructured and non-restructured states confirms the core conclusion from our prior article: rate increases have been very similar since restructuring was implemented in the 1990s. While it is correct that average retail rates in restructured states are 37 percent higher than in non-restructured states, that difference already existed prior to restructuring. New data for 2007 through 2010 show that retail rates in restructured states have increased and decreased more directly with changes in wholesale power markets. Over the 1997 through 2010 period, since restructuring was implemented, rates in fully restructured retail access states have increased 42 percent, while rates in non-restructured states have increased 48 percent.
Surprisingly, higher retail rates don’t necessarily translate into higher average residential bills nor do the higher rates necessarily mean that electricity is less affordable. While rates are significantly higher, average residential bills in fully restructured states are only slightly above residential bills in non-restructured states. Given higher average household incomes, electricity bills on average are a smaller portion of household incomes in restructured states than in non-restructured states. This unexpected result is particularly striking in California, where much lower average electricity use and higher household incomes make electricity bills more affordable on average than in either restructured or non-restructured states.
1. Pfeifenberger, Basheda and Schumacher, “Restructuring Revisited: What we can learn from retail-rate increases in restructured and non-restructured states,” Public Utilities Fortnightly, June 2007.
2. The update is based on data reported by EIA in Form 826 through 2010, including revisions made by DOE to previously issued data back to 1990.
See http://www.eia.doe.gov/cneaf/electricity/page/sales_revenue.xls, last accessed on April 17, 2011.
3. Non-restructured states include: AK, AL, AR, CO, FL, GA, HI, IA, ID, IN, KS, KY, LA, MN, MO, MS, NC, ND, NE, NM, OK, SC, SD, TN, UT, VT, WA, WI, WV, and WY.
4. Full retail access states include: CT, DC, DE, IL, MA, MD, ME, NH, NJ, NY, PA, RI, and TX.
5. Suspended restructuring states include: AZ, MI, MT, NV, OH, OR, and VA (excluding CA, a large state with suspended restructuring, for which we report data separately as a case study).
6. These bills are calculated by dividing 1/12 of total annual residential electricity revenues by annual counts of residential customers. For customer counts, see: http://www.eia.doe.gov/cneaf/electricity/epa/customers_state.xls, last accessed on April 17, 2011. 2010 customer counts are not yet available from EIA 861, so we assume no change in the number of residential customers between 2009 and 2010.
7. Averages for the three categories of states reflect weighted averages of these percentages, where the weights are each individual state’s number of residential customers. Because 2010 state level median household income data are not yet available from the Bureau of Census, we used the Bureau of Economic Analysis’s estimated 2009-2010 percentage change in state level personal income to escalate the 2009 state level median household income data to 2010. For the core Bureau of Census data, see: http://www.census.gov/hhes/www/income/data/historical/household, last accessed on April 17, 2011.
For the Bureau of Economic Analysis escalators, see: http://www.bea.gov/newsreleases/regional/spi/2011/xls/spi0311.xls, last accessed on April 17, 2011.
8. See: http://www.bls.gov/cex/2009/standard/multiyr., last accessed on April 17, 2011.