On the campaign trail, then-Senator Obama made ambitious statements regarding renewable energy investment and greenhouse-gas (GHG) reduction goals. For example, in October 2007, Senator Obama announced plans to, if elected president, reduce emissions to 1990 levels by 2020 and 80 percent below 1990 levels by 2050.1 Achieving such ambitious goals will require major changes on many fronts. In the electric power sector, an essential component of significant GHG reduction is energy efficiency.
In the arena of electricity efficiency, much attention has been given to building codes and weatherization, efficient lighting and appliance standards, and other measures that can be undertaken by businesses and households, often with incentives from the local electric utility. Rate design has received less attention. However, building on a survey the authors performed for BC Hydro in its 2008 residential rate-design application,2 a study by the authors suggests that rate design—in particular residential inclining block rates—can help achieve GHG-reduction goals. The same opportunity does not exist for time-varying rates.
Admittedly rough and based on simplifying assumptions, the study’s calculation suggests rate redesign could reduce GHG emissions by one to two percent. While seemingly negligible, this GHG reduction easily could be obtained at low cost and in short time. Thus, both regulators and electric utilities should consider residential inclining block rate design as part of their efforts in complying with the forthcoming GHG-reduction targets.
An inclining block rate has a per-kilowatt hour charge that increases with a consumer’s monthly kWh consumption. Most inclining block rates use a two-tier design, though three- or more tier designs do exist. The consumption and price levels set for each tier depend on the specific goals of the utility and the characteristics of its residential customer class. To collect the same revenue as an otherwise applicable flat rate, a revenue-neutral inclining block rate’s lowest tier charge must be below, and the highest tier charge above, the flat rate. For example, a hypothetical two-tier inclining block rate might provide an original flat rate of 10-cents per kWh, with a tier-1 rate 15-percent lower, and a tier-2 rate 25-percent higher (see Figure 1).
To see how such an inclining block-rate design can be revenue-neutral, while still providing a strong incentive to conserve, consider the simplified case of two hypothetical customers with monthly consumption of 667 kWh and 2,000 kWh, respectively. Under the flat rate, utility revenue is given by total consumption multiplied by the flat rate, or (667 kWh + 2,000 kWh) * $0.10/kWh = $267. Under the new rate, if the tier-1 quantity is set at 1,000 kWh, then 1,667 kWh will be billed at the tier-1 rate (all 667 kWh of the small customer’s consumption plus the first 1,000 kWh of the large customer’s consumption) and the remainder (1,000 kWh) will be billed at the tier-2 rate.3 Revenue under the new rate is equal to the quantity times the price in each tier: (1,667 kWh * $0.085/kWh) + (1,000 kWh * $0.125/kWh), or $267, identical to the revenue collected under the original flat rate.
Although the rate is revenue-neutral, the majority of the kWh sales (75 percent = 2,000 kWh ÷2,667 kWh) will see the tier-2 rate as the marginal rate, providing a strong conservation incentive. Higher prices lead to lower electricity demand. A 2004 meta-analysis of residential price elasticity studies reports 123 short-run estimates between -0.004 and -2.01, with an average of -0.35, and 125 long-run estimates between -0.04 and -2.25, with an average of -0.85.4
An inclining block rate is consistent with accepted criteria for utility ratemaking:5 It promotes efficient consumption. Since the per-kWh charge rises with consumption, it has the correct price signal in a rising marginal-cost environment. Plus it fairly apportions the costs of service. In a rising marginal-cost environment, it assigns a higher proportion of costs to large customers, who bear greater responsibility for the increasing costs.
Additionally, the inclining block rate maintains universal affordability. Low-income customers, who tend to consume less energy than other customers, enjoy the lower tier-1 rate. To be fair, an inclining block rate may result in less stable bills than the flat rate. But large bill spikes can be mitigated by an optional payment plan that aims to partly smooth large bill fluctuations.
The inclining block rate is non-discriminatory and easy to understand. The rate applies to all customers in the residential class, with bill differences reflecting consumption differences. Though more complicated than a flat rate, an inclining block rate remains easy to understand.
Finally, unlike time-varying or dynamic pricing rates, an inclining block rate can be implemented quickly and at very low cost using an electric utility’s existing billing and metering system.
One possible objection to residential inclining block rates in some jurisdictions is the need to maintain affordable electric space and water heating, particularly for low-income customers. This can be addressed, however, through the use of a design that offers a large tier-1 quantity for customers who have electric heating and no access to natural gas.
Inclining block rates already are used throughout the United States.6
Utility rates fall into four categories: inclining, flat, declining, and mixed. Flat rates provide a single price for all consumption, while declining block rates have per-kWh charges that decrease with consumption. Mixed rates vary by season (see Figures 2 and 3).
Summer inclining block rates are well established in the West Coast and Southwest states, where in most cases at least one of the two largest utilities has inclining block residential rates. They also are prevalent in the Southeast and, to a lesser extent in the Northeast and around the Great Lakes.
However, a significant portion of the country employs flat rates. This category includes Maine and Texas, where the two largest residential providers are competitive energy providers rather than regulated utilities. In these cases, rate structure is not readily apparent, but the small amount of published rate data available shows flat rates are used.
In the West, Southeast, and Great Lakes regions, inclining block rates also are widely used in non-summer seasons. Much of the country, however, employs declining block rates in non-summer seasons, particularly a central swath of the country and much of the Northeast. In seven states—Iowa, Indiana, Mississippi, North Dakota, Ohio, Pennsylvania, and West Virginia—at least one of the two largest utilities uses declining block rates year-round.
Utilities that emphasize demand-side management (DSM) programs might be expected also to use inclining block rates, which provide a strong incentive to conserve and shorten the payback period for energy-efficiency measures. However, this is not entirely the case for the sample of utilities studied in the authors’ review of rate designs, as shown by a comparison of rate structures and DSM expenditures reported on EIA Form 861.7
To be sure, utilities with higher DSM expenditures are more likely to employ inclining block-rate structures. The energy providers in the survey with relatively higher DSM expenditures were more than twice as likely as others to use year-round residential inclining block rates—28 percent vs. 12 percent, respectively.8 Nevertheless, the comparison also reveals room for improvement. Many utilities with higher DSM expenditures don’t yet employ residential inclining block rates (56 percent), or employ them during summer only (16 percent); several employ declining block rates for part or all of the year. These utilities miss an easy opportunity to boost the effectiveness of their DSM programs.
They also miss an opportunity to reduce their aggregate GHG emissions as estimated under the following assumptions.
According to EIA data, the jurisdictions in the study sample with flat or declining block rates serve approximately 350 TWh of residential load per year. This sales assumption excludes: A) sales by utilities in the sample that use inclining block rates in any portion of the year; and B) sales by utilities not in the sample. Including A) or B) magnifies the sales assumption and the savings opportunity.
These jurisdictions with flat or declining block rate structures adopt simple two-tier inclining block residential rates that are 15-percent lower than the original rate in the first tier, and 25-percent higher than the original rate in the second tier, in keeping with the example presented earlier.
Also in keeping with the earlier example, 75 percent of the 350 TWh sees the tier-2 rate as the marginal price, while the remainder sees the tier-1 rate as the marginal price.
Small users (1,000 kWh and below) facing the tier-1 rate as the marginal price have an average short-term price elasticity of -0.05; larger users facing the tier-2 rate as the marginal price have a moderately higher average short-term price elasticity of -0.1. These elasticities are conservatively low, given the meta-analysis of other studies, and are applied under the assumption that users respond to marginal price changes.9 The percentage in consumption by user group is estimated as the percentage change in price times the elasticity value.
GHG-emission rates among the affected utilities are equal to the U.S. average.
Using the above simplifying assumptions, the percentage change in total sales is 1.7 percent,10 or 5.9 TWh of energy savings. Assuming CO2 emissions intensity of 0.67 metric tons per MWh,11 this amounts to 3.96 million metric tons of CO2 savings, about one percent of what would be required to reduce the electric sector’s total CO2 emissions to the 1990 level.12 This number would be roughly doubled if the calculation were expanded to encompass utilities not in our review and those with inclining rates in other seasons.
While a one to two percent CO2 reduction might seem negligible, it’s significant when one considers the ease of implementing the rate redesign. Further, where marginal cost is high, upper-tier rates might be increased beyond the modest levels considered in this study, spurring even greater reductions. California’s large IOUs for example, have upper tiers that are multiples of lower tiers, nearly 30 cents/kWh in the case of PG&E.
Finally, the calculation does not account for long-term customer price response that entails energy-efficient purchase decisions, nor does it attempt to measure the enhanced value to existing DSM programs.
Time-varying pricing encompasses time-of-use (TOU) rates, real-time pricing (RTP), and critical-peak pricing (CPP).13 Peak-shaving benefits notwithstanding, there is little GHG reduction potential for alternative rate designs based on time-varying pricing.
To achieve meaningful GHG reduction, a rate redesign must induce a reduction in a customer’s overall kWh consumption. Time-varying rates, in contrast, mainly result in load shifting. To understand this point, consider the case of optional time-varying pricing. A customer likely joins a time-varying rate option, whether TOU, RTP or CPP, if he or she can achieve bill savings with relative ease. The bill savings can be obtained by shifting consumption from the high-price peak hours to low-price off-peak hours.14 While the participating customer may achieve the desired reduction in the per-kWh charge, there is little or no conservation incentive.
Making the time-varying rate designs mandatory doesn’t alter their inability to induce significant conservation. For example, a revenue-neutral two-period TOU rate design necessarily has a peak rate above, and an off-peak rate below, an existing flat rate. While the peak rate reduces peak kWh consumption, the off-peak rate increases off-peak kWh consumption. Thus, the total kWh effect of the TOU design is small. The same line of reasoning applies to an RTP that has hourly rates above and below the existing flat rate. It also applies to a CPP that has high rates during critical peak hours but low rates in non-critical-peak hours.
Inclining block rates offer a low-cost and timely opportunity to achieve electricity conservation and efficiency improvements, and resulting GHG-emissions reductions. Residential inclining block rates are easy to implement and to understand. Unlike time-varying and dynamic pricing rates, they don’t require new billing and metering infrastructure. Moreover, inclining block rates can spur residential customers to make long-term consumption decisions that incorporate investments in energy efficiency. Efforts to reduce national GHG emissions should include this easy-to-implement and low-cost measure.
1. Jeff Zeleny, “Obama Proposes Capping Greenhouse Gas Emissions and Making Polluters Pay,” The New York Times, Oct. 9, 2007.
2. Filed in February 2008, BC Hydro’s Residential Inclining Block Rate Application is available at: http://www.bchydro.com/etc/medialib/internet/documents/info/pdf/info_2008_residential_inclining_block_application.Par.0001.File.info_2008_residential_inclining_block_application.pdf.
3. Revenue-neutrality is calculated prior to any consideration of price-induced changes in consumption.
4. James A. Espey and Molly Espey, “Turning on the Lights: A Meta-Analysis of Residential Electricity Demand Elasticities,” Journal of Agricultural and Applied Economics, April 2004, Vol. 36, No.1, pp.65-81. See also: Ahmad Faruqui, "Inclining Toward Efficiency: Is Electricity Price-Elastic Enough for Rate Designs to Matter?" Public Utilities Fortnightly, August 2008, Vol. 146, No. 8, pp.22-27
5. Charles F. Phillips. The Regulation of Public Utilities, Public Utilities Reports, Arlington, Virginia, 1993, p.434.
6. DOE EIA-0348, 2007.
7. DOE EIA Form 861. EIA-861 does not provide DSM expenditures by customer class; we assume that utilities with high overall DSM expenditures include residential programs in their portfolio.
8. We defined “high” DSM expenditures as $0.75/MWh or greater, which results in a “high” label for energy providers with DSM expenditures in roughly the upper quartile of our sample. Where utilities did not report a value for DSM expenditures on EIA-861, we assume expenditures were, in fact, zero.
9. Applied microeconomics typically models customer responsiveness based on marginal price changes, see Jerry A. Hausman, "The Econometrics of Nonlinear Budget Sets," Econometrica, Vol.53, No.6, pp.1255-1282.
10. Percentage change in total sales = (share of sales with marginal rate at tier-1 rate * price elasticity for small users * percentage of tier-1 rate change) + (share of sales with marginal rate at tier-2 rate * price elasticity for large users * percentage of tier-2 rate change). Thus, (25 percent * -0.05 * -15 percent) + (75 percent * -0.10 * +25 percent) = 1.7 percent. Changing the tier-1 sales share assumption to 50 percent would result in a total sales change of -0.9 percent.
11. The U.S. average based on EIA 2006 sales and emissions data.
12. The total emissions reduction requirement is estimated based on EIA and EPA sources: http://www.eia.doe.gov/cneaf/electricity/epa/epat5p1.html; EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990–2006, Apr. 15, 2008, pp.2-4.
13. For a discussion of time-varying pricing options, see C.K. Woo, Eli Kollman, Ren Orans, Snuller Price and Brian Horii, “Now that California Has AMI, What Can the State Do with It?” Energy Policy, April, 2008, Vol. 36, pp.1366-74.
14. For empirical evidence on customer response to time-varying pricing, see: Chris King and Dan Delurey, "Efficiency and demand response: twins, siblings, or cousins?" Public Utilities Fortnightly, March 2005, 58-61; and DOE (2006) “Benefits of Demand Response in Electricity Markets and Recommendations for Achieving Them,” Department of Energy, Washington D.C. (Available at: http://www.oe.energy.gov/DocumentsandMedia/congress_1252d.pdf).