COMPATIBILITY OF THE SE4ALL ENERGY EFFICIENCY OBJECTIVE WITH RENEWABLE ENERGY, ENERGY ACCESS, AND CLIMATE MITIGATION TARGETSJay 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

ABSTRACTThe 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.

1

INTRODUCTION

BACKGROUNDEnergy 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 GtCO 2/year)1 by 20542 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 EFFICIENCYEnergy 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 1 1 tC is equal to 3.67 tCO2, given the atomic mass ratio of 44/12.2 The study by Pacala et al. considered a 50-year window from the present (then 2004) and made reference to 2054 as a target year, rather than the more commonly used 2050 benchmark. 3 AEEI is discussed in more depth in the model input data assumptions.

2

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 EFFICIENCYThe 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).

3

Doubling global rate of

improvement of energy effi ciency

Doubling share of renewable

energy in global mix

Electrification CookingEnergy

Effi ciencyRenewable

Energy1990 76 47 -1.3 16.62010 83 59 -1.3 17.82012 84.6 58.4 -1.7 18.12030 (projected) 89 72 -2.2 242030 (target) 100 100 -2.6 36

Universal access to modern energy services

Year

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.

0

5

10

15

20

25

30

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

Prim

ary E

nerg

y Int

ensi

ty (M

J/do

llar P

PP)

Africa

Australia

Canada

China

Central and South America

Eastern Europe

Former Soviet Union

India

Japan

Middle East

Mexico

Other Developing Asia

South Korea

United States

Western Europe

Europe

GLOBAL

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

4

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.

-3.5%

-3.0%

-2.5%

-2.0%

-1.5%

-1.0%

-0.5%

0.0%

0.5%

1.0%

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

Rate

of C

hang

e (%

/yea

r)

Annual Change5-year average10-year average20-year average5-year CAGR10-year CAGR20-year CAGRGTF 10-year CAGRGTF 20-year CAGR

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 EFFICIENCYEngineering 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

5

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).

A summary of many of the other potential political, economic, social, technological, legal, environmental, and governmental (PESTLEG) barriers to energy efficiency improvements is presented in Table 2.

Table 2. Barriers to energy efficiency improvements

Barrier Description

Politi

cal

Financial priorityOther investments or expenditures might take precedence over energy efficiency measures in municipality budgets, even though measures have a high return on investment (Ecorys, 2012).

Value priorityOther values such as cultural and historical could be prioritized over desired measures such as energy efficiency improvements (Meijer et. al, 2009)

High initial cost A high initial cost could stop the measure adoption (Anderson & Newell, 2004; Yearwood Travesan, et al., 2013).

Lack of funding possibilities Access to capital is limited even though the investment is judged as profitable. A bank might not be familiar with the measure or the financial return of the measure is too far in the future, leading to

6

Economic

restricted lending (Yearwood Travezan, et al., 2013).

Low rate of return A low financial rate of return could hinder measure adoption (Ástmarsson, et al. 2013).

Long payback period A long payback period could hinder measure adoption (Anderson & Newell, 2004; Yearwood Travezan, et al., 2013).

Incentive for owner(s)/user(s)

Principal-agent dilemma could exist: e.g. the owner of a building has no direct incentive to improve energy-efficiency, through retrofitting, since the costs for energy (and water) are paid by the tenant (Ástmarsson, et al., 2013). The building owner should also have an incentive in those cases where some of the energy costs are included in the rent. This incentive gets lowered if the rent is regulated since the owner does not want to implement measures if the costs cannot be transferred to the tenants (Lind, 2012).

Uncertainty

The measure might get uneconomical due to cost increases, lower improvements than expected, non-optimal operation, delayed construction or energy cost changes (Yearwood Travezan, et al., 2013).

Transparency of funding possibilities

Lack of transparency makes it hard to find funding from public sources (Yearwood Travezan, et al., 2013).

Soci

al

Split incentivesThe decision to invest in a measure is shared between several actors and seeking agreement from all actors can be difficult (Meijer et al., 2009).

Low awareness

Low awareness and knowledge of the potential and multiple benefits of energy efficiency (energy savings, increased comfort, reduced costs, etc.) could affect measure adoption (Schleich, 2009). Low knowledge can also lead to rebound effects (Immendoerfer, et al., 2014).

Limited resources Competent staff for the assessment of potentials and risks and for leading the implementation of improvements could be limited.

Risk averse behavior Risk adverse behavior, especially for measures that are not well known, could affect measure adoption (Farsi, 2010).

Opposition/resistanceSocial opposition could lower measure adoption, e.g. opposition against renewable energy production with wind turbines, against refurbishing certain buildings or against development of cycling paths.

Tech

nolo

gica

l

Infrastructure lock-in

Existing infrastructure, with associated business models and human behavior, could hinder the implementation of a measure, e.g. the operator of an existing thermal heat grid could oppose investments in building insulation.

Lack of enabling infrastructure The lack of enabling infrastructure, e.g. smart meters, could hinder other measures such as the adoption of smart household appliances.

High technology adaptationThe involved technologies could require a certain amount of adaptation to fit with the intention of a measure. High need for adaptation might lower measure adoption (Fleiter, et al., 2012).

Lega

l

Limited data access

Access to quality data, e.g. on electricity, heat, gas and water consumption, is crucial to the identification, implementation and monitoring of measures. Unavailability of data, data censoring because of privacy issues or proprietary rights, limited data collection or poor data structuring, could lower measure adoption (McKenna et al., 2012).

Permitting procedure Permitting procedures and support structures could be spread out over several authorities (Ecorys, 2012).

7

Envi

ronm

enta

l

Environmental side-effects

A measure that will result in environmental side-effects could face challenges. An example could be energy efficiency improvements with materials (e.g. refrigerants) that use environmentally harmful substances.

Gove

rnan

ce

Procurement process clarityThe procurement process of organizations, mostly public sector agents, is not uniform for all departments, has conflicting objectives or lacks clarity (Borg, et al., 2006; Yearwood Travezan, et al., 2013).

Monitoring of quality Lacking quality control could decrease the success of a measure (Immendoerfer, et al., 2014).

CURRENT POLICIES FOR ENERGY EFFICIENCYMany countries have future targets for reducing energy intensity, with varying degrees of ambition. A selection of major economies that have adopted targets to reduce energy intensity is given in Table 3. Individual national targets within the EU are assumed to be subsumed by the EU Energy Efficiency Directive (2012/27/EU). The goal of this directive is to reduce primary energy consumption in 2020 by over 15,000 PJ, relative to a reference scenario provided within the policy. This is an ambitious target, nearly doubling the historic rate of energy intensity improvement, and it covers countries that represent a large amount of energy consumption. Japan seeks to reduce energy intensity of GDP by 30% by 2030, relative to 2003 (ABB, 2012a). This is also quite ambitious considering historic rates of reduction, and also on account that Japan already has a low energy intensity of GDP. South Korea seeks to reduce energy intensity by 46% between the years 2007 and 2030 (ABB, 2013b). Like Japan, this is quite an ambitious target, given the historic trend in energy intensity. Russia and Kazakhstan has the goal to reduce energy intensity by 40% by 2020 relative to 2007 (ABB, 2012b) and 2008 (Kazakhstan Energy Charter Secretariat & Kazenergy, 2014), respectively. Turkey seeks to reduce energy intensity of GDP by 20% between 2008 and 2023 (ABB, 2013a). In Brazil, implementation of the National Policy for Energy Efficiency is expected to result in a gradual energy savings up to 106 TWh/year to be reached in 2030 (ABB, 2013c). Brazil also has policies specifically designed to reduce electricity consumption (ABB, 2013c), not included in Table 3. The New Zealand Energy Policy promotes energy intensity improvement of 1.3 percent per annum for the years 2010-2030 (New Zealand Ministry of Economic Development, 2011). As part of their 12th 5-year plan, China sought to reduce energy intensity of GDP by 16% by 2015 (ABB, 2013d). This is now nearly a historic target, but the 2015 data are not yet available. The 13th 5-year plan will be released in early 2016. Finally, India seeks to reduce energy intensity 20% by 2020 from 2005 levels, as part of their 12th Five Year Plan (Planning Commission Government of India, 2013). Many other countries have energy efficiency policies targeted at improving specific technologies or sectors, with various metrics for assessment (e.g. the US CAFE standards for vehicle fuel economy). Those are not included in Table 3 , as they are not a policy directly targeting national energy intensity of GDP.

8

Table 3. Current Energy Intensity reduction policies of major economies. Historic CAGR is calculated from Enerdata (2015). The estimated annual energy savings at the target year is that reported by the specific policy or projected from historic CAGR values versus the target value, calculated with per capita GPD PPP projections from OECD (2014) and population projections from the World Bank (2014). *No per capita GDP projections were available for Kazakhstan; therefore, it is assumed that the ratio of per capita GDP to Russia in 2010 is the same in 2020.

Country/ RegionHistoric CAGR (1990-2010)

Target year

Target CAGR (2010-target year)

Estimated Annual Energy Savings at Target Year (PJ)

EU -1.6% 2020 -2.7% 15407

Japan -0.3% 2030 -1.6% 5844

Russia -1.5% 2020 -2.7% 4345

Turkey -0.2% 2023 -1.9% 1604

South Korea -0.1% 2030 -3.2% 1172

Brazil 0.2% 2030 0.1% 382

New Zealand -0.8% 2030 -1.3% 154

Kazakhstan -2.1% 2020 -2.4% 136*

India -2.1% 2020 -1.2% -4110

China -4.7% 2015 -3.6% -7787

India and China are interesting cases, as the targets for improvement in energy intensity are below the historic rates of reduction. This leads to a negative energy savings, and can be interpreted as targets that are not particularly ambitious. On the other hand, the historic rate of reduction was higher from 1990-2000 for China (Figure 1), in particular. There is also a lot of uncertainty concerning both China’s GDP and China’s energy consumption (Akimoto, et al., 2006; Gregg, et al., 2008; Sinton, 2001). Nevertheless, India and China are projected to become a larger share of the global economy and also dramatically increase their energy consumption in the future. How these countries develop will greatly influence the global energy intensity, and whether or not the SE4ALL energy efficiency objective is ultimately met.

OUTLINEThis study provides background Energy Technology System Analysis Program TIMES Integrated Assessment Model (ETSAP-TIAM), and then creates scenarios that constrain the technical pathways for meeting different combinations of the SE4ALL objectives. From there, we identify regions and sectors where the most potential lies for energy efficiency improvements from the criteria of cost effectiveness, and the extent to which the various SE4ALL objectives support each other in terms of investment costs and greenhouse gas emissions.

9

The main objective of this study is to determine how the global SE4ALL objectives on energy efficiency (in terms of the rate of energy intensity reduction) can be achieved by 2030 given the simultaneous goals for renewable energy and universal energy access.

METHOD

OVERVIEWFirst, a base scenario is run which represents the default assumptions for technology improvement and cost optimization with no policy incentives or other constraints on technology development. Next, reference scenarios are created based on the historical rates of EIIR, a default energy system (described below), and current carbon taxes. Alternative scenarios are created that represent different constraints (universal access as expressed in residential electricity consumption and phase out of traditional biomass, renewable energy targets, and energy efficiency for various regions). The scenarios are modeled using ETSAP-TIAM. When comparing the alternative scenarios to the reference scenario(s), it is possible to determine the effect the SE4ALL targets have on the technological and structural development within the energy system. The framework for the analysis is presented in Figure 3.

Changes in:Renewable Energy Profile

Energy Consumption,GHG Emissions

CostsAlternative Scenario

-Renewable Energy Targets for 2010-2030

-Energy Intensity Targetsfor 2010-2030

-Increased Energy Access & Phase-out Traditional Biomass

Current: -Carbon price

-Renewable Energy Profile-Energy efficiency trends-Traditional biomass use

-Technology profiles

ETSAP-TIAMReferenceScenario

Assessment of PathwaysRegional PotentialsSector & Subsector

PotentialsPolicy Recommendations

ETSAP-TIAM

Figure 3. Diagram of framework for analysis and work flow.

10

SCENARIOSThe following six scenarios are constructed and input into ETSAP-TIAM. In all scenarios, ETSAP-TIAM optimizes the energy systems based on resource availability, existing infrastructure stock, and prices given the exogenous constraints. Thus, constraints on resources can define the technology choices because the process of switching across energy carriers generally accompanies technological changes.

(i) Baseline (BASE): This scenario includes the basic model structure and available technologies, but no policy constraints or targets for energy efficiency or renewable energy, no carbon price, and no barriers to efficiency improvements. Thus it represents a cost optimal solution to meeting the energy service demands.

(ii) Reference (REF): This scenario reflects the development of the global, regional and sectoral energy demand if current trends are continued and current policies and pledges come to fruition. This scenario takes into account current technological mixes, performance and cost data for conventional technologies, and default assumptions for AEEI. It also takes into account the current carbon price, holding it constant until 2030. It does not, however, take into consideration any major energy efficiency improvements and policy interventions. Two alternate Reference scenarios are constructed:a. Regional historic trends (REFReg): For each ETSAP-TIAM region, improvements in energy intensity

were projected using OECD (2014) GDP PPP projections until 2030, and the historic average annual reduction rate of energy intensity for the years 1990-2013, calculated from Enerdata (2015) (Figure 1 and Table 5).

b. Global historic trend (REFGbl): Improvements in energy intensity were projected at the global level using OECD (2014) GDP PPP projections until 2030, and the historic average annual reduction rate of energy intensity for the years 1990-2013, calculated from Enerdata (2015) (Figure 1 and Table 5). No regional constraints are applied, allowing ETSAP-TIAM to optimize the regional allocation of energy efficiency improvements.

(iii) Renewable Energy Scenario (RE): This scenario sets region-specific targets for renewable energy deployment based off the realistic potential outlined by International Renewable Energy Agency (IRENA) global renewable roadmap (REMap2030). It uses the energy efficiency assumptions of the reference scenario.

(iv) Energy Efficiency Scenario (EE): This scenario aims at achieving SE4ALL objective related to energy efficiency by achieving a 2.6% reduction in global energy intensity in 2030, relative to 2010.

(v) Energy Efficiency and Renewable Energy Scenario (EE&RE): This combines the constraints from both the renewable energy and Energy Efficiency scenarios. Two alternative versions of this scenario are created:a. EE&RE: A renewable energy that does not phase out traditional biomass phase-out or electricity

access as part of the scenario.b. EE&RE&EA Renewable energy that also phases out the use of traditional biomass phase-out, and

meets a minimum electricity demand, thus achieving the three SE4ALL objectives.

CARBON PRICEThe current carbon price is included in all scenarios except the base scenario. The World Bank (2014) released a report that documented the current state of carbon taxes and carbon emission trading schemes (ETS) and their price levels. Some changes and updates to these carbon pricing schemes have occurred: e.g., the carbon tax in Australia was scrapped in July 2014 (Dayton, 2014) and an ETS started in the Republic of Korea, changing the carbon price levels (World Bank, 2015a). Further information on ETS was taken from the International Carbon

11

Action Partnership (ICAP, 2015) and from other nation specific sources (China Carbon, 2015; Cho, 2015; OTC-X, 2015).

While ETSAP-TIAM is capable of simulating cap-and-trade carbon markets such as the ETS, for simplicity, carbon markets were modeled as a tax by taking the current carbon price. Some nations have more than one pricing mechanism operating simultaneously, e.g. a national tax and ETS. In such cases, the prices were summed into one price applicable to the specific sector and region. Some regions have several carbon prices applying to different sectors, and this was retained in the ETSAP-TIAM input. Mexico has a carbon tax applying to fossil fuels, where the tax is the difference between the emissions from combustion of petroleum and coal versus the emissions that would have occurred were natural gas used instead, in effect, creating a tax on emissions from petroleum and coal. For Mexico, we applied a 25% ratio for petroleum, and a 40% ratio for coal, representative of the approximate ratios in emissions per unit of energy relative to natural gas. In the case where a country has both an upper and lower bound for carbon, then the upper bound was used.

The carbon prices were then aggregated to the ETSAP-TIAM regions. This aggregation was done by computing the nation’s share of energy (and cement production) carbon emissions relative to the total emissions from its corresponding ETSAP-TIAM region. The carbon price was then converted to 2005 US dollars 4 and scaled by this amount. An analogous computation was performed for carbon prices applying only to specific states in the USA, provinces in Canada, and cities in China and Japan. Data on greenhouse gas emissions and the share for different nations, states and cities were taken from the Carbon Dioxide Information Analysis Center (CDIAC) (Boden, Andres, & Marland, 2010), from the Global Carbon Atlas (Global Carbon Project, 2014), from Environment Canada (Environment Canada, 2015), from United States Environmental Protection Agency (EPA, 2014), from Wang, Zhang, Liu, & Bi (2012) and from the World Bank (2014). Carbon taxes are summarized in Table 4 and are applied in ETSAP-TIAM for the periods 2015-2030 in the reference scenario.

Table 4. Current carbon prices in 2005 USD per Tonne CO2.

Region Industry Power Heat BuildingsTransport (excluding Aviation)

Agriculture Oil Coal

AfricaAustralia/ New ZealandCanada 4.68 6.39 1.00China 0.88 0.88 0.88Central and South AmericaEastern Europe 5.51 5.51 5.51Former Soviet Union 0.72 0.72 0.72 0.63 0.63

IndiaJapan 1.16 1.16 1.16 1.16 1.16Middle EastMexico 0.62 1.00

4 Exchange rates from:

https://www.ecb.europa.eu/stats/exchange/eurofxref/html/eurofxref-graph-usd.en.htmlhttp://www.xe.com/http://www.bankofcanada.ca/rates/exchange/daily-converter/

12

Other Developing AsiaSouth Korea 4.93 4.93 4.93 4.93 4.93 4.93USA 0.06 0.14Western Europe 7.02 11.35 7.02 5.43

ENERGY EFFICIENCYThe historic average rate of annual change in energy intensity is calculated from the Enerdata (2015) statistics for the historic years 1990-2013 for each ETSAP-TIAM region, and for the world. In the REF-Regional scenario, the historic regional EIIR were extended for the years 2010-2030 (Table 5). By multiplying these energy intensity projections by the OECD (2014) GDP PPP projections, a total primary energy constraint was created for each region. When summed, this produces a global reduction rate higher than the historic global reduction rate ( Table5). This is because developing countries, which have higher historic rates of reduction in energy intensity, are projected to make up a larger proportion of the global GDP in the future. REF-World, the average reduction rate for the years 1990-2030 was extrapolated to the years 2010-2030. This variant of the reference scenario leaves more flexibility of ETSAP-TIAM to make regional improvements where they are most cost effective.

Similar to the process for establishing the bounds in the reference scenario, the SE4ALL objective of a 2.6% EIIR by 2030 (relative to 2010) was determined from the exogenous global GDP PPP projections from the OECD (2014) and applying a 2.6% reduction in energy intensity for the years 2010-2030. From here, 2030 global total primary energy constraints were established within ETSAP-TIAM (Table 5).

Table 5. EIIR for the reference and alternative energy efficiency scenarios. These are based off the historical average change over the years 1990-2013, and applied to the years 2010-2030 within ETSAP-TIAM.

Region

REFReg(CAGR 2010-2030)

REFGbl(CAGR 2010-2030)

EE Scenarios(CAGR 2010-2030)

AFR -0.8%

Optimized by ETSAP-TIAM Optimized by ETSAP-TIAM

AUS -1.4%CAN -1.4%CHI -4.3%CSA -0.4%EEU -3.5%FSU -1.4%IND -2.1%JPN -0.3%MEA 1.5%MEX -0.7%ODA -0.2%SKO 0.0%USA -1.7%WEU -1.2%Global -1.4% -1.2% -2.6%

The calculations for EIIR do not distinguish between changes in energy intensity due to improvements in technological efficiency and changes in energy intensity due to structural change in the economy. This is an active

13

area of research (see, for example, the adjusted EIIR analysis done by the Global Tracking Framework (SE4ALL, 2015)). Furthermore, there are barriers to both technological improvements and structural changes in the economy, as discussed in the background of this report. ETSAP-TIAM does not have any other way to explicitly model these various barriers. Nevertheless, the barriers are implicitly assumed in ETSAP-TIAM, because the reference scenario is defined by the historic EIIR; given that energy service demands are a result of exogenous drivers, this has the effect of limiting the EIIR and therefore representing barriers to energy efficiency improvements. In other words, this constraint would limit the EIIR to not exceed historical rates of improvement for the reference scenario. The alternative scenarios for energy efficiency are far more ambitious, and therefor may result in a regional improving efficiency at a rate greater than what has historically transpired.

RENEWABLE ENERGYRE targets were obtained for each of the ETSAP-TIAM regions from the IRENA REMap2030 study (IRENA, 2014). In the REMap study, 26 countries were analyzed, and included the renewable energy share of final energy consumption in the base year, 2010. Targets for a 27th country, Poland, have been obtained from IRENA, representing the first country within the EEA (Eastern European Union) region in ETSAP-TIAM. Further adjustments and alignments of the RE model scenario have been included after a workshop meeting with IRENA experts.

The REMap 2010 data and reference scenario 2030 projections were used to define the reference scenario in ETSAP-TIAM. To do so, the IRENA countries’ renewable energy targets were weighted according to their share of the Total Final Energy Consumption (TFEC) (IEA, 2014) within their respective regions. The IRENA REMap study also created an optimistic scenario for renewable energy deployment based on what they determined to be realistic potential (IRENA, 2014). Based on this, targets for renewable energy shares were also created for the ETSAP-TIAM regions in 2030 for the RE scenarios. A summary of the IRENA renewable energy targets is given in Table 6.

Table 6. Regional renewable energy shares of final energy consumption for 2010 and 2030 reference and 2030 RE scenarios (IRENA, 2014).

Region

REMap 2010 REMap 2030 Reference Scenario 2030 REMap Scenario

AFR 15% 9% 21%AUS 7% 12% 23%CAN 21% 22% 33%CHI 7% 16% 25%CSA 41% 40% 54%EEU 4% 6% 9%FSU 4% 8% 15%IND 17% 12% 25%JPN 4% 10% 19%MEA 5% 10% 15%MEX 4% 10% 21%ODA 5% 6% 21%SKO 3% 8% 13%USA 8% 10% 27%WEU 10% 21% 33%

14

UNIVERSAL ACCESSThe IEA estimates in their central scenario the number of people in 2030 without access to electricity to below 1 billion and without access to clean cooking facilities to just above 2.5 billion (IEA, 2014). The universal energy access target for 2030 is defined as 100% access to electricity and 100% primary reliance to non-solid fuel (SE4ALL, 2013a). The SE4ALL initiative stresses that these binary targets fail to capture many aspects of energy access, such as not considering energy applications outside of the household sector (SE4ALL, 2013b). An official target for energy electricity consumption is lacking. The SE4ALL scenario in the Global Energy Assessment assumes a 100% electrification rate and household electricity consumption of 420 kWh/year (SE4ALL, 2013b). This corresponds to the use of lighting, air circulation, televisions and light appliances according to World Bank’s tiered electricity consumption framework. The level can be traced back to a study in a Tanzanian village where the average household electricity consumption was 35 kWh/month (Ilskog, Kjellström, Gullberg, Katyega, & Chambala, 2005). Bazilian & Pielke (2013) criticize this level of energy access, pointing out that the per capita electricity consumption in wealthy countries is at least ten times higher. They emphasize the importance of electricity for businesses, industries and hospitals for economic development and want more focus on universal modern energy access that alleviates poverty.

Many studies have looked into the correlation between energy or electricity consumption and economic development (e.g. Asafu-Adjaye, 2000; Lee, 2006; Shiu & Lam, 2004; Wolde-Rufael, 2006). The direction of causality cannot always be shown, but the fact that energy or electricity consumption has a positive connection with economic development (measured in GDP) is clear (Ozturk, 2010). From this follows that one part of a crude target for energy access could be to set a level of electricity consumption that is close to that of wealthy countries. Statistics for per capita electricity consumption in 2011 and the share of the population that had access to electricity in 2010 are shown in Table 7 for certain countries and regions (World Bank, 2015b).

Table 7. Per capita electricity consumption and electricity access for selected countries

Country Per capita electricity consumption (kWh/capita/year)

Electricity access (share of population)

USA 13246 100%European Union 6115 100%South Africa 4606 82.7%China 3298 99.7%World 3045 83.1%India 684 75%Least developed countries: UN classification5

174 31.5%

Haiti 32 33.9%

Other ways of measuring energy access have been proposed. For example, Nussbaumer, Bazilian, & Modi (2012) review energy access (or poverty) metrics and suggest a multidimensional energy poverty index that focuses on energy services; cooking, lighting, entertainment and education, communication and services from household appliances.

Chakravarty & Tavoni (2013) calculate the additional energy consumption in 2030 resulting from eradicating energy poverty. They first map the number of people on different levels of energy consumption by using a model

5 The least developed countries consist of 48 countries with a total population of around 900 million people (World Bank, 2015c).

15

that builds on income distribution data. They then estimate the additional residential electricity consumption in 2030 that comes from raising energy poor people’s electricity use to at least 750 kWh/capita/year. This level is called Productive uses by the UN and corresponds to the level in the IEA’s energy access model used to calculate the investments needed to achieve the SE4All objectives (SE4ALL, 2013b). The level assumes electricity for lighting, health, education, communication and use in the agricultural sector. It should be noted that the availability of more efficient technologies will reduce the electricity use target over time, whereas electrification of transport and boilers and heaters will act in the other direction. Chakravarty & Tavoni (2013) also assume that everyone uses at least 150 kg oil/capita/year. The estimated additional residential consumption of electricity and oil in the ETSAP-TIAM regions based on their data is shown in Table 8. Only the additional residential electricity consumption in 2030 is introduced in the universal energy access scenario. The additional oil consumption is left out since some extra fossil fuel consumption will occur due to the phase out of traditional biomass as described below.

Table 8. Additional residential energy consumption in 2030 to eradicate energy poverty

Electricity (PJ) Oil (PJ)Africa 349 4501India 122 1578Other Developing Asia 140 1810China 31 404Central and South America 30 390

In order to represent the 100% non-primary reliance on solid fuel SE4ALL goal, traditional biomass is set to decrease 7.5% per year, and is phased out entirely by 2030 for the alternative scenarios (Figure 4). Lacking any detailed literature on pathways to phase out traditional biomass by 2030, this rate of decrease (7.5%/year) was chosen to create a roughly linear decline to 2030. Additionally, the constraints (on the minimum amount of biomass to be used in the residential water heating and space heating) were relaxed in ETSAP-TIAM for energy sources to hot water and space heating, allowing a greater degree of fuel switching for these end use demands.

16

0

5000

10000

15000

20000

25000

30000

35000

2005 2010 2015 2020 2025 2030

Annu

al C

onsu

mpti

on (P

J)

MEX

CSA

ODA

IND

CHI

AFR

Figure 4. Phase out of traditional biomass use for alternative scenarios.

ASSESSMENTThe scenarios are assessed ETSAP-TIAM, which optimizes the energy system to meet the energy service demands on the basis of total system cost. ETSAP-TIAM is one of the most detailed IAMs in terms of its technology database. Furthermore, there are many options for creating constraints, targets, or other policy incentives, allowing for the creation of the scenarios described above. Each alternative scenario’s model output will be subtracted from output in the reference scenario, yielding an estimate of the ‘saving’ effect of the alternative scenario.

ETSAP-TIAMTIMES Architecture BackgroundThe TIMES (The Integrated MARKAL-EFOM System) model generator, is an evolved version of MARKAL (MARket Allocation model), developed under the IEA implementing agreement, ETSAP. TIMES is a model generating set of optimization equations6 that computes an inter-temporal dynamic partial equilibrium on energy and emission markets based on the maximization of total surplus (defined as the sum of supplier and consumer surpluses). In essence, a model generated by TIMES finds the least-cost solution for the entire energy system with flexibility in terms of time resolution and sectorial focus.

Model StructureAs ETSAP-TIAM is based on the TIMES equations, it is a perfect foresight, linear optimization model (ETSAP-TIAM optimizes all time periods simultaneously). The objective function that is maximized is the discounted net present value7 of the total surplus8 for the entire world. The surplus maximization can be subject to many exogenously-

6 A complete description of the TIMES equations appears in http://www.iea-etsap.org/web/Documentation.asp.

7 A discount rate of 5% is assumed. Net present value is calculated to 2005.8 Total surplus is here defined as the sum of supplier and consumer surpluses.

17

defined constraints on a regional, sectoral or global basis, such as supply bounds (in the form of detailed supply curves that describe resource availability at different price points) for the primary resources, technical constraints governing the creation, operation, and abandonment of each technology, balance constraints for all energy forms and emissions, timing of investment payments and other cash flows, and the satisfaction of a set of demands for energy services in all sectors of the economy.

As an integrated energy system model, ETSAP-TIAM is built to represent the total energy chain, including energy extraction, conversion and demand (e.g., fossil and renewable resources), potentials of storage of CO 2 (which comes into play with a carbon price and can be adjusted via cost parameters) and region-specific demand developments. The region and sector-specific demands for end-use energy and industrial products are driven by socio-economic parameters which are described below. The model contains explicit detailed descriptions of hundreds of technologies as well as hundreds of energy, emission and demand flows within each region (region-specific parameters can be defined), logically interconnected to form a Base Energy System (Figure 5). Such technological detail allows precise tracking of optimal capital turnover, and provides a precise description of technology and fuel competition. The long-distance trade of energy between the regions of ETSAP-TIAM is endogenously modeled for coal, natural gas (gaseous or liquefied), crude oil, various refined petroleum products, and biofuels. Global and regional (partial agreement) GHG emission trading is also possible. ETSAP-TIAM is driven by a set of demands for energy services in agriculture, residential buildings, commercial buildings, industry, and transportation. Each technology has a hurdle rate that varies from 5% to 20%, depending on the sector. The hurdle rate is used to convert the capital cost in an annual cash flow: discounted multi-year interest rate payments are included when calculating an annual payment for an investment and payback time (a technology with a high hurdle rate means a short payback rate is required, while a technology with a low hurdle rate allows a longer payback time). Learning curves are exogenously assumed for each technology through the price inputs contained in the ETSAP-TIAM database. Thus technologies generally become cheaper in future time periods.

The model's variables include the investments, capacities, and activity levels of all technologies at each period of time, as well as the amounts of energy, material, and emission flows in and out of each technology, and the quantities of traded energy between all pairs or regions. For sectors that use electricity and heat, the flow variables are defined for each of six time-slices: three seasons (summer, winter, and autumn/spring) times two diurnal (day and night) divisions. ETSAP-TIAM is a partial equilibrium model, and although it does not include macroeconomic variables beyond the energy sector, there is evidence that accounting for price elasticity of demands captures the majority of the feedback effects from the economy to the energy system (Bataille, 2005; Labriet, et al., 2012; Scheper & Kram, 1994).

18

Climate Module

Atm. Conc.ΔForcing ΔTemp

Used for reporting &

setting targets

Biomass Potential

Other Renewables

Nuclear

Fossil Fuel Reserves

(oil, coal, gas)Extraction Upstream

Fuels

Trade

Secondary Transformation

OPEC/NON-OPEC regrouping

Power and Heat Fuels

Electricity

Cogeneration

Heat

Hydrogen production and distribution

End Use Fuels

Industrial Service

Composition Auto Production

Cogeneration

Carbon captureCH4 options Carbon

sequestrationTerrestrial

sequestration

Landfills Manure Bio burning, rice, enteric ferm Wastewater

CH4 options

N2O options

CH4 options

OI****

GA****CO****

Trade

ELC***

WIN SOL GEO TDL

BIO***

NUC

HYD

BIO***

HETHET

ELCELC

SYNH2

BIO***

CO2

ELC

GAS***COA***

Industrial Tech.

Commercial Tech.

Transport Tech.

Residential Tech.

Agriculture Tech.

I***

I** (6) T** (16)R** (11)C** (8)A** (1)

INDELC

INDELCIS**

Demands

IND*** COM***AGR*** TRA***RES***

Non-energy sectors (CH4)

OIL***

Figure 5. Base energy system within ETSAP-TIAM. Technological efficiencies are included in the industrial, agriculture, commercial, residential, and transport technology boxes. Other efficiency adjustments are possible within the fuel production chains.

ETSAP-TIAM integrates a climate module permitting the computation and modeling of global changes related to GHG concentrations, radiative forcing and global temperature increase. The climate module was originally inspired by the Nordhaus and Boyer (1999) model, but now consists of three sets of equations, dynamically calculating the atmospheric concentrations of the three main GHGs (CO2, CH4, and N2O), the atmospheric radiative forcing of these three gases, and the resultant change in mean global temperature. The climate module has been calibrated and compared to other, more detailed climate modules, during several past multi-model experiments (Loulou, et al., 2009). The CO2, CH4, and N2O emissions related to the energy sector are explicitly represented in the model at the level of the individual technologies. The emissions from non-energy sectors (landfills, manure, rice paddies, enteric fermentation, wastewater, agriculture, land-use) are also included in the model, but in a more rudimentary way. The other GHGs (CFCs, HFCs, SF6, etc.) are not explicitly modeled, but their radiative forcing is represented in an exogenous manner. Options for GHG emission reductions available in the model include: specific CH4 and N2O destruction, mitigation of emissions from agriculture, CO2 capture (upstream, power plants, biofuel refineries, hydrogen generation) and sequestration (in geological sinks), biological sequestration via reforestation, and finally, numerous fuel and technology switching options in each sector (which would simultaneously improve energy efficiency and correspondingly induce a reduction in energy intensity). Thus, carbon price can be used as a simple lever for policy intervention, and this can be applied globally or differentially across regions.

19

Regions and Time FrameIn ETSAP-TIAM, the world is divided into 15 regions (Figure 6).

ETSAP-TIAM Regions

AFR Africa

AUS Australia & NZ

CAN Canada

CHI China

CSA Central and South America

EEU Eastern Europe

FSU Form er Soviet Union

IND India

JPN Japan

MEA M iddle East

MEX Mexico

ODA Other Developing Asia

SKO South Korea

USA United States

WEU W estern Europe

Figure 6. Fifteen regions of the Energy Technology System Analysis Program TIMES Integrated Assessment Model (ETSAP-TIAM).

The model architecture of ETSAP-TIAM is built off of 2005 data. Primary energy consumption and demand driver data (population, GDP, number of households, etc.) have been updated and constrained within ETSAP-TIAM to match key 2010 historical data, thus making 2010 a de-facto base year. This is done in order to avoid optimization of the past. 2010 also serves as a base year in energy efficiency improvement calculations. From 2010, ETSP-TIAM is run on 5-year time steps to the SE4ALL target year of 2030.

20

RESULTS

Figure 7 displays pathways for energy intensity, by scenario. The EE&RE and EE&RE&EA scenarios follow the same path (by design; they achieve the SE4ALL energy efficiency target of 2.6% compound annual reduction relative to 2010). The RE scenario has a pathway that closely follows the BASE scenario. The two reference scenarios have the lowest rate of reduction.

Figure 7. Global primary energy intensity pathways, by scenario.

21

In Figure 8, total global net present value (NPV) of investment costs are shown, per sector. In the RE scenario, most of the investment is needed in the transportation sector, while the residential sector can realize cost savings by meeting renewable energy targets. In the EE scenario, the majority of the investment must also be directed toward the transport sector. In comparing the EE&RE to the EE&RE&EA scenarios, it is clear that the later requires far more investment in the residential sector.

Figure 8. Global sectoral investment (net present value), by scenario

22

In Figure 9, large savings in renewable energy can be found in Other Developing Asia. Western Europe and the United States will require the most investment in order to meet the SE4ALL Energy Efficiency objectives. Meeting the energy access objective will require substantially more investment in currently developing regions: Africa, Other Developing Asia, India, and Central and South America.

Figure 9. Regional investment (net present value), by scenario.

23

The global CO2 emissions pathways for the energy system are presented in Figure 10. The Representative Concentration Pathway that maintains 2.6 W/m2 radiative forcing (RCP 2.6), which would likely prevent Earth from warming over 2° C, is the more ambitious than any of the scenarios modeled in this study. The EE scenario reduces emissions more than the RE scenario, and taken together, both together (EE&RE) reduce emissions more than either alone. The EE&RE&EA has slightly higher emissions than the EE&RE scenario through 2020 (fossil resources can substitute for traditional biomass in this scenario), but then ends on a pathway with lower emissions by 2030.

Figure 10. Global greenhouse gas emission pathways, by scenario.

24

When comparing total NPV investment versus the mitigation (Figure 11), the reference scenarios have the highest emissions. Locking in regional trends requires more total investment than a scenario with one global target. The BASE scenario is the cheapest, and actually has lower emissions than the reference scenarios, and only slightly higher emissions than either of the EE or RE scenarios. The EE scenario requires more investment than the EE&RE scenario, yet has higher emissions. The EE&RE and EE&RE&EA scenarios have the lowest emissions, with the latter requiring the most investment.

Figure 11. Total investment costs (net present value) versus total emissions (2010-2030), by scenario.

ANALYSIS

SUMMARY OF FINDINGSThe results above suggest that the SE4ALL renewable energy target is cost-effective (it saves cost over the reference scenarios). Achieving the renewable energy objective also makes achieving the energy efficiency objective cheaper, and together, they reduce greenhouse gas emissions more than either target alone. Moreover, the EE scenario promotes slightly more renewable energy deployment relative to the reference scenarios (16% versus 15%). Thus there is a synergistic relationship between these two SE4ALL objectives.

There is some likelihood that the SE4ALL Universal Energy Access objective makes the other two objectives more difficult to achieve. It is very ambitious to phase out traditional biomass by 2030, and the most economic near-term option to replace this fuel is likely to be fossil-based. Moreover, meeting the increased service demands requires more energy consumption in the developing regions of the world. Thus meeting the universal access

25

objective tends to reduce the percentage of renewable energy, even when traditional biomass is not counted as a renewable energy source. It also potentially has an effect on energy intensity, as the distribution and availability of fossil fuels would likely increase fossil energy consumption. Much of the increased investment and energy consumption would be in developing regions of the world, thus this is primarily a sustainable development initiative.

Given OECD (2014) projections of GDP, the SE4ALL energy intensity objective of 2.6% reduction, if achieved, will reduce global energy consumption by nearly 200 EJ/year in 2030 versus the historic 1.3% reduction rate. This will still mean an increase in global energy consumption of nearly 90 EJ/year relative to 2010. As such, the SE4ALL objectives are not sufficient in themselves to meet the target of remaining below 2° C global warming. Therefore, additional climate policies will be necessary to achieve a path to a 2° target, such as a price on carbon or other climate policy mechanisms.

BARRIERS TO ENERGY EFFICIENCYThe optimal BASE model solution is much cheaper than all the scenarios, including the reference scenarios, and yet the energy system in the BASE scenario has emissions that are roughly equivalent to the EE and RE scenarios. Moreover the energy efficiency pathway from the BASE scenario tracks closely with that of the RE scenario. This suggests that energy efficiency through technology improvement is a cost effective means of reducing emissions. Moreover, it shows that there are indeed real world barriers that prevent us from transitioning to an “optimal” business-as-usual solution. These barriers are thus implicitly represented in the difference between the BASE and reference scenarios. Nevertheless, it suggests the investment in energy efficiency and renewable energy makes sense, in terms of cost, as an end in itself if these barriers can be overcome (through policies, for example). It also shows there are limitations to applying results from cost optimization based modeling to the real world, thus highlighting the importance of research to better understand the barriers to adopting energy efficiency.

LIMITATIONS AND UNCERTAINTIESAs a measure of energy efficiency, energy intensity is not the most robust of statistics. This is because it incorporates uncertainties in both energy consumption and the economy. Economic uncertainty is compounded by uncertainties in purchasing power parity, which depends on the relative buying power across economies. Globalization trends dilute the interpretation of energy intensity, as production and consumption of goods are geographically separated. Furthermore, given the vicissitude of the economy, energy intensity can vary widely year to year without any discernable change in the energy technology or conservation. This is problematic when establishing future targets for energy intensity. In effect, there are two degrees of freedom, both with high degrees of uncertainty. The hidden variable in efficiency metrics is time; in the calculation of energy intensity (GDP/year divided by primary energy consumption /year), the year cancels algebraically, but tacitly, it still understood to be there. Efficiency ultimately concerns the rate of production over the rate of consumption, and greater efficiency can be attained while increasing both, so longs as the rate of production increases at a greater rate than the rate of consumption. SE4ALL objectives to promote development are obscured by potential drops in consumption by employing the energy intensity metric (e.g. a global recession). Likewise, the SE4ALL objectives are to address the climate challenge are also obscured by potential increases in production. In summary, though widely used, energy intensity of GDP as a statistic and a target is highly uncertain, difficult to forecast, and does not necessarily guarantee development or environmental goals are achieved. On the other hand, one principle advantage of energy efficiency is that it is a “no regrets” option.

26

ETSAP-TIAM is a linear model, and the results presented are an algebraically optimal solution to a set of input data and constraints. In this sense, they give a picture of how to most economically achieve an a priori energy intensity or renewable energy pathway. However, ETSAP-TIAM does have its limitations. As a linear model it cannot handle feedback effects, such as Jevons Paradox, structural changes to the economy, or economic development as a result of technological development. Thus, ETSAP-TIAM, as with all integrated assessment models, should not be considered a truth-machine and the results contained herein are not predictions or forecasts, but rather solutions to preset scenarios. IAMs with large databases quickly become outdated; this is a particular challenge with working with IAMS, as naturally, results are only as good as the input assumptions. While the demand drivers were updated for this analysis, there are thousands of technology parameters that are based on 2005 technology development assumptions. Therefore, there is some uncertainty stemming from outdated input data, particularly for technology subsectors that are rapidly developing.

Moreover, with population and GDP as exogenous demand drivers, there is some additional uncertainty in modeling the universal access objective. The phasing out of traditional biomass would likely coincide with rapid economic development, thus affecting GDP. ETSAP-TIAM is unable to handle such non-linear feedbacks.

CONCLUSIONSETSAP-TIAM is useful in determining where the largest potentials for energy efficiency lie and where the most cost effective investments in energy efficiency can be made, regionally, and technologically. This can aid in crafting efficient policies to meet the SE4ALL objectives for energy efficiency, renewable energy, and energy access, as well as provide a pathway for dramatically reducing global greenhouse gas emissions. This suggests that renewable energy is an economically attractive means to meet energy efficiency targets, and that likewise, policies that promote renewable energy make it easier to achieve energy efficiency targets. Achieving universal access will require more investment globally, and this invites more research into the larger question of values surrounding sustainable development.

27

REFERENCESABB (2013d). China energy efficiency report. http://new.abb.com/docs/librariesprovider46/EE-Document/china-report-en.pdf?sfvrsn=2 (Accessed 25 Jun, 2015)

ABB (2013c). Brazil energy efficiency report. https://library.e.abb.com/public/c6d0b52cc84505a2c1257be80052c5a7/Brazil.pdf (Accessed 25 Jun, 2015)

ABB (2013a). Turkey energy efficiency report. https://library.e.abb.com/public/a2c92d1d4f7f2405c1257be9002c5060/Turkey.pdf (Accessed 25 Jun, 2015)

ABB (2013b). South Korea energy efficiency report. https://library.e.abb.com/public/557d50223ed20a76c1257beb0044f3bc/South%20Korea.pdf?filename=South%20Korea.pdf (Accessed 25 Jun, 2015)

ABB (2012b). Russia energy efficiency report. https://library.e.abb.com/public/1112256e3dbeb710c12579e6003937a3/Russia%20Energy%20efficiency%20Report.pdf (Accessed 25 Jun, 2015)

ABB (2012a). Japan energy efficiency report. https://library.e.abb.com/public/4a34976258d9db4bc12579e600391cf7/Japan%20Energy%20efficiency%20Report.pdf (Accessed 25 Jun, 2015)

Akimoto, H., T. Ohara, J. Kurokawa, and N. Horii (2006), Verification of energy consumption in China during 1996–2003 by using satellite observational data, Atmos. Environ., 40, 7663– 7667

Allcott, H., & Greenstone, M. (2012). Is there an “energy gap”? Journal of Economic Perspectives, 26(1), 3–28. http://doi.org/10.1257/jep.26.1.3

Anderson, S. T., & R. G. Newell (2004). Information programs for technology adoption: the case of energy-efficiency audits. Resource and Energy Economics 26(1): 27–50. doi:10.1016/j.reseneeco.2003.07.001

Asafu-Adjaye, J. (2000). The relationship between energy consumption, energy prices and economic growth: time series evidence from Asian developing countries. Energy Economics 22: 615–625. http://doi.org/10.1016/S0140-9883(00)00050-5

Ástmarsson, B., P. A. Jensen & E. Maslesa (2013). Sustainable renovation of residential buildings and the landlord/tenant dilemma. Energy Policy 63: 355–362. doi:10.1016/j.enpol.2013.08.046

Bataille, C. G. F. (2005). Design and application of a technologically explicit hybrid energy-economy policy model with micro and macro-economic dynamics (Doctoral dissertation, School of Resource and Environmental Management, Simon Fraser University).

Bazilian, M., & Pielke, R. (2013). Making energy access meaningful. Issues in Science and Technology, 29(4), 74–78.

Boden, T., Andres, B., & Marland, G. (2010). Ranking of the world’s countries by 2010 total CO2 emissions from fossil-fuel burning, cement production, and gas flaring. Retrieved May 27, 2015, from http://cdiac.ornl.gov/trends/emis/top2010.tot

28

Borg, N., Y. Blume, S. Thomas, W. Irrek, H. Faninger-Lund, P. Lund, & A. Pindar (2006). Release the power of the public purse. Energy Policy 34(2): 238–250. doi:10.1016/j.enpol.2004.08.027

Chakravarty, S. & M. Tavoni (2013) Would Universal Energy Access Boost Climate Change? Review of Environment, Energy and Economics (Re3), Forthcoming. Available at SSRN: http://ssrn.com/abstract=2255974

China Carbon. (2015). Trading data. Retrieved May 27, 2015, from http://chinacarbon.net.cn

Cho, M. (2015). Trading dries up in South Korea’s new carbon market. Retrieved May 27, 2015, from http://www.reuters.com/article/2015/02/05/emission-southkorea-idUSL4N0VF1QN20150205

Dayton, L. (2014). Australia scraps carbon tax. Retrieved May 27, 2015, from http://news.sciencemag.org/asiapacific/2014/07/australia-scraps-carbon-tax

DeCanio, S. J. (1998). The efficiency paradox: bureaucratic and organizational barriers to profitable energy-saving investments. Energy Policy 26(5): 441-454.

De Groot, H. L., E. T. Verhoef, & P. Nijkamp (2001). Energy saving by firms: decision-making, barriers and policies. Energy Economics 23(6): 717-740.

Ecorys. (2012). Local investments options in Energy Efficiency in the built environment - Identifying best practices in the EU, (November).

Enerdata (2015). Global Energy Statistical Yearbook 2014. Retrieved May 20, 2015. https://yearbook.enerdata.net/energy-primary-production.html

Environment Canada. (2015). Greenhouse gas emissions by province and territory, Canada, 1990, 2005 and 2013. Retrieved May 27, 2015, from http://open.canada.ca/data/en/dataset/9a673fa1-22af-41c1-9ae5-de38de714c83

EPA (Environmental Protection Agency). (2014). State Energy CO2 Emissions. Retrieved May 27, 2015, from http://epa.gov/statelocalclimate/resources/state_energyco2inv.html

Farsi, M. (2010). Risk aversion and willingness to pay for energy efficient systems in rental apartments. Energy Policy 38(6): 3078–3088. doi:10.1016/j.enpol.2010.01.048

Fleiter, T., S. Hirzel & E. Worrell (2012). The characteristics of energy-efficiency measures – a neglected dimension. Energy Policy 51: 502–513. doi:10.1016/j.enpol.2012.08.054

Gregg, J.S., R. Andres & G. Marland (2008). China: Emissions pattern of the world leader in CO2 emissions from fossil fuel consumption and cement production. Geophysical Research Letters 35:L08806, doi:10.1029/2007GL032887

Gupta S., J. Harnisch, D. C. Barua, L. Chingambo, P. Frankel, R. J. Garrido Vázquez, L. Gómez-Echeverri, E. Haites, Y. Huang, R. Kopp, B. Lefèvre, H. Machado-Filho, & E. Massetti (2014). Cross-cutting Investment and Finance Issues. In: Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Edenhofer, O., R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J. Savolainen, S.

29

Schlömer, C. von Stechow, T. Zwickel and J.C. Minx (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

ICAP. (2015). ETS Map. Retrieved May 27, 2015, from https://icapcarbonaction.com/ets-map

Ilskog, E., B. Kjellström, M. Gullberg, M. Katyega, & W. Chambala (2005). Electrification co-operatives bring new light to rural Tanzania, Energy Policy, 33:10, 1299-1307

Immendoerfer, A., M. Winkelmann & V. Stelzer (2014). Energy Solutions for Smart Cities and Communities - Recommendations Policy Makers (p. 128).

IEA (International Energy Agency) (2014). World Energy Investment Outlook. Paris, France.

IESG (International Energy Study Group) (2015). Appliance Energy Efficiency. Lawrence Berkeley National Laboratory, US Department of Energy. https://ies.lbl.gov/research-area/appliance-energy-efficiency Accessed 4/12/2015

IRENA (International Renewable Energy Agency) (2014). REMAP 2030 - A Renewable Energy Roadmap. Abu Dhabi. http://irena.org/remap/REmap_Report_June_2014.pdf

Jaffe, A. B., Newell, R. G., & Stavins, R. N. (1999). Energy-efficient technologies and climate change policies: issues and evidence.

Kazakhstan Energy Charter Secretariat & Kazenergy (2014). Review of the National Policy of The Republic of Kazakhstan in the Area of Energy Saving and Energy Efficiency. Brussels, Belgium. http://www.encharter.org/fileadmin/user_upload/Publications/Kazakhstan_EE_2014_ENG.pdf (Accessed 25 Jun, 2015)

Kober, T. (2014). Impact of Energy Efficiency Measures on Greenhouse Gas Emission Reduction, ECN (ECN-E–14-038)

Kriegler, E., J. P. Weyant, G. J. Blanford, V. Krey, L. Clarke, J. Edmonds, A. Fawcett, G. Luderer, K. Riahi, R. Richels, S. K. Rose, M. Tavoni, & D. P. van Vuuren (2014). The role of technology for achieving climate policy objectives: overview of the EMF 27 study on global technology and climate policy strategies. Climatic Change 123(3-4): 353–367. doi:10.1007/s10584-013-0953-7

Lee, C. C. (2006). The causality relationship between energy consumption and GDP in G-11 countries revisited. Energy Policy, 34(9), 1086–1093. http://doi.org/10.1016/j.enpol.2005.04.023

Lind, H. (2012). Pricing principles and incentives for energy efficiency investments in multi-family rental housing: The case of Sweden. Energy Policy 49: 528–530. doi:10.1016/j.enpol.2012.06.054

Loulou, R., M. Labriet, & A. Kanudia (2009). Deterministic and stochastic analysis of alternative climate targets under differentiated cooperation regimes. Energy Economics 31: S131-S143.

Markandya, A., K. Halsnaes, A. Lanza, Y. Matsuoka , S. Maya, J. Pan, J. Shogren, R. S. de Motta, T. Zhang, & T. Taylor (2001). In Climate Change 2001: Mitigation, Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change [Jochem, E. (ed.)] Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

30

McKenna, E., I. Richardson & M. Thomson (2012). Smart meter data: Balancing consumer privacy concerns with legitimate applications. Energy Policy 41: 807–814. doi:10.1016/j.enpol.2011.11.049

McNeil, M.A., V. E. Letschert, S. de la Rue du Can, & J. Ke (2012). Bottom-Up Energy Analysis System – Methodology and Results. Lawrence Berkeley National Laboratory. http://www.superefficient.org/Resources/~/media/Files/BUENAS%20Methodology.pdf

Meijer, F., Itard, L., & Sunikka-Blank, M. (2009). Comparing European residential building stocks: performance, renovation and policy opportunities. Building Research & Information, 37(5-6), 533–551. doi:10.1080/09613210903189376

New Zealand Ministry of Economic Development (2011). New Zealand Energy Strategy 2011–2021: Developing our energy potential. http://www.med.govt.nz/sectors-industries/energy/pdf-docs-library/energy-strategies/nz-energy-strategy-lr.pdf (Accessed 25 Jun, 2015)

Nordhaus W. D. & J. Boyer (1999). Roll the DICE Again: Economic Models of Global Warming. Yale University, manuscript edition.

Nussbaumer, P., Bazilian, M., & Modi, V. (2012). Measuring energy poverty: Focusing on what matters. Renewable and Sustainable Energy Reviews, 16(1), 231–243. http://doi.org/10.1016/j.rser.2011.07.150

OECD (2014). Economic Outlook No 95 - May 2014 - Long-term baseline projections. http://stats.oecd.org/OECDStat_Metadata/ShowMetadata.ashx?Dataset=EO95_LTB&ShowOnWeb=true&Lang=en

OTC-X. (2015). CO2-Emissionsrecht CHU. Retrieved May 27, 2015, from https://www.otc-x.ch/markt/instrument/valor/999999.html

Ozturk, I. (2010). A literature survey on energy-growth nexus. Energy Policy, 38(1), 340–349. http://doi.org/10.1016/j.enpol.2009.09.024

Pacala, S. & R. Socolow (2004). Stabilization Wedges: Solving the Climate Problem for the Next 50 Years with Current Technologies. Science 305: 968-972

Planning Commission Government of India (2013). Twelfth Five Year Plan (2012–2017) Economic Sectors Volume II. http://planningcommission.gov.in/plans/planrel/12thplan/pdf/12fyp_vol2.pdf (Accessed 25 Jun, 2015)

Rogelj, J., D. L. McCollum & K. Riahi (2013). The UN’s ‘Sustainable Energy for All’ initiative is compatible with a warming limit of 2 °C. Nature Climate Change 3: 545-551

Shiu, A., & Lam, P.-L. (2004). Electricity consumption and economic growth in China. Energy Policy, 32(1), 47–54. http://doi.org/10.1016/S0301-4215(02)00250-1

Sinton, J. E. (2001), Accuracy and reliability of China’s energy statistics, China Econ. Rev., 12, 373– 383.

SE4ALL (Sustainable Energy for All). (2013a). Executive Summary. In Global Tracking Framework (pp. 9–16). Vienna, Austria: United Nations.

31

SE4ALL (Sustainable Energy for All). (2013b). Universal Access to Modern Energy Services. In Global Tracking Framework (pp. 74–132). Vienna, Austria: United Nations. Retrieved from http://www.sustainableenergyforall.org/tracking-progress

SE4ALL. (2015). Progress Toward Sustainable Energy 2015. 09/10/2015. Retrieved from http://www.se4all.org/wp-content/uploads/2013/09/GTF-2105-Full-Report.pdf

Scheper, E., & T. Kram (1994). Comparing MARKAL and MARKAL-MACRO for The Netherlands, ECN Policy Studies, Draft, presented at the May 1994 meeting of ETSAP.

Schleich, J., & Gruber, E. (2008). Beyond case studies: Barriers to energy efficiency in commerce and the services sector. Energy Economics 30(2): 449-464.

Sugiyama, M., O. Akashi, K. Wada, A. Kanudia, J. Li, & J. Weyant (2013). Energy efficiency potentials for global climate change mitigation. Climatic Change 123(3-4): 397–411. doi:10.1007/s10584-013-0874-5

U.S. Department of Energy (DOE). (2015). Barriers to Industrial Energy Efficiency. Retrieved October 9, 2015, from http://energy.gov/sites/prod/files/2015/06/f23/EXEC-2014-005846_5 Study__0.pdf http://energy.gov/eere/amo/downloads/barriers-industrial-energy-efficiency-study-appendix-june-2015

Yearwood Travezan, J., R. Harmsen, & G. van Toledo (2013). Policy analysis for energy efficiency in the built environment in Spain. Energy Policy 61: 317–326. doi:10.1016/j.enpol.2013.05.096

Wang, H., Zhang, R., Liu, M., & Bi, J. (2012). The carbon emissions of Chinese cities. Atmospheric Chemistry and Physics, 12(14), 6197–6206. http://doi.org/10.5194/acp-12-6197-2012

Wolde-Rufael, Y. (2006). Electricity consumption and economic growth: A time series experience for 17 African countries. Energy Policy, 34(10), 1106–1114. http://doi.org/10.1016/j.enpol.2004.10.008

World Bank (2014). Health Nutrition and Population Statistics: Population estimates and projections (Last: 09/26/2014). http://databank.worldbank.org/data/home.aspx

World Bank. (2014). State and Trends of Carbon Pricing 2014. Washington, DC.

World Bank. (2015a). Carbon Pricing Watch 2015. Washington, DC.

World Bank. (2015b). World Bank Open Data. Retrieved June 16, 2015, from http://data.worldbank.org/

World Bank. (2015c). Least developed countries: UN classification. Retrieved June 16, 2015, from http://data.worldbank.org/region/LDC

32