Jay Gregg1, Olexandr Balyk1, Ola Solér1, Simone La Greca1, Cristian Hernán Cabrera Pérez1

Compatibility of the SE4ALL Energy Efficiency Objective with Renewable Energy, Energy Access, and Climate Mitigation Targets

Jay Gregg1, Olexandr Balyk1, Ola Solér1, Simone La Greca1, Cristian Hernán Cabrera Pérez1, Tom Kober2

1Systems Analysis, Technical University of Denmark

2Energy Research Centre of the Netherlands

Abstract

The objectives of the Sustainable Energy for All (SE4ALL), a United Nations (UN) global initiative, are to achieve, by 2030: 1) universal access to modern energy services; 2) a doubling of the global rate of improvement in energy efficiency; and 3) a doubling of the share of renewable energy in the global energy mix (United Nations, 2011; SE4ALL, 2013a). The purpose of this study is to determine to what extent the energy efficiency objective supports the other two objectives, and to what extent the SE4ALL objectives support the climate target of limiting the global mean temperature increase to 2° C over pre-industrial times. To accomplish this, pathways are constructed for each objective, which then form the basis for a scenario analysis using the Energy Technology System Analysis Program TIMES Integrated Assessment Model (ETSAP-TIAM). We find that, in general, the energy efficiency objective is reinforced by the renewable energy objective, but not by the universal access objective. Achieving the energy efficiency objective is made cheaper (in terms of the net present value of investment costs) when the renewable energy target is also achieved. However, achieving both the renewable energy and energy efficiency targets require more investment than achieving the renewable energy objective alone. Furthermore, we find that the universal access objective requires much more investment in the residential sectors of developing regions of the world, and makes the meeting the other two objectives more expensive. Meeting any of the objectives also requires increased investment in the transportation sector. While achieving the SE4ALL objectives does not limit warming to 2° C on its own, it makes a substantial contribution toward that goal, particularly if the renewable energy and energy efficiency objectives are met.

Introduction

Background

Energy savings through energy efficiency are viewed as a one of the primary avenues to address anthropogenic climate change for many years. Investing in energy efficiency has long been viewed as “win-win”, as the same level of end-use service is delivered at reduced cost and reduced emissions (Jaffe et al., 1999). Three of the fifteen “stabilization wedges”, actions that can reduce emissions by 1 GtC/year (3.34 GtCO2/year)[1] by 2054[2] for solving climate change identified by Pacala, et al. (2004) involve energy efficiency improvements. It is estimated that energy efficiency improvements in vehicles, buildings, and baseload power generation could contribute approximately a third of total reductions in emissions necessary to address climate change (Pacala et al., 2004). Additionally, technologies for achieving these reductions in emissions are largely available (McNeil, et al. 2012), thus presenting a particularly attractive mitigation option for the short- and medium-term.

While engineering studies estimate that carbon emissions could be inexpensively reduced by 20-25% by globally switching to more efficient end-use technologies (e.g. lighting, appliances, insulation, etc.), economic studies emphasize that consumers are primarily motivated to switch products when there is a price incentive to do so (though other qualitative features can play a role); thus technological developments into more efficient products are motivated by price differentials (Markandya, 2001). Moreover, the existence of a huge “energy efficiency gap”, i.e. the gap between potential energy efficiency gains and realized ones, has recently been questioned by some economists (Allcott & Greenstone, 2012). Likewise, though the Working Group III of the Intergovernmental Panel on Climate Change (IPCC WG3) indicated that in order to remain below 2°C warming (equivalent to a stabilization of 430-530 ppm CO2) by the end of the century, annual investments in energy efficiency for transport, buildings, and power would need to be increased by USD 336 billion compared to reference scenarios, this conclusion had only limited evidence, medium agreement and high uncertainty (USD 1 – 641 billion) (Gupta, et al., 2014).

Modeling of Energy Efficiency

Energy efficiency was one of the main topics addressed by the 27th Energy Modeling Forum (EMF) in Stanford, California, 2014, providing a detailed comparison of energy-economic models and integrated assessment models. A suite of 18 different models ran multiple scenarios for energy efficiency, and included both simulated technological improvements as well as simulated structural changes in the economy (Sugiyama, et al., 2014). Some of the models accomplished this simulation by modifying end-use efficiencies while others adjusted the “autonomous energy efficiency improvement” (AEEI)[3] rates to meet the specified energy demand reduction targets defined in each scenario. The comparative study showed that energy efficiency improvements occurred quicker under a climate policy, and this improvement rate was enhanced when technology was constrained (with fewer technological options for reducing emissions, efficiency improvements become more essential to achieving a climate target) (Sugiyama, et al., 2014). The study showed that the second objective of the SE4ALL initiative was feasible, that without a climate policy, Energy Intensity Improvement Rates (EIIR) were around 2% per annum (Sugiyama, et al., 2014). This result is corroborated by Kriegler, et al. (2014), who found that doubling the rate of improvement in energy intensity of GDP significantly reduced global mitigation costs. Kriegler, et al. (2014) has noted, however, that most models have a very crude representation of demand side investments and costs, and that the models may therefore be underestimating the mitigation costs. Likewise, Sugiyama, et al. (2014) found large variances across models in the efficiency improvement rates and potentials, particularly at the regional and sectoral levels. General equilibrium models tended to reduce service demands, while in contrast, partial equilibrium models preferred technological substitution to meet climate targets (Sugiyama, et al., 2014).

The Times Integrated Assessment Model at the Energy Research Centre of the Netherlands (TIAM-ECN) was recently used to examine the question of energy efficiency at the specific technology level (Kober, 2014). The study focused on the G20 countries, and primarily on the transportation sector, using energy efficiency parameters from the International Energy Agency (IEA) (resulting in 46% and 42% reduction in energy use for cars and trucks, respectively). Kober (2014) used four different climate policy scenarios: BAU and 3 carbon tax prices of $40, $70, and $100 per tonne of CO2e; and found that energy efficiency measures are more effective with an increased carbon price ($40/t, $70/t and $100/t). Emissions are reduced by 2-3 GtCO2e by 2030, representing 15-25% of greenhouse gas (GHG) reductions in relation to the BAU baseline.

Rogelj, et al. (2013) conducted an analysis of the SE4ALL objectives using the MESSAGE IAM framework, which included climate impacts addressed by using the climate model (MAGICC). MAGICC was used to limit the global temperature increase to less than 2° C by the end of the century (Rogelj, et al. 2013). They found that the SE4ALL objectives were compatible with climate goals- that sustainable energy and providing universal access to energy were important steps to mitigating climate change and remaining below 2° C warming (Rogelj, et al. 2013). This study emphasizes the importance of making a sensitivity analysis in both the renewable energy and the energy efficiency scenarios because the GDP projections (and accordingly energy demand) will change in the future, therewith affecting the quality of our results and robustness of our model. Second, the provision of universal energy access has a limited impact on the achievement of the SE4ALL objectives and on climate protection. Furthermore, this is unlikely to be achieved before the 2060s (Rogelj, et al. 2013). The universal access goal also would in turn reduce the global renewables share of final energy by about 2%- this may be due to the replacement of conventional biomass (use for cooking and heating) to electricity or LPG (Rogelj, et al. 2013). On the other hand, this demand may be met by distributed renewable energy which would increase the share of renewables. Third, the energy intensity indicator cannot be used as the sole yardstick to measure climate action since climate action can only be measured and assessed in terms of the actual effectiveness of policies in limiting and reducing the absolute amount of GHG emissions (Rogelj, et al. 2013). However, the scenario analyses by Rogelj, et al (2013) do not include policy instruments such as feed-in tariffs or carbon tax that would trigger the implementation of specific measures. Other current projects, including the Bottom-Up Energy Analysis (BUENAS) project model appliance energy demand and efficiency improvements to determine the effect on greenhouse gas emission reduction (IESG, 2015).

Historic Trends in Energy Efficiency

The Global Tracking Framework (SE4ALL, 2015) highlights some success in attaining the SE4ALL objectives: over the last 20 years, over 1 billion people gained access to electricity, global renewable energy share has increased from 16% to 18%, and energy intensity has dropped. Nevertheless, faster progress is necessary if the objectives are to be achieved, summarized in Table 1.

Table 1. Progress in achieving the SE4ALL objectives (SE4ALL, 2015).

National level energy intensity (of GDP PPP) and primary energy consumption data were attained from Enerdata (2015) for the years 1990-2013, representing 88% of global energy consumption. Countries not represented in the Enerdata (2015) database were estimated by calculating the difference between the regional totals and the reported national statistics that comprise the respective regions. The energy intensity statistics were divided by the energy consumption statistics and inverted to produce internally consistent data for GDP PPP. In Figure 1, the historic regional trends are depicted. China and the Former Soviet Union have the highest energy intensity (highest levels of energy consumption per unit of economic output) whereas Europe has among the lowest energy intensity. There is not a discernable relationship between the level of economic development and level of energy intensity. The global rate of energy intensity has decreased rather steadily over the 23 years in the data.

Figure 1. Historical Energy Intensity, by region based on 2005 GDP PPP.

Moreover, the rate of change in energy intensity varies substantially year to year. In Figure 2, different rates of change for global energy intensity are plotted together. The data indicate that the rate of improvement can vary wildly from year to year (which depends on both the economy and the quality of the data). This is the case both for the average rate of change and the compound annual growth rates (CAGR) calculated from the endpoints. In general, there is an upward trend in these lines, suggesting that the energy intensity improvement rate (EIIR) is diminishing as time goes on. These data align well with the Global Tracking Framework (GTF) estimates, plotted as diamonds. The estimates from the GTF match well with the analysis from Enerdata (2015) data. Using CAGR, the decadal global change in energy intensity is between -0.9 and -1.6%. The long term point average is about -1.3%, as seen also in Table 1.

Figure 2. Rate of change in global energy intensity of GDP PPP, using 2005 as a basis: 5-, 10- and 20- year average smoothing has been applied, as well as calculation of the compound annual growth rate (CAGR) for the previous 5, 10, and 20 years. For reference, the CAGR estimates reported in the Global Tracking Framework (SE4ALL, 2015) are also included here.

Barriers to Energy Efficiency

Engineering estimates of energy efficiency potentials are often not achieved in the real world due to these barriers in adoption of energy efficiency (Allcott & Greenstone, 2012; DOE, 2015). Achieving energy savings through energy efficiency measures is more than just a technical problem or a question of cost. Schleich, et al. (2008) surveyed 2848 German companies and public institutions, and developed a model for efficiency improvements for each subsector of organization types. They found that not only do barriers to energy efficiency vary consistently across sub-sectors, but that there is no clear pattern of combination of barriers (Schleich, et al., 2008). At the firm level, Decanio (1998) also examined barriers to energy-saving investments, using a multiple regression analysis of economic and organizational favors on firm profitability resulting from lighting upgrades. Dacanio (1998) found that economic factors alone were not enough to explain firm behavior, and in fact, when it came to energy-savings investments, firms sometimes strayed from profit maximization, and thus did not act economically. Thus, potential energy-saving investments are often not realized due to internal impediments of the organization (Decanio, 1998). Nevertheless, the economic potential for cost savings is still the most important motivation in investing in energy efficiency (De Groot, et al., 2001). While most firms accept government regulation, they prefer it at the international level (e.g., the EU) and such that the policies maximize their freedom and flexibility for meeting the regulation requirements (De Groot, et al., 2001).

However, Schliech, et al. (2008) noted that “organisations with rented buildings and office space also tend to know less about energy consumption patterns” with regular consistency, indicating a lack of financial incentive to improve building efficiency if the tenant is responsible for the energy costs. Additionally, more energy intensive sub-sectors (which had higher economic incentive to improve energy performances) had significantly less barriers to efficiency improvements; whereas sub-sectors with public or quasi-public ownership exhibited higher impact from barriers to efficiency improvements (Schleich, et al., 2008). In the commercial, private, and public sectors, lack of information about energy efficiency measures was a significant barrier (due to weak technical expertise) (Schleich, et al., 2008). This corroborates a finding by De Groot, et al. (2001), from a survey of 135 Dutch firms, where 30% of firms were not aware of existing new technologies and a further 20% had only limited knowledge on technologies that were in use by other firms. They furthermore noted that competitive firms tended to delay the adoption of new technology on account of uncertainty, particularly in regard to future price reductions (De Groot, et al. 2001). Thus, information campaigns on energy savings technologies should concentrate on these sectors, combined with an effective mix of policies to reduce investors' transaction costs for energy savings measures (De Groot, et al., 2001; Schleich, et al., 2008). Employing Energy Services Companies (ESCOs) could potentially be an effective way to overcome some barriers (e.g., risk, lack of capital, lack of time, and lack of staff for energy monitoring and assessment); however, ESCOs are generally reluctant to do business in the commercial sector because they have lower financial risk when operating in the public sector, and because ESCOs prefer large projects (most of which are outside the commercial sector) where the savings can be found in minimizing transaction costs (Schleich, et al., 2008).