Overview

This paper describes the price responsiveness incorporated into the Annual Energy Outlook 1999 (AEO99) versions of the National Energy Modeling System (NEMS) buildings sector models. The emphasis here on price responsiveness should not be taken to imply that price responsiveness is the main determinant of energy consumption either in NEMS or in general—it is not. Sectoral growth, the development and penetration of new technologies, and the penetration of existing or new end uses all have important effects on long-term energy consumption.

The Residential and Commercial Demand Modules (RDM and CDM) are separate models within NEMS. While the two models generally respond similarly, differences in accounting and equipment choice algorithms result in cases where one model may include effects or exhibit behavior different from the other. In such cases, differences are noted. The discussion of model features and algorithms provided here is intentionally brief, because detailed information is provided elsewhere.¹

The NEMS buildings sector models exhibit both short-run and long-run responses to changes in energy prices. Conventionally defined, short-run responses are the immediate behavioral effects of a change in energy prices on the intensity of utilization of a fixed stock of energy-consuming capital equipment. Long-run price responses occur through changes in the stock of energy-consuming capital equipment installed in buildings. As described below, for computational tractability, short-run elasticities are computed here as any change occurring in the first year of a price change. Long-run elasticities are computed as a persistent change in price after an interval of 20 years. Examples of short-run responses include adjusting thermostats on heating and cooling equipment, being more or less careful about leaving lights on or equipment running when not in use, or consuming more or less hot water.

The energy-using capital stocks convert energy from its raw potential into the desired end-use services. The NEMS buildings sector models are “stock turnover” models—they alter capital stocks by simulating equipment purchases for new construction, the replacement of worn-out equipment, and the retrofitting of still functioning but economically obsolete equipment.²

For buildings, capital service lives generally range from 12 years (e.g., air conditioners and heat pumps) to 30 years (e.g., boilers). Because of the persistence of the equipment stock, full responses to energy price changes occur over an extended interval. Long-run price responses occur in the models through potentially altered equipment purchases that may be projected to occur under different energy price regimes with all other factors and policies affecting energy consumption being equal. During periods of higher energy prices, examples of long-run responses include the purchase of more efficient lighting fixtures and bulbs, adding lighting timers and motion sensors, the purchase of higher efficiency space heating equipment, installing higher R-value insulation, and switching fuels when price increases vary by fuel (e.g., replacing an electric clothes dryer with a natural gas dryer or vice versa).³ During periods of lower energy prices, the purchasing tendencies simply reverse.

Own-Price Responses

Empirical studies of energy demand generally have found inelastic short-run responses to energy prices—that is, for a given percentage change in energy prices, a less than proportional percentage adjustment occurs in energy consumption.⁴ The short-run elasticity parameters in the buildings models are for each individual end use (heating, lighting, etc.). For both models, all end uses except refrigeration (which is assumed to be unresponsive in the short run) include a short-run price response. For all end uses with simulated equipment choices (including refrigeration), long-run adjustments to the efficiency of the equipment stock can also occur in response to price changes.⁵

Long-run responses to energy prices in the buildings models are determined endogenously through potentially altered equipment choices. Long-run responses occur through the interaction of installed equipment costs, equipment efficiencies, energy prices (“own” prices, and where fuel switching is a possibility, the prices of other energy sources), discount rates, and annual equipment utilization rates.⁶

In the RDM, the equipment cost versus equipment efficiency tradeoffs are modeled by a logistic functional form which provides a continuous adjustment of equipment market shares as prices change. The shares of equipment adjust smoothly from one year to the next in the model unless existing equipment types are removed (because of equipment standards) or new types are introduced (because of technological developments).

In the CDM, equipment shares are determined by comparing annualized capital costs plus operating costs in 1,782 discrete choice “segments.” The market segments for equipment are by Census Division (9), building type (11), choice set (there are three choice sets—unrestricted, restricted to the same fuel, and restricted to the same technology), and discount rate (6). Within each of the 1,782 segments, only one piece of equipment is selected. Selections are simulated on the basis of minimizing life-cycle costs among the available alternatives. The 18 combinations of choice sets and discount rates are intended to capture the varied behavior motivating building owners and occupants. The model segmentation also prevents the CDM from necessarily gravitating to a single equipment choice—a situation that would be highly unrealistic in most cases.

As described above, the long-run effects of equipment choice occur over an extended interval, and because of the multi-year equipment lives, the effects persist once purchases are made. Thus, for example, the effects of a temporary price increase, “wear off” over an extended interval. For the RDM, price-induced increases in building shell efficiency (e.g., insulation, caulking, thermally efficient windows) persist longer than equipment purchase decisions, because adjustments to the shell are not retired until the housing unit is retired.⁷ Thus, if prices decline in subsequent years, the effects of the installed shell measures will act as a damper on consumption levels, as illustrated below in a simulation that includes a temporary price increase. Equipment purchases other than shell adjustments have a persistence that is less than the life of the structure, and after a price shock they can wear off more rapidly than shell measures. For the equipment-related component of long-run price response there is a 10- to 20-year interval before full adjustment occurs (depending on the end use).

Another aspect of long-run price response is what has been referred to as the efficiency “rebound effect.”⁸ Efficiency rebound effects occur because the marginal cost of an end-use service is affected by the efficiency of purchased equipment. Higher efficiency equipment lowers the marginal cost of the service (the “price” of the service to the consumer), and the price response is increased consumption. Rebound effects influence consumption in the long run because of their link to equipment efficiency, which only changes gradually as equipment stocks turn over.

Cross-Price Effects

Another type of price effect occurs when one fuel’s consumption is affected by changes in another fuel’s price. These are referred to as cross-price effects, which can be either short-run or long-run. An example of a short-run cross-price effect would be altering the relative amount of food prepared using electricity relative to that prepared using gas. Although many homes have options to use both fuels (e.g., a home with both a gas oven and an electric microwave oven), short-run fuel switching rarely occurs in the buildings sector, and the buildings models do not include short-run cross-price effects.

Over the long run, the buildings models do exhibit some cross-price responsiveness, because certain equipment choice decisions include the consideration of competing equipment types using different fuels (e.g., electric versus gas water heaters). Thus, some equipment choices are based on more than just the price of a single fuel and result in measurable long-run cross-price elasticities. For example, in choosing residential space heating equipment for new construction, life-cycle costs of various types of equipment (gas furnace, electric resistance, electric heat pump, ground source heat pump) are compared in the model.

Elasticity Estimates and Simulations

To estimate responses to energy price changes, a series of alternate simulations were made with adjustments to the energy price paths from AEO99⁹ The adjustments begin in the year 2000, and continue through the end of the model run, 2020. Short-run price responses are defined here to be those that occur in the initial year of a price change.¹⁰ Long-run price responses are defined as the percentage change relative to a baseline after 20 years of a persistent change in energy prices.¹¹ This choice in measuring long-run responses is somewhat arbitrary, and for very long-lived equipment (such as space heaters) some additional responsiveness could potentially occur. Table 1 shows the results of a 10-percent increase in individual energy prices over the AEO99 levels for all years, one fuel at a time.

Table 1. NEMS AEO99 Buildings Sector Fuel Price Response Summary

The short-run own-price elasticities range from -0.23 to -0.47. Included in the estimated effects are the direct short-run effects plus one year’s worth of altered fuel choices and equipment purchases (fuel choice effects were not isolated from equipment purchase effects in the simulations). Long-run own-price effects are larger than short-run own-price effects in both models, as expected.¹² For the commercial sector, however, the long-run elasticity for electricity is only slightly higher than its short-run value.

The relatively small difference between the short-run and long-run price sensitivities for commercial electricity can be understood by isolating “major” end uses from the “minor” end uses. Major end uses—space heating and cooling, water heating, ventilation, cooking, refrigeration, and lighting—have endogenous, price-sensitive usage intensities. Minor end uses—office equipment and other miscellaneous uses¹³—are based on exogenous parameters. Growth in minor end uses is a function of non-price-responsive factors such as floorspace additions and the increasing penetration of office equipment.¹⁴ The calculated short-run and long-run elasticities for the major end uses are -0.24 and -0.31, respectively.

The spread between long-run and short-run elasticities is wider for residential than for commercial use of natural gas, in part because of the price responsiveness of building shells in the RDM, where opportunities for easy shell upgrades are available for many older housing units. Building shells are assumed not to be price responsive in the commercial sector. Commercial shell improvements generally are options for major building overhauls rather than incremental responses to price. The availability and cost of energy-efficient equipment are also factors, because some end-use efficiency opportunities are greater for the residential sector.¹⁵ For distillate, the CDM is more responsive than the RDM both in the short run and in the long run, because the CDM allows somewhat more price-responsive fuel switching.¹⁶

Long-run cross-price effects generally are negligible in both models except for the response of distillate consumption to a change in natural gas prices. As gas prices increase, there are some small shifts from gas to distillate. Because projected distillate consumption in 1999 is only about 10 percent of commercial gas consumption and 17 percent of residential gas consumption, any shift from gas to distillate will be magnified by a factor of nearly 6 for residential and just under 10 for commercial. For example, if 10 trillion British thermal units (Btu) of energy consumption shift from gas to distillate, gas consumption in the commercial sector declines by only 0.3 percent, but distillate consumption increases by 2.7 percent (the corresponding changes are 0.2 percent and 1.1 percent for the residential sector). This leveraging of any movement away from gas causes the relatively large cross-price elasticity for distillate in response to gas price changes. For an increase in distillate prices, distillate’s small share would cause a much smaller percentage effect on gas—thus the nearly negligible cross-price effects for natural gas in response to changes in distillate prices.

Price Shock Cases

To illustrate the responses of the NEMS buildings models under conditions other than simple, permanent price changes, the AEO99 reference case can be compared with two cases in which energy prices are doubled relative to the reference case beginning in 2000 for different lengths of time. In one case, prices are permanently doubled. In the other, prices return to the reference case path after a 5-year doubling shock.

Reviewing the results for the RDM (Figure 1), two effects are notable. First, under persistent doubled prices, there is an initial reduction in energy consumption of approximately 1.8 quadrillion Btu, which gradually widens to 2.8 quadrillion Btu by 2020.¹⁷ The widening gap is attributable to continued choices of higher efficiency equipment under the higher price regime. Its gradual nature is the result of different simulated equipment choices as equipment is retired and then replaced.

Figure 1. Response to Price-Doubling Sensitivity Cases:  Residential Sector Total Delivered Energy Consumption [source]

The second observation is that, for the case in which prices return to the reference path, there is still a slight gap that narrows over time but does not completely disappear. The gradual narrowing reflects the return to baseline equipment choices after the shock has ended. It is gradual for the same reason that the widening in the permanently price-doubled case is gradual—it occurs as equipment is retired and replaced. Over the 20-year course of the simulation, the gap between the reference case and the price shock still remains, because building shells responded to higher prices during the shock period. Any installed shell efficiency measures remain in place until the buildings themselves are retired from the stock. Similar results are shown for the CDM in Figure 2; however, the effects are not quite as persistent, because in the CDM there is no price-responsive retrofitting of building shells.

Figure 2. Response to Price-Doubling Sensitivity Cases:  Commercial Sector Total Delivered Energy Consumption [source]

Cross Price Effects From Equipment Choices

The second set of comparison cases illustrates long-run cross-price effects and uses distillate consumption as the example for both sectors. The comparisons include the reference case and three alternative cases—one with all prices increased by 10 percent, another with only the natural gas price increased by 10 percent, and a third with only the distillate price increased by 10 percent. As for the previous cases, all price increases begin in the year 2000. Comparing these three cases against the reference case illustrates the effects of relative prices on fuel choices in the two models.

Figure 3 illustrates the RDM results. When all prices increase, relative energy prices are the same as in the reference case, and fuel switching beyond that already in the reference case is minimized. When only the natural gas price increases, relative energy prices are altered, and equipment using other fuels becomes more attractive relative to natural gas equipment for end uses potentially served by different fuels. The slight increase in the demand for distillate fuel relative to the reference case is the result of the cross-price elasticity effects in the RDM. When only the distillate price increases, the result is a slightly greater suppression of distillate consumption than in the case in which all prices increase, because both the absolute price of distillate fuel and also its price relative to those of other fuels have increased, further suppressing its demand.

Figure 3. Illustration of Cross-Price Effects:  Residential Sector Distillate Fuel Consumption [source]

Figure 4 shows the results of a set of parallel cases for the CDM. When all fuel prices increase, demand for distillate is suppressed, as was the case for the RDM. When only the natural gas price increases, however, distillate fuel consumption is projected to be somewhat higher than in the reference case. This represents the switching of commercial gas-fueled services to distillate-fueled services. When only distillate prices increase, a small additional suppression of distillate consumption occurs, similar to that seen for the RDM.

Figure 4. Illustration of Cross-Price Effects:  Commercial Sector Distillate Fuel Consumption [source]

In 1993, the Energy Information Administration commissioned a survey of energy demand elasticities by Professor Carol Dahl,¹⁸ as background for the development of NEMS. The survey incorporated results from previous survey articles as well as from more recent studies (referred to as “new studies” below) that had been performed after the last major surveys. The previous survey articles included data primarily from the 1970s or earlier. A limited number of the new studies included data as recent as 1990, but many of the time-series-based new studies also included pre-energy-crisis intervals, and one used data from 1937 through 1977. Thus, the new studies do not necessarily represent studies of more recent consumer responses to prices.

In addition to short-run and long-run elasticities, Dahl also categorized the results of some models as “intermediate run” price elasticities—generally, from studies based on models that did not explicitly recognize a time path of adjustment to prices. Such models usually mix both short-run and long-run effects into a single estimate—hence the “intermediate run” nomenclature. A few of the studies reported results for the combined residential and commercial sectors, but they are not summarized here because the comparisons to the individual model results are less appropriate. Finally, because the Dahl study focused on own-price elasticities, comparisons here are limited to own-price elasticities.

Table 2 summarizes the information from the Dahl survey for the residential and commercial sectors. The table reports ranges derived from Dahl’s extensive tables of individual model results. Table 2 highlights the wide range of estimates that have been made for price responses. For example, residential short-run electricity demand elasticities range from +0.57 to -0.97. For intermediate- and long-run residential electricity demand, the range is from +0.77 to -2.5.

Table 2. Summary of Ranges of Residential and Commercial Elasticities from Dahl (1993)

In order to allow comparisons with the NEMS results presented above, the ranges from Table 2 have been aggregated by sector and fuel in Table 3. Furthermore, to make the comparisons more meaningful, the ranges have been narrowed by eliminating models reporting positive own-price elasticities. Also, because details on the scope of the new studies were readily available, only new studies with results that are nationally representative (i.e., not based on regional, State-level, or utility-level data) are included in the Table 3 ranges.

Table 3. Summary of Adjusted Overall Buildings Sector Fuel Own-Price Response from Dahl (1993)

National-level studies are the most comparable to the national estimates for NEMS shown in Table 1. Finally, because the intermediate run elasticities generally include effects beyond the initial short-run effects, they were combined with the long-run elasticities from Table 2. Comparing the results from Table 1 with those in Table 3, the NEMS short-run and long-run own-price elasticities fall within the reported overall ranges.

Summary

The behavior of end-use energy consumption under different fuel price paths has been described for the NEMS residential and commercial models. Both short-run and long-run adjustments to prices are included in the models. Responses categorized as short-run represent the immediate behavioral effects of energy price changes on the intensity of utilization of energy-consuming equipment. The long-run elasticities are a function of the cost and performance attributes of available equipment. As the projected equipment availability and cost and efficiency characterizations change, long-run responses to prices also change. The magnitudes of the estimated own-price elasticities for NEMS are consistent with the ranges from a 1993 survey of econometric studies.