By: Joel Hicks, Ph.D. student and CESP GRA
The following provides a summary of a quantitative study the author conducted to better understand the impacts of state-level energy efficiency measures over time on both energy conservation as well as on economic performance. The two questions that were proffered included: 1) is there evidence to support the relationship between aggressive energy efficiency policies and efficacy in reducing carbon emissions? and 2) are there economic benefits to doing so? The result was not only yes, but significantly so, on both accounts.
Of course, aggregating thousands of energy efficiency programs at the state level would be optimal in answering this question. However, although much of this information exists within the Database of State Incentives for Renewables and Efficiency (DSIRE), the American Council for an Energy Efficient Economy (ACEEE) has, since 2007, successfully aggregated state level energy efficiency programs into a scoring criteria based on several distinct areas addressing energy efficiency: 1) Utility and public benefits programs and policies (2008-2013), 2) Transportation policies (2008-2013), 3) Building energy codes (2007-2013), 4) Combined Heat and Power (CHP) (2007-2013), 5) State government Initiatives (2009-2013), and 6) Appliance and equipment efficiency standards (2007-2013). For each category the allocation of points is designed to “reflect the relative magnitude of energy savings possible through the measures scored”. The study examined whether these ACEEE scores could be good proxy for answering the research questions.
Using a fixed effects panel regression of data (2007-2013) provided by the EIA [energy consumption, energy intensity (BTU per unit of GDP), and electricity rates], ACEEE [Total Score, Categorical Scores, and spending on energy efficiency programs], and the EPA [Cooling and Heating degree days] the study found that, after controlling for weather effects, that, on average, a state with a 1-point increase in ACEEE TOTAL SCORE relative to another state yields a 0.45 percent decrease (p<0.001) in EI, a 0.30 percent decrease in energy consumption (p<0.001), and a 0.16 increase in GDP (p<0.1). The reduction in overall consumption showed that, at a macroeconomic level, states with higher ACEEE scores do not exhibit a rebound effect, for conservation (a reduction in consumption) is observable at a statistically significant level. Because it is easy to relate how both energy consumption and GDP contribute to changes in EI [taking the logs of each yields the following interpretation in [log(EI) = log(Energy consumption) – log(GDP)], the data show that GDP and energy consumption each account for approximately one-third and two-thirds, respectively, in explaining how TOTAL SCORE drives EI downward. In other words, demand side policies have a significant net-friendly effect on carbon reductions [the 1/3 increase in GDP is offset in the Kaya equation by the GDP term itself (the product of terms (3) and (4))]. Thus, a 1-point increase in ACEEE TOTAL SCORE results in a net 0.30 percent reduction in carbon emissions relative to other states while still providing statistically significant positive growth benefits.
Why should states be focusing on energy efficiency as a, if not the, major pillar for addressing carbon reductions? There are two good reasons. First, energy policy in the U.S. has essentially been relegated to individual states. While the federal government does have the responsibility to regulate GHG’s under the Clean Air Act, the scale and scope of implementation tools are limited. In February 2016, SCOTUS granted a stay on a major federal tool for reducing GHG emissions from generating facilities, the Clean Power Plan, further mitigating federal influence on state carbon emission profiles. Although federal legislation theoretically holds potential in unifying state policies, congressional gridlock may prevent the timely implementation of policies to sufficiently address the problem. The failure of the Waxman-Markey Bill (HR 2454) to come to a vote in the U.S. Senate in 2009 signaled the most recent attempt to comprehensively address carbon emissions through federal legislation. Even under unified Democratic Party control of the White House and Congress this became politically unviable. With divided government possible for the foreseeable future, the prospects of comprehensive climate legislation appear slim. Consequently, the burden of meeting near-term, national carbon reduction targets may fall predominantly on the aggregated actions of individual states.
The second reason to focus on EE is because it is economically efficient to do so. This is true both as a cost-effective way of addressing carbon reductions (relative to production side measures), but also as a profit maximizing measure. The latter reason may help explain why states with higher ACEEE scores have relatively higher GDP growth, for instance. To better understand the impact on carbon reductions, the Kaya Equation is a useful way of showing how production and demand side policies influence carbon emissions; in fact, the Energy Information Agency (EIA) uses it as the primary determinant of U.S. state carbon emissions by tracking state production and consumption levels and assessing carbon emissions based on several factors, including fuel types, energy loads, and characteristics of generating facilities. Because each type of fuel has a unique emission profile (e.g. carbon emissions per BTU per fuel type) a rather accurate carbon footprint can be calculated.
Each of these terms has an expression associated with them related to the energy economy:
(1) = Carbon Intensity of Energy (CI) [kilograms of energy-related CO2 per million BTUs]
(2) = Energy Intensity of the economy (EI) [thousand BTU’s per dollar of GDP]
(1)x(2) = Carbon Intensity of the Economy [kilograms of energy-related CO2 per dollar of GDP]
(3) = GDP per Capita [dollars per person]
(4) = Population growth
Terms (1) and (2) account for all factors that can be managed within the realm of energy policy and represent, respectively, the supply and demand variables of the energy economy. For instance, improvements to electricity generation facilities’ carbon emissions affect CI, or the amount of carbon emitted for a given unit of energy produced. By comparison, improvements in building efficiencies and conservation (e.g. window insulation, turning lights off that are not in use, energy efficient water heaters), are examples of factors that affect EI, or the amount of energy consumed for a given unit (product or service) of GDP. What the Kaya equation conveniently reveals is that a given percentage change in any of these variables will yield a corresponding percentage change in carbon emissions. Reductions or increases in any of these variables contribute to changes in carbon emissions.
A recent report by Lawrence Berkeley National Laboratory (LBNL) found that it cost utility providers about 2.4 cents per kWh over the lifetime of an EE program. The magnitude of these costs is very consistent with the findings of this research, for this study found that a state with a TOTAL SCORE of 24, relative to a state with no EE program, would incur the same cost (2.4 cents per kWh). LBNL’s study levelized demand-side costs against production-side (all attributable to CI reductions) costs by equating costs of energy savings with costs of energy production, as shown in figure 1. The EE costs, which are rolled into the utility rates (as shown in the green bar), do not reflect any actual cost savings to the end user, which would be reflected only in macroeconomic indicators such as GDP, likely due to increased consumer spending. What figure 1 shows is that, even if EE programs did not return cost savings to a customer (which they certainly do), they would still be a more cost-effective way of managing carbon emissions than wind, natural gas, or solar.
As states scramble to develop carbon reduction programs within their respective energy sectors, the policy tools have tended to focus on the production side of the energy economy as opposed to demand side management (DSM). This has resulted in the search for less carbon-intensive energy sources and technologies throughout the various energy sectors (transportation, residential, industrial, electrical). For instance, there has been a large “fuel switching” trend within the electricity sector, most notably the substitution of coal for natural gas. Also, there has been a large shift toward more renewable sources of energy, primarily through the integration of wind and solar power into the grid.
However, despite state and federal government focus on supply side policies figure 2 shows that, since 1990, reductions in EI have yielded, by far, the greatest reductions in carbon output in the U.S. For instance, in Virginia, between 2005-2013, the average CI reduction per year was 1.31 percent versus 2.15 percent for EI. In absolute terms they amounted to a reduction of 10.45 percent and 17.16 percent in CI and EI output, respectively. Yet, this reality has yet to translate into a proportionately wider focus on EE and DSM policies relative to production side policies.
A closer look at Virginia
Interestingly, Virginia seems to be a bit of an outlier to the outcome of the research. Why? Virginia has ranked no better than 32nd (out of 51) between 2007 and 2013 on the ACEEE scorecard. Yet, the Commonwealth’s total energy consumption (-8.8%) is below the U.S. average of (-3.8%) [figure 3]. However, her GDP growth is more consistent with the evidence [figure 4] (4.7% versus 5.6%).
This may suggest that consumers in Virginia could benefit from incentives to invest in energy efficiency improvements. Regulations and behavioral changes seem to be making improvements in consumption patterns, but may not be targeted at economic efficiency. There are many barriers to overcome, including principal agent challenges that reflect landlord/tenant relationships as well as informational asymmetries. Much more research is encouraged to better understand Virginia’s unique situation.
EI remains the primary means by which carbon emissions have been reduced (or limited) in the U.S. Few studies have looked at the quantitative costs incurred by state policies and incentives aimed at promoting energy efficiency. This study demonstrates that the aggregation of EE programs, using the ACEEE energy efficiency scorecard, can be a strong predictor of EI and consumption reductions as well as growth increases. This information, in turn, can serve as a policy tool for states to manage their carbon emissions relative to production-side options. The relatively low cost of implementing EE programs in the electricity sector confirms previous research, comparing the levelized costs of CI and EI investments in reducing carbon emissions. The evidence here suggests it is still cheaper to meet new energy needs through reducing demand than by increasing production. Other economic benefits appear to derive from these EE policies as well.
 “The 2014 State Energy Efficiency Scorecard | ACEEE.”
 One of the criticisms of energy efficiency was put forward as early as 1865 by economist, William Jevons, who proposed that energy efficiency is counterproductive due to the “rebound effect” or “backfire”.
 Adler, “Supreme Court Puts the Brakes on the EPA’s Clean Power Plan.”
 Raupach et al., “Global and Regional Drivers of Accelerating CO2 Emissions.”
 “U.S. Energy Information Administration – EIA – Independent Statistics and Analysis.”
 “How Much Does Energy Efficiency Cost? | ACEEE.”
 Wilson et al., “Marginalization of End-Use Technologies in Energy Innovation for Climate Protection.”