Realizing the Potential of Energy Efficiency Ruoyu Chen Roberto Estevez Magnasco Andrea Heins Fany Rocha Jevgenijs Steinbuks in Latin Claudia Ines Vasquez Suarez America and the Caribbean Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Authors Ruoyu Chen Assistant Professor of Economics, University of Windsor Roberto Estevez Magnasco © 2022 December| International Bank for Reconstruction and Energy Specialist, Energy and Extractives Global Practice, World Bank Development / The World Bank • 1818 H Street NW, Washington, DC 20433 • Telephone: 202-473-1000; Internet: www.worldbank.org Andrea Heins Independent Senior Specialist, Energy Efficiency and Sustainable Development Some rights reserved This work is a product of the World Bank with contributions given by Fany Rocha the staff and consultants listed in the Acknowledgments. The findings, Independent Energy Efficiency Specialist interpretations, and conclusions expressed in this work do not necessarily reflect the views of the World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the Jevgenijs Steinbuks accuracy of the data included in this work. Nothing herein shall constitute Senior Economist, Office of the Chief Economist for Infrastructure Practice, or be considered to be a limitation upon or waiver of the privileges and World Bank immunities of The World Bank, all of which are specifically reserved. ESMAP is a collaborative effort of the World Bank and 24 partners to Claudia Ines Vasquez Suarez help low- and middle-income countries reduce poverty and boost growth Lead Energy Specialist, Energy and Extractives Global Practice, World Bank through sustainable energy. ES­ MAP’s analytical and advisory services are fully integrated with the World Bank’s country financing and energy policy Graphic Project and Design dialogue. ESMAP works to accelerate the energy transition required to ensure access to affordable, reliable, sustainable, and modern energy for Nicolas Carvajal and Inti Alonso, Puntoaparte Editores all, consistent achieve Sustainable Development Goal 7 (SDG7). It helps to shape strategies and programs to achieve the World Bank Group’s December 2022 Climate Change Action Plan targets Rights and permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// creativecommons.org/licenses/by/3.0/ igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: World Bank. 2022. Realizing the Potential of Energy Efficiency in Latin America and the Caribbean. World Bank, Washington, DC. License: Creative Commons Attribution CC BY 3.0 IGO Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org.y Abstract Realizing the Potential of Energy Efficiency in Latin America and the Caribbean responds to the urgent need to relaunch Acknowledgments the energy efficiency agenda on the continent in the context of post-COVID recovery, the challenges of climate change The team is immensely grateful to Martina Bosi (Senior Cli- mitigation, and high energy prices. The report assesses the mate Change Specialist), Cesar Adrian Arreola Croda (En- current state of energy efficiency policies and measures in ergy Specialist), and Guillermo Beylis (Economist, World the region, identifies key challenges and drivers for improve- Bank) for their constructive comments. The team is thank- ment, and proposes ways forward. Starting with a high-level ful for the full-throttled support of Stephanie Gil (Practice review of how energy efficiency policies have evolved in the Manager, Latin America and Caribbean Region, Energy and region over the past two decades, the report proceeds to Extractives Global Practice). Stephen Spector, Steven Ken- disaggregate broad trends into discrete efficiency improve- nedy and Claudia Viviana Nunez De Urquidi provided excel- ments and changes in economic activity. It then identifies lent, timely editing. Agostina Signorini (Program Assistant, key drivers of energy intensity reductions at the sector level. Energy and Extractives Global Practice) provided outstand- Finally, it provides recommendations on policy instruments ing administrative support. to support energy efficiency improvements. Contents Executive Summary Page viii  1 2 The region’s record on energy efficiency Page 14 Why should the region’s countries be concerned about energy efficiency? Page 16  Main barriers to energy efficiency Quantitative Page 17  analysis Is the region energy efficient? Page 34 Page 17  Institutional context, regulation, and policies Data and measurement Page 21  Page 35  Energy efficiency potential in the region Methodological framework Page 32  Page 36  4 3 How can the region improve its energy efficiency? Some recommendations Page 50 Results of the Fisher References Page 52  decomposition Appendix A analysis Page 53  A.1. Note on data sources Page 38 Page 53  A.2. Results What have been the drivers of changes in energy Page 56  intensity? Page 40  A.3. Key questions informing RISE energy efficiency scores used for this report Econometric analysis and results Page 95  Page 41  A.4. Types of energy efficiency policies and Counterfactual simulations and results regulations in LAC Page 45  Page 98  Figures Figure 3.2. Energy intensity decomposed by changes in activity and efficiency in LAC’s 10 largest economies, 2000–19  40 Figure ES.1. Growth of primary energy supply, final Figure 3.3. Projected reductions in energy intensity energy consumption, and GDP in LAC, 2000–19  ix from eliminating subsidies  45 Figure ES.2. Reductions in energy intensity implied by Figure 3.4. Projected reductions in the EE component of better RISE scores, by country  xi energy intensity, due to elimination of subsidies 46 Figure ES.3. Improvements in energy efficiency Figure 3.6. Projected percentage changes in energy resulting from better RISE scores, by country  xii intensity with improved RISE scores  47 Figure 1.1. Growth of primary energy supply, final Figure 3.7. Projected percentage changes in energy energy consumption, and GDP in LAC, 2000–19  17 efficiency with improved RISE scores  47 Figure 1.2. Energy-intensity of primary energy in Figure 3.8. Potential energy savings produced different regions, 2000–15  18 by reforms to subsidies and regulations  48 Figure 1.3. Historical and projected total final energy Figure A.1. Fisher indices, sectoral composition, consumption in the region, 2010–50  19 and energy efficiency - LAC region  57 Figure 1.4. Evolution of final energy intensity in LAC, Figure A.2. Fisher indices, sectoral composition, 2000–19 (MJ/GDP, PPP 2017 US$)  19 and energy efficiency - Argentina  58 Figure 1.5. Ratio of final energy intensity to GDP Figure A.3. Fisher indices, sectoral composition, in major LAC countries, 2000–19  20 and energy efficiency - Belize  59 Figure 1.6. Changes in final energy intensity by sector, Figure A.4. Fisher indices, sectoral composition, 2000–19  21 and energy efficiency - Bolivia  60 Figure 1.7. Number of measures per year per country  22 Figure A.5. Fisher indices, sectoral composition, and energy efficiency - Brazil  61 Figure 1.8. Number of measures by sector, country, and type of measure, 1985-2019  23 Figure A.6. Fisher indices, sectoral composition, and energy efficiency - Chile  62 Figure 1.9. Discontinued and ongoing measures by sector, country, and type of measure, 1985-2019  25 Figure A.7. Fisher indices, sectoral composition, and energy efficiency - Colombia  63 Figure 1.10. RISE energy efficiency scores in the major countries of the region and in the rest of the region, Figure A.8. Fisher indices, sectoral composition, 2010 and 2019  30 and energy efficiency - Costa Rica  64 Figure 1.11. Overall RISE energy efficiency score, Figure A.9. Fisher indices, sectoral composition, 2010 and 2019  31 and energy efficiency - Dominican Republic  65 Figure 1.12. Correlation between RISE scores on energy Figure A.10. Fisher indices, sectoral composition, efficiency and the cumulative number of energy and energy efficiency - Ecuador  66 efficiency policies implemented in LAC, by year  31 Figure A.11. Fisher indices, sectoral composition, Figure 1.13. Projections of ratio of final energy and energy efficiency - El Salvador  67 intensity to GDP in STEPS and SDS, 2000–40  33 Figure A.12. Fisher indices, sectoral composition, Figure 2.1. Evolution of energy prices in LAC, 2000–19  35 and energy efficiency - Grenada  68 Figure 3.1. Energy intensity decomposed by changes Figure A.13. Fisher indices, sectoral composition, in activity and in efficiency, 2000–18  39 and energy efficiency - Guyana  69 Figure A.14. Fisher indices, sectoral composition, Figure A.21. Fisher indices, sectoral composition, and energy efficiency - Jamaica  70 and energy efficiency - China  77 Figure A.15. Fisher indices, sectoral composition, Figure A.22. Fisher indices, sectoral composition, and energy efficiency - Mexico  71 and energy efficiency - India  78 Figure A.16. Fisher indices, sectoral composition, Figure A.23. Fisher indices, sectoral composition, and energy efficiency - Nicaragua  72 and energy efficiency - Russian Federation  79 Figure A.17. Fisher indices, sectoral composition, Figure A.24. Fisher indices, sectoral composition, and energy efficiency - Panama  73 and energy efficiency - United States  80 Figure A.18. Fisher indices, sectoral composition, Figure A.25. Fisher indices, sectoral composition, and energy efficiency - Peru  74 and energy efficiency - EU28  81 Figure A.19. Fisher indices, sectoral composition, Figure A.26. Fisher indices, sectoral composition, and energy efficiency - Suriname  75 and energy efficiency - EU12  82 Figure A.20. Fisher indices, sectoral composition, Figure A.27. Energy efficiency policies and regulations and energy efficiency - Uruguay  76 of various types across the LAC region  98 Tables Table 2.1. Link between RISE scores, energy intensity Table A.4. Effect of removing energy subsidies index, and energy price  41 on energy efficiency  84 Table 2.2. Link between RISE scores, energy efficiency, Table A.5. Effects of policies on energy intensity  85 and energy price  42 Table A.6. Effects of policies on energy efficiency  86 Table 2.3. Link between RISE scores, economic activity, and energy price  42 Table A.7. Relationship between energy prices, regulatory policies, and energy intensity index  89 Table 3.4. Link between national planning RISE score, energy intensity, and energy price  43 Table A.8. Relationship between energy prices, regulatory policies, and energy efficiency index  90 Table 3.5. Link between the RISE score of energy efficiency entities, energy intensity, and energy price  44 Table A.9. Relationship between energy prices, regulatory policies, and economic activity index  91 Table 3.6. Link between financing mechanism RISE scores, energy intensity, and energy price  44 Table A.10. The role of national planning (RISE score)  92 Table A.1. Data sources  54 Table A.11. The role of entities responsible for promoting energy efficiency (RISE score)  93 Table A.2. Selected countries and regions  55 Table A.12. The role of financing mechanisms Table A.3. Effect of removing energy subsidies (RISE score)  94 on energy intensity  83 Boxes Box 1.1. Mexico’s voluntary agreements Box 1.3. Argentina’s energy efficiency program for energy efficiency  24 funded by the Global Environment Facility  26 Box 1.2. Mexico’s EE Labelling and MEPS  25 Box 1.4. Mexico’s PRESEMEH program  27 viii Executive The rest of the world significantly outpaced the Summary region’s modest improvements Realizing the Potential of Energy Efficiency in Latin America and the Caribbean in energy intensity Energy efficiency is a critical but underutilized resource through- out Latin America and the Caribbean. Investing in it should be The region’s economies have become less energy intensive an integral part of every country’s energy policy. Countries that in recent decades. Since 2005, there has been a decoupling have consistently invested in energy efficiency over the last de- of energy variables from changes in GDP, with GDP grow- cades have seen lower consumer costs, a more reliable energy ing faster than energy supply and consumption from 2005 supply, less volatile energy prices, and lower emissions of green- to 2019 (figure ES.1). These trends indicate that the region’s house gases (GHG) (IEA 2022). Energy efficiency improvements economies have become less energy intensive. In fact, the are necessary for sustainable economic development because region’s energy intensity is lower than that of all other world they help reduce the economic and environmental costs of pro- regions except the European Union. ducing goods and services and shrink their overall carbon foot- print. As noted by the International Energy Agency’s (IEA) Glob- However, improvements in energy intensity in the region have al Commission for Urgent Action on Energy Efficiency, success stagnated, while those of the rest of the world have acceler- depends on whole-of-government responses to align actions ated. In recent decades, the United States and the European across economic sectors, engage public support and participa- Union reduced their energy intensity by around 2 percent and tion, and dismantle barriers (IEA 2020a). 1.8 percent per year, respectively.1 In contrast, the indicator for LAC oscillates at practically constant values in the same On average, improvements in Latin America and the Carib- period, with an average annual reduction of 0.5 percent, far bean (LAC) have been modest, with energy demand project- below the 4 percent per year between 2020 and 2030 need- ed to increase significantly in coming decades. Though some ed to fulfill net-zero GHG reduction objectives, according to countries have taken measures to improve energy efficien- the IEA’s World Energy Outlook 2020 (IEA 2021a). cy, those measures have been unevenly and consistently implemented. Energy efficiency remains underutilized in the Low energy intensity in the region does not necessarily imply region because of technical, financial, and policy barriers. high energy efficiency. It also reflects the underutilization of household appliances, a lack of affordable residential energy ser- Successful implementation of energy efficiency programs re- vices, and less use of technology. The use of energy in industrial quires long-term and holistic engagements on all these fronts. production is also less intensive, as LAC’s economies are not as But government actions in the region have lacked sustainabili- industrialized as those of other developing regions (IDB 2019). ty and have been unable to attract private financing to invest in energy efficiency initiatives or reduce the risks associated with Sectoral energy intensity indicators have varied widely from projects (Loureiro et al. 2021). There is now an urgent need to country to country. The variations have evened out at the relaunch the energy efficiency agenda across the continent in regional level, with aggregate values across the residential, the context of post-COVID recovery, the challenges of climate manufacturing, and services sectors showing relatively small change mitigation, and high energy prices. changes. Colombia and Peru show consistent energy intensity reductions in the three sectors. Chile, Mexico, and Argentina This report assesses the current state of energy efficiency display reductions in two of the three sectors, while Brazil and policies and measures in the region, identifies key regional other countries of the region have increased their energy inten- challenges and drivers for improved energy efficiency, and sity overall. Countries with a higher share of energy-intensive proposes ways forward. Starting with a high-level review industries show higher energy intensity. Greater access to ap- of how the policies throughout the region have evolved pliances in the residential sector also results in greater energy over the past two decades, the report then pinpoints key intensity (unless those appliances are energy efficient). regional policy levers that drive improvements. Specifical- ly, the report applies the Fisher decomposition method Under favorable economic conditions, energy demand in (Boyd and Roop 2004) to disaggregate broad trends in LAC could increase significantly with greater affordability and energy trends into discrete efficiency improvements and penetration of appliances. If that happens, energy intensity changes in economic activity. It then analyzes which policy will stabilize or even increase. For example, under the right instruments are most likely to reduce the energy intensi- economic conditions, households would be able to purchase ty of key sectors. Finally, it provides recommendations on additional appliances, which, absent energy efficiency poli- Executive Summary policy instruments to support energy efficiency improve- cies, would push up the residential sector’s energy intensity. ments in the region. 1 World Bank data, https://datos.bancomundial.org/indicador/EG.EGY.PRIM.PP.KD Figure ES.1 GDP (PPP$ 2017) ix Growth of primary energy supply, final energy Primary Energy Supply (ktoe) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean consumption, and GDP in LAC, 2000–19 Final Energy Consumption (ktoe) 170 160 Change from Base Year (2000) 150 140 130 120 110 100 90 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: WBG, based on Latin-American Energy Organization (OLADE) and World Bank database. Note: Total energy supply comprises final energy consumption, non-energy consumption, and consumption and losses of the energy sector. The region needs to overcome adequate prioritization in a very energy-intensive economy many barriers to greater energy might erroneously focus on tackling insulation of the exist- ing building stock as the first measure to be implemented. efficiency Actually, other steps (such as improving labeling and stan- dards for appliances and vehicles) can be more immediate, Commonly acknowledged barriers to energy efficiency are effective, and efficient. low energy prices, low exposure to energy supply shocks, legacy assets operating past their design life, and obstacles Many of the programs developed in the region have been to the uptake of technology. The latter might include im- based on international grants or technical assistance and pediments to technology transfers, such as limits on trade were not sustained over time despite their good results with certain countries or financial barriers that limit the (ECLAC 2021). International cooperation primarily seeks government and the private sector’s capacity to invest in to achieve a quick knowledge transfer and seldom yields updated technologies. In addition, lack of coordination and programs that are self-sustaining without an enabling en- prioritization of policy measures across sectors, lack of pol- vironment. Although most of these programs include a gov- icy stability, and lack of capabilities to implement and con- ernment-financed capacity-building component, local gov- trol policy implementation are also important barriers. ernments do not always provide continuity and scale up the implemented measures. Energy efficiency policies, which are notoriously difficult to approve and implement, are effective if they are well-tar- Lack of financing has also hampered advances in energy effi- Executive Summary geted to specific sectors and sensitive to the local context. ciency. Based on Latinobarometer’s 2018 survey, 71 percent However, their implementation is often limited by several of households in the region would be willing to spend money factors, including information gaps, the high up-front cost on appliances that allow them to lower their electricity bill, of investments, and the diffuse nature of benefits, low visi- but 22 percent of them said that they did not have the re- bility, and challenges in measuring results. For example, in- sources to cover the cost of the appliances (IDB 2019). The lack of information on the impacts of implemented policies Countries in the region have taken different approaches to x and measures has delayed the uptake of energy efficiency mea- energy efficiency planning. Some have approved a national sures in the region. For example, lack of awareness was a signif- plan or strategy.2 In these, the policies and measures have Realizing the Potential of Energy Efficiency in Latin America and the Caribbean icant impediment for energy efficiency policies in the Argentin- focused on (i) energy labeling systems and minimum ener- ean electricity sector (Recalde and Guzowski 2012). Improved gy performance standards (MEPS); (ii) national energy effi- information on impact would provide an opportunity to eval- ciency planning; (iii) incentives and mandates for the private uate and disseminate results and share best practices among sector; (iv) building codes; (v) incentives and mandates for countries, enabling the industrial and commercial sectors to the public sector; and (vi) financing mechanisms. implement successful measures without public support. Improvements in energy efficiency regulations related to planning and access to financing have the potential to re- Energy demand policies duce energy intensity and improve sectoral energy efficiency. and regulatory reforms can Countries with weak regulatory policies could significantly improve their energy efficiency if they implemented better significantly improve the policies and raised their score on the Regulatory Indicators region’s energy efficiency for Sustainable Energy (RISE) to that of the leading country (Mexico in 2019, the last year for which updated indicator values were obtained). Over the past four decades, LAC countries have implement- ed various energy efficiency policies and measures. Eleven For example, Argentina, Dominican Republic, Guatemala, and countries have adopted national legislation; six others Peru—where energy efficiency policies are suboptimal—have have a bill under discussion. Since 1985, twenty-two LAC the potential to achieve additional reductions of 6 to 8 per- countries have implemented almost 300 measures and cent in their energy intensity index and additional improve- programs, with the implementation rate rising after 2007. ments of 6.5 to 9.5 percent in their sectoral energy efficiency. Using information from ECLAC’s Base of Energy Efficiency Improving RISE scores to Mexico’s levels in countries with Indicators (ECLAC 2021) and the IEA’s policies database, the somewhat better regulatory policies, such as Brazil, Chile, World Bank compiled a database of policies and programs Colombia, and Ecuador, could produce additional declines of in LAC countries. That information was supplemented with 2 to 3.5 percent in their energy intensity index and additional additional measures from official country websites to cover improvements of 2.5 percent to 4 percent in their sectoral countries that do not appear in the IEA and BIEE databases energy efficiency. The average energy intensity and sectoral and those that appear but whose information is incomplete. gains in energy efficiency resulting from strengthening regu- It should be borne in mind, however, that well-structured latory policies in the region are estimated at 2.3 percent (fig- and comparable information on implementation and effec- ure ES.2) and 2.7 percent and (figure ES.3), respectively. tiveness is hard to find, and information presented at the national level may mask significant regional discrepancies. 2 In Argentina, Guatemala, and the Dominican Republic the plans are still under parliamentary discussion. Executive Summary Figure ES.2 xi Reductions in energy intensity Δ Energy Intensity Realizing the Potential of Energy Efficiency in Latin America and the Caribbean implied by better RISE scores, Additional Δ RISE by country 20% 10% 11,3% -3,5% 0,4% Percentage Change in Energy intensity 0% -3,4% -28,6% -24,7% -21,6% -20,6% -19,8% -11,3% -10,3% -4,4% -6,6% -2,1% -10% -2,3% 0,0% -7,3% -20% -7,9% -6,9% -3,2% -2,7% -6,0% -30% -40% Ecuador Dominican Republic Colombia Guatemala Peru Chile Argentina Mexico Brazil Costa Rica LAC Source: WBG’ estimates based on counterfactual regression analysis using OLADE, UN data, Penn World Tables, IEA. Note: Yellow bars show the baseline changes in energy intensity of ten largest LAC countries between the first year Executive Summary and the last year of available data. Blue bars show additional energy intensity reductions if each country had raised their RISE score to that of the leading country for 2019 RISE EE scores. RISE scores have been periodically updated, and we used the version available at the time of the analysis. xii Figure ES.3 Δ Energy Intensity Improvements in energy efficiency Additional Δ RISE Realizing the Potential of Energy Efficiency in Latin America and the Caribbean resulting from better RISE scores, by country 10% 5% 4,8% 1,5% 0,6% -2,2% 0% -3,9% -29,4% -24,6% -19,9% -15,4% -14,7% -8,1% -6,0% -3,6% -2,5% -5% -2,7% Percentage Change in Energy intensity -8,6% 0,0% -10% -15% -9,5% -3,8% -20% -3,2% -25% -7,4% -30% -6,6% -35% -40% Ecuador Dominican Republic Guatemala Colombia Chile Peru Mexico Argentina Costa Rica Brazil LAC Source: WBG’ estimates based on counterfactual regression analysis using OLADE, UN data, Penn World Tables, IEA. Note: Yellow bars show the baseline changes in energy intensity, due exclusively to energy efficiency improvements, Executive Summary of ten largest LAC countries between the first year and the last year of available data. Blue bars show additional energy intensity reductions if each country had raised their RISE score to that of the leading country for 2019 RISE EE scores. RISE scores have been periodically updated, and we used the version available at the time of the analysis. Improvements in energy efficiency resulting from phas- ciency measures would enable the private sector to under- xiii ing out energy subsidies and reforming regulatory regimes stand options and implement successful measures without can bring substantial energy savings. Counterfactual sim- the need for public support. Realizing the Potential of Energy Efficiency in Latin America and the Caribbean ulations based on econometric analysis show consider- able combined savings for the countries of the region, with Wider access to financing is effective—but only under cer- smaller and poorer economies of Central America and the tain conditions. For example, energy efficiency investments Caribbean experiencing the largest relative savings. For ex- require an adequate enabling environment characterized by ample, energy reforms could save up to an estimated 13.38 clear information, trusted parties that can provide needed percent of total energy consumption in Haiti, 9.93 percent services, and essential equipment and material, all of which in El Salvador, and 8.5 percent in Honduras. Argentina and can be hard to come by. To ensure their presence, national Bolivia would also see significant energy savings from en- governments must implement clear long-term plans and ergy reforms (9.73 percent and 9.45 percent, respectively). streamlined procedures to support investments (IEA 2022). These savings would be even larger when accounting for en- ergy security and environmental benefits. LAC countries must continue to reduce energy subsidies while protecting vulnerable populations. Energy subsidies directly affect how energy is used and the choice to acquire Steps to improve energy efficient technologies. Although the currently high energy efficiency in the region prices make subsidy reduction difficult, they also offer an opportunity to establish frameworks that will allow subsi- dies to be phased out automatically as soon as prices drop. Exploiting synergies between sustainable programs, tech- nology transfer, financing, and adequate energy pricing will Finally, the importance of coordinated government actions be essential to realize the region’s energy efficiency poten- and a comprehensive and integrated approach to address tial. Based on the IEA’s bottom-up scenarios (IEA 2021a), the multiple challenges to energy efficiency implementation considerable improvements in the region’s energy intensity cannot be understated. Success with energy efficiency de- are possible, from 1.1 percent to 2.3 percent annual reduc- pends on the actions of policymakers responsible for ener- tions through 2040. These scenarios mark a clear upside gy, industry, housing, transport, and finance, as well as the compared with the region’s paltry improvement trend over equivalent actors at the subnational and local levels. the past 20 years (around 0.5 percent per year). There is an urgent need for sustainable and well-financed pol- icies and programs that focus on efficient technology integra- tion complemented by a just phase-out of energy subsidies. Energy efficiency policies in the region have shown mixed results depending on the country and sector, with more policies and measures often not translating into consistent improvements in energy efficiency and final energy intensity. National and local governments should take advantage of ex- isting programs and improved capacities to set up long-term programs and scale up the measures already implemented. Their focus should be on enhancing technology transfer, im- proving the sustainability of programs, improving access to financing, and reducing energy subsidies while ensuring that vulnerable social groups are protected. Incentivizing the integration of the most efficient technol- ogies in various sectors can be done by establishing spe- cial tax schemes to promote efficient technology uptake (either in the form of tariffs or taxes that are removed or lowered for efficient technology or, conversely, in the form of increased tariffs or taxes for inefficient technology). Other drivers include better access to information on the impact of improved technologies and offers of concessional financ- Executive Summary ing for select investments. Particularly for the industrial and commercial sectors, programs should be designed so that their sustainability and scaling are backed by co-financing from the private sector. Improved frameworks for monitor- ing and reporting of the results and impacts of energy effi- 1 14 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean The region’s record on energy efficiency Why should the region’s countries be concerned about energy efficiency? Page 16  Main barriers to energy efficiency Page 17  Chapter 1. The region’s record on energy efficiency Is the region energy efficient? Page 17  Institutional context, regulation, and policies Page 21  Energy efficiency potential in the region Page 32  Energy efficiency improvements in Latin America and the Carib- The report begins with a review of how energy efficiency pol- 15 bean (LAC) are not keeping pace with growth in energy demand, icies in the region have evolved over the last two decades. which is projected to grow substantially in coming decades. It then proceeds to identify key policy levers that drive im- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Though some countries have taken measures to improve en- provements in energy efficiency across the region. Specif- ergy efficiency, those measures have been unevenly and con- ically, the report applies the Fisher decomposition meth- sistently implemented.3 There is an urgent need to revisit the od (Boyd and Roop 2004) to disaggregate broad trends in energy efficiency agenda in the continent and relaunch it in the energy intensity into discrete efficiency improvements and current context of post-COVID recovery and high energy prices. changes in economic activity. It subsequently analyzes which policy instruments would have the largest effects in The objective of this report is to take stock of current en- reducing sectoral energy intensities. The report concludes ergy efficiency policies and measures in the region, identify with recommendations for policy instruments to support key regional challenges and drivers for improvement, and energy efficiency improvements in the region. propose some ways forward. 3 LAC includes Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay and the Bolivarian Republic of Venezuela. Chapter 1. The region’s record on energy efficiency 16 Why should their overall carbon footprint. As noted by the International Energy Agency’s (IEA) Global Commission for Urgent Action the region’s on Energy Efficiency, success depends on whole-of-govern- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean ment responses to align actions across economic sectors, engage public support and participation, and dismantle countries be barriers (IEA 2020a). concerned However, despite its great potential, energy efficiency remains underutilized in the region because of technical, financial, and policy barriers. Successful implementation of energy efficien- about energy cy programs requires long-term and holistic engagements on all these fronts. But government actions in the region have efficiency? lacked sustainability and have been unable to attract private financing to invest in energy efficiency initiatives or reduce the risks associated with projects (Loureiro et al. 2021).4 Investing in energy efficiency should be an integral part of At the country and regional level, the evolution of energy ef- every country’s energy policy. Countries that have consis- ficiency can be inferred from trends in total energy supply, tently invested in energy efficiency over the last decades final energy consumption, and GDP.5 Figure 1.1 shows that have seen lower consumer costs, a more reliable energy primary energy supply and total final energy consumption supply, less volatile energy prices, and lower emissions of grow at similar rates.6 Since 2005, there has been a decou- greenhouse gases (GHG) (IEA 2022). Energy efficiency im- pling of energy variables from changes in GDP, with GDP provements are necessary for sustainable economic devel- growing faster than energy supply and consumption from opment because they help reduce the economic and envi- 2005 to 2019. These trends indicate that the region’s econ- ronmental costs of producing goods and services and shrink omies have become more energy efficient. 4 Owing mainly to a lack of commercial and financial sector experience with EE projects, it is difficult to obtain statistical data on the actual energy and cost savings achieved by implemented energy efficiency projects. There is also a lack of statistics on sector default rates (Loureiro et al. 2021). 5 Primary energy supply includes final energy consumption, non-energy consumption, and consumption and losses of the transformation sector (or energy sector). 6 During this period some LAC countries experienced changes in their power generation mix that affected energy supply but not necessarily final energy consumption. However, in the regional aggregated values, these variations are compensated for, such that both variables show a similar evolution. Chapter 1. The region’s record on energy efficiency Figure 1.1 GDP (PPP$ 2017) 17 Growth of primary energy supply, final energy Primary Energy Supply (ktoe) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean consumption, and GDP in LAC, 2000–19 Final Energy Consumption (ktoe) 170 Change from Base Year (2000) 160 150 140 130 120 110 100 90 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: WBG, based on Latin-American Energy Organization (OLADE) and World Bank database. Note: Total energy supply comprises final energy consumption, non-energy consumption, and consumption and losses of the energy sector. Owing to general equilibrium effects and structural changes, also important barriers. For example, inadequate prioritization energy efficiency is difficult to measure and track across an in a very energy-intensive economy might erroneously focus entire economy. It is usually measured as improvements in on tackling insulation of the existing building stock as the first energy consumption for specific processes or services. Giv- measure to be implemented. Actually, other steps (such as im- en these measurement problems, trends can be gauged by proving labeling and standards for appliances and vehicles) can changes in both primary and final energy intensity, defined as be more immediate, effective, and efficient. the ratio of output to energy. While energy intensity is a crude indicator of energy efficiency, it can provide some macro-level guidance on how energy efficiency has developed over time Is the region energy efficient? by controlling for observed structural changes. Energy intensity is lower in LAC than in all other world regions except the European Union (figure 1.2), but improvements in the Main barriers to energy region have stalled, while the rest of the world has reduced its efficiency energy intensity significantly. The European Union, the coun- tries of the Organization for Economic Co-operation and De- velopment (OECD), the United States, and the world as a whole Commonly acknowledged barriers to energy efficiency are low have shown much more progress in lowering their energy in- Chapter 1. The region’s record on energy efficiency energy prices, low exposure to energy supply shocks, legacy tensity than have the LAC countries. In the 2000–15 period, the assets operating past their design life, and obstacles to the United States reduced its energy intensity by around 2 percent uptake of technology. The latter might include impediments per year, and the European Union and the OECD by around 1.8 to technology transfers, such as limits on trade with certain percent per year.7 By contrast, the indicator for LAC oscillates at countries or financial barriers that limit the government and practically constant values in the period, with an average annu- the private sector’s capacity to invest in updated technolo- al reduction of 0.5 percent. According to the IEA’s World Energy gies. In addition, lack of coordination and prioritization of policy Outlook 2020 (IEA 2021a), this reduction is considerably below measures across sectors, lack of policy stability, and lack of ca- the 4 percent per year between 2020 and 2030 needed to fulfill pabilities to implement and control policy implementation are net-zero GHG reduction objectives (IEA 2021a). 7 World Bank Data, https://datos.bancomundial.org/indicador/EG.EGY.PRIM.PP.KD 18 Figure 1.2 LAC United States Energy-intensity of primary energy in European Union World Realizing the Potential of Energy Efficiency in Latin America and the Caribbean different regions, 2000–15 OECD 8 7,5 7 MJ/GDP (PPP 2011 US$) 6,5 6 5,5 5 4,5 4 3,5 3 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: WBG, based on World Bank Database, World Development Indicators. Low energy intensity in the region does not necessarily imply Final energy intensity is dropping slowly in LAC, with an high energy efficiency. It also reflects the underutilization of average annual change of around –0.4 percent (figure 1.4). household appliances, a lack of affordable residential energy Although final energy intensity changes at a rate similar services, and less use of technology. The use of energy in in- to that of final energy supply, using the energy intensity dustrial production is also less intensive, as LAC’s economies indicator based on final consumption allows us to remove are not as industrialized as those of other developing regions the impact of variations in the transformation sector8 and (IDB 2019). With energy demand projected to increase in the those caused by nonenergy consumption, leaving only en- region owing to greater affordability and penetration of ap- ergy consumption from the residential, industrial, services, pliances (figure 1.3). If that happens, energy intensity is likely transportation, agriculture and fishing, and construction to stabilize or even increase. This is so because, under favor- sectors. Between 2000 and 2019, the cumulative reduction able economic conditions, households would be able to pur- was around 8 percent, with no consistent annual reduction. chase additional number of appliances which, absent energy efficiency policies targeted to the sector, would mean an in- Trends in final energy intensity in the region show an irregular crease in the residential sector’s energy intensity. decreasing trend between 2000 and 2019 (figure 1.4). Although countries present different realities, the average is shaped chief- Though most countries in the region have energy efficien- ly by changes in the region’s economic composition (marked by cy policies, they appear to have little impact on the energy a reduction in the industrial share of GDP) and an average im- Chapter 1. The region’s record on energy efficiency intensity index. However, their effects are difficult to isolate provement in the energy efficiency of the region’s residential and owing to the combined effects of other variables, such as services sectors. In addition, irregularities can be largely explained access to equipment and technology, economic structures, by the effect of economic instability in the region over the years the power generation mix, economic crises, the climate, studied and changes in the population’s access to energy-con- habits of consumption, and general economic development. suming appliances (which increased per-capita energy use). 8 The “transformation sector” is understood as activities that result in the transformation of primary energy forms into secondary energy forms (for example, a coal power plant transforming energy in coal into energy in an electric current). This concept s closely linked to that of Final Energy , which is the primary or secondary energy that is directly used by socio- economic sectors, and does not include losses due to the intermediate processes (transformation, transmission, transport, distribution and storage losses). Figure 1.3 Central and South America Announced Pledges Scenario 19 Historical and projected total final energy Stated Policies Scenario Sustainable Development Scenario Realizing the Potential of Energy Efficiency in Latin America and the Caribbean consumption in the region, 2010–50 35 30 25 Exajoules 20 15 10 5 0 2010 2019 2020 2030 2050 Source: WBG, based on IEA 2021a and World Bank Indicators. Note: The scenarios are described in IEA’s World Energy Outlook 2020 (IEA 2021a). The scenarios are discussed in greater detail in the last section of this chapter of the report. Figure 1.4 Annual percentage change Evolution of final energy intensity in LAC, Final energy intensity (MJ/ US$ PPP 2017) 2000–19 (MJ/GDP, PPP 2017 US$) 2% 2,8 1,7% 2,7 1% Final energy intensity (MJ/ US$ 1,0% Annual percentage change 0,3% 0,2% 0,1% 0,6% 2,6 0% PPP 2017) -0,6% -2,1% -1,3% -0,7% -2,4% -1,0% 2,5 -1,0% -0,4% -0,5% -0,5% -0,5% -0,6% -0,6% -1% Chapter 1. The region’s record on energy efficiency 2,4 -2% 2,3 -3% 2,2 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: World Bank, based on OLADE and World Bank database. The evolution of energy intensity shows wide differences across tions have evened out at the regional level, with aggre- 20 LAC countries. Colombia, Peru, and Chile show the fastest re- gate values across residential, manufacturing, and ser- duction in energy intensity (figure 1.5). However, all three showed vices sectors showing relatively small changes. Colombia Realizing the Potential of Energy Efficiency in Latin America and the Caribbean considerably greater improvement in the 2000–10 period than and Peru show consistent energy intensity reductions in in 2010–19. Argentina and Brazil saw a change in the growth the three sectors. Chile, Mexico, and Argentina display trend between the 2000–10 and 2010–19 periods, initially low- reductions in two of the three sectors, while Brazil and ering energy intensity and later increasing it. Mexico, in contrast, other countries of the region have increased their ener- initially showed an upward trend that was later reversed, leading gy intensity overall. Countries with a higher share of en- to consistent reductions. The “Other LAC countries” category ergy-intensive industries show higher energy intensity. exhibited more reduction in the period 2010–19 period. Within Greater access to appliances in the residential sector this group, Costa Rica, the Dominican Republic, Guyana, Hondu- also results in greater energy intensity (unless those ap- ras, Nicaragua, Panama, Paraguay, and Suriname improved their pliances are energy efficient). energy intensity between 2000 and 2019. In the Dominican Republic, Guyana, Panama, and Suriname, the improvement Reductions in energy intensity for the residential and man- exceeded 40 percent; in Honduras, Nicaragua, and Panama, 20 ufacturing sectors in Colombia and Peru could be attribut- percent; and in Paraguay and Costa Rica, 10 percent. ed to improvements in energy efficiency. These improve- ments could potentially be driven by increased uptake of Sectoral energy intensity indicators have varied widely more efficient technology, resulting in lower intensity per from country to country (figure 1.6). However, the varia- capita and as a share of value added, respectively. Figure 1.5 Brazil Argentina Ratio of final energy intensity to GDP in major Chile Peru LAC countries, 2000–19 Other Mexico Latin America & Caribbean Colombia 3,5 3,3 3,1 GDP (MJ/GDP (PPP 2017 US$) 2,9 2,7 2,5 2,3 Chapter 1. The region’s record on energy efficiency 2,1 1,9 1,7 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: WBG, based on OLADE and World Bank database, change in energy intensity over time. Figure 1.6 Residential per capita energy Services energy intensity/value 21 intensity (GJ per capita) added (MJ/constant 2010 US$) Changes in final energy intensity Manufacturing energy intensity/ Population growth Realizing the Potential of Energy Efficiency in Latin America and the Caribbean by sector, 2000–19 value added (MJ/constant 2010 US$) (2000 to 2019) 40% Percentage change in final energy intensity 36% 30% 20% 18% 16% 16% 10% 12% 2% 1% 1% 0% -1% -1% -3% -3% -5% -14% -7% -10% -17% -17% -18% -20% -21% -24% -20% -28% -28% -38% -30% -40% Colombia Peru Chile Other Mexico Argentina Brazil Latin America countries LAC & Caribbean Source: WBG, based on OLADE and World Bank database. Institutional context, regulation, do not appear in the IEA and BIEE databases and those that and policies appear but whose information is incomplete. It should be borne in mind, however, that well-structured and compara- ble information on implementation and effectiveness is hard Evolution of energy efficiency policy to find, and information presented at the national level may implementation in the region mask significant regional discrepancies. Countries in the region have taken different approaches to Over the past four decades, LAC countries have implemented energy efficiency planning. Some have approved a national various energy efficiency policies and measures. Eleven coun- plan or strategy.10 In these, the policies and measures have tries have adopted national legislation; six others have a bill focused on (i) energy labeling systems and minimum ener- under discussion. Since 1985, twenty-two LAC countries have gy performance standards (MEPS); (ii) national planning; (iii) implemented almost 300 measures and programs, with the incentives and mandates for the private sector; (iv) building implementation rate rising after 2007. Using the aggregated codes; (v) incentives and mandates for the public sector; indicators proposed in the document “Indicadores de Políticas and (vi) financing mechanisms. Chapter 1. The region’s record on energy efficiency Públicas en Materia de Eficiencia Energética en América Latina y el Caribe” (ECLAC-GTZ, 2010), the information from ECLAC’s Energy efficiency policies and actions in Colombia, Brazil, Mex- Base of Energy Efficiency Indicators (BIEE)9 (ECLAC 2021) and ico, Argentina, Chile, and Peru have shown increased momen- the IEA’s policies database (IEA 2020b), the World Bank com- tum but wide differences (figure 1.7). Some countries have piled a database of policies and programs in LAC countries. shown regular actions and efforts since 1985 (Brazil) and 1993 This information was supplemented with additional mea- (Mexico) while others exhibit sporadic activity. Since 2007 sures from official country websites to cover countries that growth in adopted measures has been strong across the region. 9 https://biee-cepal.enerdata.net/en/ 10 In Argentina, Guatemala, and the Dominican Republic the plans are still under parliamentary discussion. 22 Figure 1.7 Rest of South America Mexico Number of measures per year Rest of Central America Colombia Realizing the Potential of Energy Efficiency in Latin America and the Caribbean per country Caribbean Chile Peru Brazil Argentina 100 80 Number of measures 60 40 20 0 1985-1989 1990-1994 1995-1999 2000-2004 2005-2009 2010-2014 2015-2019 Source: WBG, based on the aggregation of policy databases (BIEE, ECLAC, IEA). Note: Only measures for which a starting year was documented are accounted for. Chapter 1. The region’s record on energy efficiency There are also wide variations by sector in the types and all sectors, the latter implying the importance of cross-cut- quantity of policies implemented 1.8). Energy labeling sys- ting approaches to address the multiple challenges of im- tems and MEPS are the measures most frequently imple- plementation. “Energy labeling and MEPS measures were mented, followed by national energy efficiency plans, and the most common across all sectors, especially in residen- incentives and mandates. There are 111 measures for the tial and services sectors, due mainly to the large number of residential sector; 84 measures for the services, commer- labeling measures implemented for household appliances. cial, and public sectors; 48 for the industrial sector; and 86 Figure A.27 in the Appendix shows energy efficiency policies identified as “cross-cutting” (affecting three or more sec- and regulations per country, across the LAC region, and pro- tors). Most measures were aimed at households or across vides further detail on this regard. Figure 1.8 Incentives & mandates: Industrial 23 Building energy codes Number of measures by sector, and Commercial end users Energy labeling systems and MEPS National energy efficiency planning Realizing the Potential of Energy Efficiency in Latin America and the Caribbean country, and type of measure, 1985-2019 Entities Incentives & mandates: Public sector Financing mechanisms Information, education, and other 7 92 9 3 Households 9 12 7 8 2 44 4 Cross-cutting 5 44 22 7 8 1 Services 21 3 24 Industrial 0 20 40 60 80 100 120 Number of measures Source: WBG, based on aggregation of policy databases (BIEE, ECLAC, IEA). Note: “Entities” refers to government, quasi-government or private body that can implement certain types of Energy Efficiency policies (e.g., establishing and enforcing regulations) or execute other functions (e.g., delivery of Energy Efficiency goods and services), such as a national energy efficiency agency. Energy efficiency policies are notoriously difficult to approve available water for hydroelectric generation in Chile, where and implement. They are effective if they are well-targeted hydroelectric generation represented 40 percent of elec- to specific sectors and take into account the local context. tricity generation in the country. The drought threatened Their implementation is often limited by several factors, in- deep electricity shortages in the country. Despite this chal- cluding information gaps, high up-front cost of investments lenge, the country was able to avoid electricity interruptions Chapter 1. The region’s record on energy efficiency and diffuse nature of benefits, low visibility, and challenges by implementing a comprehensive short-term package of in measuring results. quick measures targeting low-hanging fruit. The measures implemented included private-public information cam- The implementation of energy efficiency measures needs paigns on the importance of saving energy and how to do it; to account for policy complexities across sectors. In 2007 a program to distribute energy-efficient lighting; and short- and 2008 a drought caused a 30 percent reduction in the term rationing.11 11 Sources for this paragraph and the next: https://www.cne.cl/wp-content/uploads/2016/07/AnuarioCNE2015_vFinal- Castellano.pdf; https://web.archive.org/web/20190507004819/http://antigua.cne.cl/noticias/energia/electricidad/2- gobierno-y-empresas-electricas-lanzan-campana-de-energia-ahorra-ahora; https://www.emol.com/noticias/ nacional/2008/08/03/315898/gobierno-lanza-nueva-campana-para-fomentar-ahorro-y-buen-uso-de-la-energia.html. Once the emergency had passed, the government tary agreements for energy efficiency (box 1.1). Two others 24 sought to build on the success of the public information are described in box 1.2. campaigns and maintain the gains by putting in place Realizing the Potential of Energy Efficiency in Latin America and the Caribbean long‑term financing for energy efficiency investments Measures aimed at providing financial incentives were the and offering financial incentives for conservation actions most often discontinued by far; these were followed by mea- that built on the information campaigns to stimulate sures designed to support energy efficiency entities, capac- sustained energy savings. ity building, energy efficiency programs, and information on energy efficiency (figure 1.9). It is worth mentioning that “EE The full and consistent implementation of energy efficien- entities” refers to government, quasi-government or private cy polices in the region has also been hampered by lack of bodies that can implement certain types of Energy Efficiency policy stability over time. As constraints to effective imple- policies (e.g., establishing and enforcing regulations) or exe- mentation of policies tend to be specific to each country, no cute other functions (e.g., delivery of Energy Efficiency goods single barrier can be easily addressed across the region, and and services), such as a national energy efficiency agency or there is no one-size-fits-all solution. An example of a long- decentralized entity, a special purpose public fund, a unit term targeted policy measure is Mexico’s program of volun- mandated to promote EE under a ministry, etc. Box 1.1 Mexico’s voluntary agreements for energy efficiency Mexico’s program of voluntary agreements (VAs) for energy efficiency, mandated under the country’s energy transition law, was launched in 2019. VAs are implemented by the Secretariat of Energy, through the Commission for the Efficient Use of Energy (CONUEE), a decentralized national-level energy efficiency entity created by law in 2008. CONUEE’s main objective is to promote energy efficiency and act as a technical body for the sustainable use of energy. Chapter 1. The region’s record on energy efficiency CONUEE signs VAs with companies that consume significant amounts of energy consumption. Under the agree- ments, the companies commit to cutting the energy intensity of their activities and to measuring the impacts of their efforts. The participants must specify the goal that they pledge to reach over the term of the agreement; CONUEE provides technical support to help them achieve that goal. The technical assistance provided includes support in setting goals; a methodology for energy audits (both for setting a baseline and for verification of goals); a technical and cost-benefit analysis of proposed energy efficiency actions; and potentially supporting an ISO 50.001 certifica- tion. Starting from a single agreement in 2019, CONUEE has signed VAs with 14 of the largest companies in Mexico, including Nestlé, Grupo Bimbo, Audi México, Bio Pappel Scribe, Cementos Fortaleza, Flex, Nemak, Vitro, and Ternium. Source: Mexican Government, Energy Secretariat (SENER), CONUEE, n.d. 25 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Box 1.2 Mexico’s EE Labelling and MEPS An energy labeling system and MEPS in Mexico are examples of an energy efficiency policy sustained over time. Start- ed in 1995, the Mexican Official Energy Efficiency Standards (NOM-ENER) are mandatory. They include the technical specifications to be met by the equipment (minimum efficiency values or maximum energy consumption), the test methods, the conformity assessment method, and the respective energy efficiency label. CONUEE leads the stan- dardization process and includes a Regulatory Impact Analysis, containing the estimated energy savings associated with each standard. 33 NOM-ENERs have been published, covering equipment from the residential, commercial, and industrial sectors, and the updating process for these standards is continuous. Source: Mexican Government, Energy Secretariat (SENER), CONUEE, n.d. Figure 1.9 Discontinued Discontinued and ongoing measures by sector, Ongoing country, and type of measure, 1985-2019 25 20 Number of measures 15 Chapter 1. The region’s record on energy efficiency 10 5 0 Finantial Fiscal Information Mandatory Mandatory Program Regulations Other information standards Source: WB team, based on ECLAC (2021) and review of public webpages of LAC countries. Note: “Other” includes support for energy efficiency entities, awareness raising, and capacity building. Many of the programs developed in the region have been governments do not always provide continuity and scale up 26 based on international grants or technical assistance and the implemented measures. Programs dependent on public were not sustained over time despite their good results budgets are often intermittent owing to economic and so- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean (ECLAC 2021). International cooperation primarily seeks to cio-political crises and lack of long-term planning of energy achieve a quick knowledge transfer and seldom yields pro- policies. Box 1.4 presents an example of a well-designed pro- grams that are self-sustaining without an enabling environ- gram that has faced implementation challenges related to ment (box 1.3). Although most of these programs include a an inflexible regulatory framework and insufficient capacity government-financed capacity-building component, local at the municipal level. Box 1.3 Argentina’s energy efficiency program funded by the Global Environment Facility The energy efficiency program in Argentina funded by the Global Environment Facility (GEF) is an example of a well-targeted program financed by an international grant that achieved some success during implementation but could not maintain its momentum after grant financing was exhausted. The program’s objective was to increase energy efficiency and reduce emissions of greenhouse gases. This was to be achieved by providing targeted financial resources aimed at removing the regulatory, financing, and information barriers to activities and investments in en- ergy efficiency and energy conservation. The program helped Argentina’s national government improve its capabilities to set energy efficiency standards, con- Chapter 1. The region’s record on energy efficiency duct labeling, prepare regulations, and provide SMEs with access to finance for investments in energy efficiency. It resulted in 20 sets of norms and standards, issued 19 labels for appliances, initiated an energy efficiency law project, and funded energy efficiency investments for 13 SMEs. Nonetheless, the SME financing facility ceased operations not long after the end of the GEF project for lack of addi- tional commitments from the government and a stable local implementing counterpart. The program might have been more sustainable had it focused on building technical capacity in the implementing partner and in helping the partner build a solid institutional structure. Source: IBRD Extranet-Projects and Operations-Project detail-P090119. 27 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Box 1.4 Mexico’s PRESEMEH program The ongoing PRESEMEH program in Mexico is an example of a well-designed program that has faced implementation challenges owing to insufficient municipal capacity barriers and the lack of a sufficiently flexible regulatory framework. The program’s objective is to promote energy efficiency in Mexico’s municipalities and other eligible public facilities through energy efficiency investments in the public sector and to contribute to strengthening the enabling environ- ment for energy efficiency. The project supports capacity development and institutional strengthening, while also providing financing for energy efficiency investments that will be partially repaid from anticipated energy savings. The principal implementing actor is FIDEE, a public-private financing facility, that has a sustainable financing stream based on a tax collected as a percentage of the electricity tariff. The main advantage of FIDEE as an implementer is that it is able to contract directly for works, instead of relying on municipalities, which might have limited capacity to handle the contracting. The remaining challenges are related to municipalities’ capacity for developing projects and submitting repayments to FIDEE’s revolving fund. The latter problem is due to the lack of a regulatory framework flexible enough to accom- Chapter 1. The region’s record on energy efficiency modate an innovative financing mechanism. The main lessons learned to date are related to the need to (i) ensure flexibility to adapt instruments in time, and (ii) create concentrated capacity in a single implementing agency, especially when local (municipal) capacity is limited. At the start of a program, subsidies or concessional financing (such as the concessional loan FIDEE obtained from the World Bank) can help stir initial participation. In addition, long-term sustainability and certainty come from the fact that FIDEE has a stable (albeit smaller) revenue stream to sustain its own operations. Even when the World Bank project ends, the implementing agency will be able to generate new projects. Source: WBG team; and IBRD Extranet-Projects and Operations-Project detail-P149872. Sanctions related to noncompliance with regulations on that behavioral barriers can indeed be addressed by in- 28 labeling programs or energy performance standards are ap- creasing awareness of energy efficiency benefits, such as by plied by 10 countries in the region. Fines are adjusted to a promoting energy audits in the building sector (Bagaini et Realizing the Potential of Energy Efficiency in Latin America and the Caribbean macroeconomic indicator (in most cases linked to the coun- al. 2020). try’s basic salaries), which helps maintain the operational value of sanctions despite macroeconomic variations. RISE scores and energy efficiency policies Lack of financing has also hampered advances in energy effi- ciency. Based on Latinobarometer’s 2018 survey, 71 percent RISE—Regulatory Indicators for Sustainable Energy—is a of households in the region would be willing to spend money set of indicators intended for use in comparing the policy on appliances that allow them to lower their electricity bill, and regulatory frameworks that countries have put in place but 22 percent of them said that they did not have the re- to support the achievement of Sustainable Development sources to cover the cost of the appliances (IDB 2019). Goal 7 on universal access to clean and modern energy (World Bank 2021). The third edition of the report (ESMAP The lack of information on the impacts of implemented poli- 2020) captures policies and regulations that enhance sus- cies and measures has delayed the uptake of energy efficien- tainable energy in the form of 30 indicators distributed cy measures in the region. For example, lack of awareness among four pillars: access to electricity, clean cooking, re- was a significant impediment for energy efficiency policies newable energy, and energy efficiency. in the Argentinean electricity sector (Recalde and Guzows- ki 2012). Improved information on impact would provide an As RISE scores seek to achieve a homogenous and compa- opportunity to evaluate and disseminate results and share rable scoring system, it is hard to capture all nuances and best practices among countries, enabling the industrial and differences in energy efficiency policies between countries commercial sectors to implement successful measures and between sectors in each country. Also, scores some- without public support. The lack of reliable information on times do not reflect the country’s updated policy landscape, the impacts of energy efficiency policies and programs lim- or they rely heavily on the enactment of regulation for scores, its comparability and is a barrier to their replication across without being able to assess if and how these policies are countries and sectors. It also misses the sizeable potential actually implemented on the ground. These measurement efficiency gains that could be achieved through behavioral issues notwithstanding, the RISE scores are the best avail- changes like promoting maximum heating and minimum able measure to benchmark advances in energy efficiency cooling temperatures in homes. Recent research shows policy and action among countries and sectors. Chapter 1. The region’s record on energy efficiency Notwithstanding their limitations, RISE scores are the best Brazil, Colombia, Dominican Republic, Ecuador, Guatemala, 29 available benchmarks of energy efficiency policies and ac- Nicaragua, Panama, Peru, Uruguay, and Venezuela), but the tions among countries and sectors and are thus taken as a combination of mechanisms that each country uses to en- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean reference framework for this analysis. The following catego- courage investments and financing for the development of ries of policy indicators are considered: (i) national planning; energy efficiency are different. (ii) specific energy efficiency entities, incentives, and man- dates for the private sector (industrial and commercial end Energy labeling systems provide reliable information to con- users); (iii) incentives and mandates for the public sector; sumers on the energy efficiency of products, while MEPS help (iv) financing mechanisms; (v) MEPS; (vi) energy labeling sys- displace the most inefficient equipment from the market. tems; and (vii) building energy codes. Product labeling and minimum energy efficiency standards are key to allow industry and consumers to make econom- The areas of national planning and entities are closely in- ic choices when purchasing appliances and equipment. They terdependent and are fundamental enablers to develop, im- also help to support oversight of large consumers’ compli- plement, and sustain long-term policies to promote energy ance with standards for processes, services, and products. efficiency. These policies help coordinate multiple actors, ensure sustainability of policies and programs, provide clar- There is little financial incentive for builders to pay the addi- ity on objectives and tools, channel concessional financing, tional capital costs required for more efficient buildings. Build- and enable information exchanges. ing codes attuned to energy efficiency address this barrier by setting minimum energy efficiency standards for building tech- Incentives and mandates for the private sector (industrial nologies and design elements that may include the building and commercial end users) are common elements of most envelope, HVAC systems, and lighting, among others. national energy efficiency plans. The industrial and com- mercial sectors include most large individual consumers. In The evolution of overall RISE energy efficiency scores in the many cases, initiatives cover both sectors with incentives region shows consistent improvements. Over the past de- such as specific energy-saving targets, compliance with cade all LAC countries have improved their RISE scores for mandatory audits, and the obligations to implement energy the overall energy efficiency policies. Scores have improved management systems, among others. in the region across all energy efficiency subindicators. The subindicators showing the largest improvement at the re- Incentives and mandates for the public sector usually gional level are those dealing with national planning, enti- hinge on a combination of initiatives. These comprise en- ties, and energy labeling systems.12 ergy-saving obligations in buildings, control and monitoring mechanisms, and mandatory guidelines for the purchase of The evolution of RISE scores in the largest countries of the products and services with certain energy efficiency charac- region differs from that in the rest of the region (figure 1.10). teristics. In addition, they sometimes include provisions for Compared to 2010 the largest LAC countries reported a no- public budget regulations to allow public entities to retain table improvement in energy labeling systems. There was a the energy savings they achieve through energy efficiency. certain level of progress in the other areas, such as financ- ing mechanisms and entities. Other countries had a gener- Financing of policy instruments and financial incentive al improvement in the areas of entities, national planning, schemes to eliminate or reduce barriers to energy efficiency and incentives and mandates for the public sector. Over the are key policy elements. Financial tools have shown good re- period 2010 to 2019, the largest countries made moderate sults in many countries, where they ensure the development progress in the categories of entities, financing mecha- of a market for energy-efficient goods and services. Howev- nisms, and energy labeling systems, while in the rest of the er, despite the critical need for financing, the current region- region, only financing mechanisms stands out. Policies for al context is expected to further restrict access to financing building energy codes were little changed in either of two for such investments, mostly due to a sharp slowdown in country categories. growth in the region, contracting fiscal space, inflationary Chapter 1. The region’s record on energy efficiency pressures, and exchange rate volatility. The most commonly Although each country’s characteristics, priorities, and per- used financing instrument is a trust (a specific fund target- formance on these indicators is different, it is relevant to ed to achieve a policy goal). Other mechanisms for promot- highlight the importance of making progress in the areas of ing investments are tax exemptions, tariff preferences, and national planning and entities since these are indispensable awards for excellence. Sixty-nine percent of the countries for the development of long-term policies to promote ener- analyzed included aspects related to this topic (Argentina, gy efficiency. 12 To ensure that all countries in the benchmark exercise were comparable, we used the IEA energy dataset, which included 19 LAC countries. Cuba and Venezuela are not included because they are considered outliers. Suriname and Trinidad and Tobago are not included because they lack household consumption data, which is needed for the disaggregation of the residential sector. Barbados is not in the IEA dataset. RISE covers EE policies in every sector, including utilities and transport. According to RISE, all LAC countries show some increase in Energy Efficiency Indicators (BIEE) (ECLAC 2021) and 30 the quantity and scope of energy efficiency policies imple- the IEA’s policies database (IEA 2021b), it was possible mented in the past decade (figure 1.11). Mexico ranks first, to compile a database of energy efficiency policies and Realizing the Potential of Energy Efficiency in Latin America and the Caribbean followed by Panama, Brazil, and Costa Rica; with Panama programs in LAC countries. This information was supple- showing the largest absolute score increase for the group. mented with additional measures from official country websites, as noted previously. Together these sources RISE scores are moderately well correlated with records were used to construct an extended database of cumu- of energy efficiency actions implemented in countries lative policies and actions. Figure 1.12 was constructed (figure 1.12). Using information from ECLAC’s Base of based on this information. Figure 1.10 2010 RISE energy efficiency scores in the major countries of the 2019 region and in the rest of the region, 2010 and 2019 National planning 100 80 Building Entities energycodes 60 40 20 Argentina, Brazil, Incentives & mandates: Energy Chile, Colombia, labelingsystems Industrialand Commercial Endusers Mexico, Peru Incentives & mandates: MEPS Publicsector Financing mechanisms National planning 100 80 Building Entities 60 energycodes 40 20 Chapter 1. The region’s record on energy efficiency Incentives & mandates: Energy Rest of LAC Industrialand Commercial labelingsystems Endusers Incentives & mandates: MEPS Publicsector Financing mechanisms Source: WBG, based on RISE indicators. Figure 1.11 2010 31 Overall RISE energy efficiency score, 2019 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean 2010 and 2019 90 80 70 60 RISE score 50 40 30 20 10 0 Ecuador El Salvador Uruguay Paraguay Mexico Panama Brazil Costa Rica Colombia Chile Jamaica Dominican Republic Nicaragua Argentina Bolivia Guatemala Peru Honduras Haiti Venezuela, RB Source: WBG, based on RISE indicators. Figure 1.12 Correlation between RISE scores on energy Number of energy efficiency policies in the region efficiency and the cumulative number of energy Region’s overall RISE score efficiency policies implemented in LAC, by year 45 20 18 40 16 Region’s Overall RISE Score Number of energy efficiency 35 14 policies in the region 30 12 25 10 Chapter 1. The region’s record on energy efficiency 8 20 6 15 4 10 2 0 0 1985 1993 1994 1995 1996 1997 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: RISE, BIEE, IEA. 32 Energy efficiency potential in jections of the expected evolution of energy intensity for the region the countries in the group are applied to the reported re- gional energy intensity for the “LAC region” (as defined by Realizing the Potential of Energy Efficiency in Latin America and the Caribbean OLADE14). The goal is to analyze the potential of energy effi- The projections developed in World Energy Outlook 2020 ciency measures to accelerate achievement of the SDG tar- (IEA 2021a) are used to estimate the potential for energy gets for the same set of countries. efficiency in LAC. The analyzed scenarios are: The SDS scenario projects a more pronounced decrease • The Stated Policies Scenario (STEPS) reflects the latest in energy intensity, with a larger impact in the short term, announced policy intentions and targets, insofar as they as a result of an intensification of energy efficiency policies. are backed up by detailed measures for their realization. The scenario foresees that the post-Covid-19 recovery will be accompanied by energy efficiency measures, achieving • The Sustainable Development Scenario (SDS) models an annual improvement of 1.5 percent between 2019 and an increment in energy efficiency and low-carbon ener- 2025. This implies continuing the measures already devel- gy policies and investment on top of what is assumed oped while implementing new policies with a higher impact in the STEPS scenario. The SDS puts the energy system on energy intensity. Further acceleration of the improve- on track to achieve the Paris Agreement goal, as well as ment in energy intensity is expected between 2025 and energy access and air quality goals. The assumptions on 2030 (at an annual rate of improvement of 2.7 percent) and public health and the economy are the same as in STEPS. between 2030 and 2040 (with an annual rate of improve- ment of 2.6 percent). This scenario marks a clear upside The scenarios suggest that the right energy efficiency mea- compared to the region’s improvement trend over the last sures can reduce cumulative global CO2 emissions by up to 20 years (around 0.5 percent each year). As shown in the one-third by 2030. In other words, energy efficiency mea- figure, there is still room for improvement between the re- sures account for one-third of the cumulative CO2 emission gion’s expected energy intensity based on current policies reductions between STEPS and SDS. Figure 1.13 shows the (STEPS scenario) and what could be achieved with policies projection for the region, calculated for the country group aimed to keep global warming at a level “well below” 2°C “Central and South America” (as defined by IEA).13 The pro- (SDS scenario). 13 In the IEA’s scenarios, Central and South America includes Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Curaçao, Dominican Republic, Ecuador, El Salvador, Guatemala, Haiti, Honduras, Jamaica, Nicaragua, Panama, Paraguay, Peru, Suriname, Trinidad and Tobago, Uruguay, Bolivarian Republic of Venezuela (Venezuela), and other Central and South American countries and territories. 14 For OLADE’s estimations, LAC countries include Argentina, Barbados, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Ecuador, El Salvador, Granada, Guatemala, Guyana, Honduras, Haiti, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, República Dominicana, Suriname, Trinidad y Tobago, Uruguay, Venezuela. Chapter 1. The region’s record on energy efficiency Figure 1.13 STEPS 33 Projections of ratio of final energy intensity to GDP SDS Realizing the Potential of Energy Efficiency in Latin America and the Caribbean in STEPS and SDS, 2000–40 2,83 Ratio of final energy intensity to GDP (MJ/GDP ( PPP 2017 US$) 2,65 2,63 2,53 2,44 2,43 2,37 2,22 2,24 2,23 2,03 1,94 1,93 1,83 1,63 1,48 1,43 1,23 1,03 2000 2010 2019 2025 2030 2040 Source: WBG, based on IEA 2020. STEPS = Stated Policies Scenario; SDS = Sustainable Development Scenario. Chapter 1. The region’s record on energy efficiency 34 Chapter 2. Quantitative analysis Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Page 35  Page 36  Data and measurement Methodological framework analysis Quantitative 2 Data and Energy prices and consumption are calculated based on se- ries for electricity, gasoline, diesel, fuel oil, kerosene, LPG, 35 measurement and natural gas. Interpolation and extrapolation methods Realizing the Potential of Energy Efficiency in Latin America and the Caribbean were used to fill gaps in the price series. The final energy end-use price is calculated as a consumption-weighted av- erage for each country. Energy subsidies are calculated us- The data used in the quantitative analysis come from vari- ing a price-gap approach. ety of sources, including the IEA, the International Monetary Fund (IMF), OLADE, Penn World Tables, and the World Bank’s Real energy price growth in the LAC region has been Open Data and RISE databases for the period 2000–19. The modest and, on average, has tracked the world energy variables of interest include final energy consumption, energy price (figure 2.1). However, trends have varied signifi- prices and value added, GDP, population, labor, capital stock, cantly across the region. Prices grew faster than the investment, exchange rates, heating and cooling degree days, regional average in some larger economies, such as and regulatory indicators. The energy consumption and val- Brazil, Chile, and Colombia. In Argentina, they have also ue-added indicators are disaggregated into four sectors (agri- increased but remain well below the regional average. culture and mining, manufacturing, services, and residential), They remained stagnant or even decreased in Mexico, while all other indicators are at the country level. Peru, and smaller countries. Figure 2.1 Argentina Colombia LAC Evolution of energy prices Brazil Mexico Other LAC in LAC, 2000–19 Chile Peru World 250 Consumption-weighted average energy price (2010 US$/boe) 200 150 100 0 Chapter 2. Quantitative analysis 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: EIA, OLADE, based on authors’ calculations. Note: Owing to limited data availability, the consumption-weighted average energy price for each country is computed using final energy consumption of gasoline, diesel, kerosene, LPG, natural gas, and electricity due. The world energy price is proxied by the Brent crude oil spot price. 36 Methodological improvements in energy efficiency or merely reflect the changing composition of the economy. It also helps identify framework policies that can better target specific sectors of the econ- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean omy. Mathematical details of the Fisher decomposition method are shown in the appendix. The methodological ad- vantage of the Fisher method is that it is exact. It allows for The methodological framework of quantitative analysis in- accurately calculating the difference between current ener- volves three stages: (i) Fisher decomposition of energy inten- gy consumption and energy consumption that would have sity index; (ii) econometric analysis of energy intensity driv- occurred had energy efficiency or the activity component of ers, and (iii) counterfactual simulations of energy efficiency the energy intensity index remained at the 0 level (time t0). policies. The first stage separates actual changes in energy efficiency within each sector of the economy from changes in The decomposition is performed based on a sample of LAC the economy’s industrial structure. The second stage identi- countries for which sector-level data were available for the fies key factors affecting energy efficiency. The third quantifies agriculture, manufacturing, services, and residential sectors the extent to which energy policies can improve energy effi- over the period 2005–18. Unfortunately, data limitations ciency in LAC countries. (A more extensive treatment of the prevent a more granular decomposition, even though there report’s methodology appears in the appendix.) is a considerable variation in energy intensity within eco- nomic subsectors.15 Fisher decomposition analysis separates the aggregate en- ergy intensity index into the index in each sector and the The econometric analysis involves estimating regressions activity index of changes in the productive structure of for panel data to determine the effects of energy prices and the economy. It aids in understanding whether changes in policies on the energy intensity index and its activity and a country’s average energy intensity are driven by sectoral efficiency components. Each energy policy proxied by the 15 For example, the mining sector tends to be more energy intensive than the agriculture sector, and there is considerable variation in energy intensity across manufacturing industries. Chapter 2. Quantitative analysis RISE subcomponent is included in a separate regression to In addition, bias may be created by omitting policies from 37 avoid problems of co-linearity. The control variables include the estimated regressions. This bias appears to be negligi- heating and cooling degree days that account for variation in ble, as the magnitude of estimated coefficients seemed lit- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean energy demand under different climate conditions. The cap- tle changed when all policies were added in the regression. ital-labor ratio and the squared capital-labor ratio variables Since the analysis is based on a sample of LAC countries, its control for potential impacts of capital intensity on energy results are not generalizable to other countries with differ- intensity. The investment-to-capital ratio variable indirectly ent income and development levels, production structures, controls for different energy intensities of the capital stock and rural-urban ratios. vintages. The population growth variable controls for poten- tial effects of adding infrastructure for energy efficiency and Counterfactual simulations use the estimated parameters intensity in slow-growing versus fast-growing countries. to quantify the extent to which policies affect improve- The time trend and the squared time trend variables control ments in energy efficiency. Two policies are considered. The for the rate of technological progress. All variables except first is the elimination of fossil fuel subsidies leading to an the time trend and population growth rate are converted to increase in end-use energy prices. The second is the change natural logarithms and carry elasticity interpretations. in energy efficiency regulations captured by RISE scores. Specifically, the simulations analyze a scenario where all Although caution is warranted in interpreting the estimat- LAC countries improve their RISE scores to the level of the ed effects of energy policies, the potential for endogeneity best-performing economy. bias is likely to be small. Energy prices are determined by trends in the global commodity market and are plausi- bly exogenous. However, estimated coefficients for energy policies may be confounded by unobserved factors (such as country-specific capabilities to implement these poli- cies through different stages) that are also correlated with outcomes. While this bias cannot be fully eliminated, it is effectively reduced by adding control variables, time and country fixed effects, and lagged values of policy variables. Chapter 2. Quantitative analysis 3 38 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Results of the Fisher decomposition analysis Chapter 3. Results of the Fisher decomposition analysis What have been the drivers of changes in energy intensity? Page 40  Econometric analysis and results Page 41  Counterfactual simulations and results Page 45  When compared to the United States, China, India, Russia, and Changes in energy intensity, economic composition, and 39 the European Union (average), the LAC region stands out for sectoral indices of energy efficiency in the region reveal the slow rate at which it is reducing its energy intensity (figure significant heterogeneity among the LAC countries. Guy- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean 3.1). Shifts to more energy-efficient sectors of the economy and ana had the largest reduction in its energy intensity index improvements in sectoral energy efficiency have contributed (42.8 percent) over the 2006–2018 period, while Jamaica to a decline in the energy intensity index in China, India, and had the largest increase between 2000 and 2018 (17.6 per- the European Union. In the United States and Russia, improve- cent). Among the major countries of the region, Colombia ments in sectoral energy efficiency have been partially offset by achieved the largest reduction in energy intensity (24.7 the reallocation of economic activity to more energy-intensive percent), followed by Peru (20.6 percent), Chile (19.8 per- sectors. In the countries of the regions, reductions in energy cent), Argentina (11.3 percent), Mexico (10.3 percent), and intensity were due to sectoral improvements in energy efficien- Brazil (4.4 percent). cy; there was no change in the industrial structure. Figure 3.1 Intensity (%Δ) Energy intensity decomposed by changes Activity (%Δ) in activity and in efficiency, 2000–18 Efficiency (%Δ) 30% 20% 10% Accumulated percentage change 2000-2018 0% -10% -20% Chapter 3. Results of the Fisher decomposition analysis -30% -40% -50% EU12 EU28 United States China India Russia LAC Source: WDI, IEA. 40 What have There are different sources of changes in energy intensity across LAC countries (figure 3.2). In the five major economies been the of the region, the shift in economic activity toward less ener- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean gy-intensive industries has played a major role in Brazil, re- ducing the energy intensity index of the Brazilian economy drivers of by 5.8 percent. In contrast, in Chile, Colombia, Mexico, and Peru, improvements in the sectoral energy efficiency index changes were more prominent. These yielded reductions of between 6 and 19.9 percent in the overall energy intensity index. In Ar- gentina’s case, the improvements in both indexes are roughly in energy equal. Costa Rica shows almost no change on all indicators, and Ecuador displays a noteworthy increase in energy inten- intensity? sity, driven both by changes in the composition of economic activity and by reductions in energy efficiency. Most of the largest economies seem to be undergoing structural chang- es in their industrial structures, with resources shifting from more to less energy-intensive industries. Figure 3.2 Intensity (%Δ) Energy intensity decomposed by changes in activity Activity (%Δ) and efficiency in LAC’s 10 largest economies, 2000–19 Efficiency (%Δ) 15% 10% Accumulated percentage change 2000-2018 5% 0% -5% -10% -15% -20% Chapter 3. Results of the Fisher decomposition analysis -25% -30% -35% Dominican Colombia Guatemala** Peru Chile Argentina Mexico Brazil Costa Rica Ecuador Republic Source: OLADE, UN data. Note: For Guatemala, Fisher Decomposition is performed using only three industries, manufacturing, services, and residential. For the other countries, agriculture and mining are also included. Note (**): The OLADE does not report energy consumption data of the agricultural and mining sector (ISIC A–C) for Guatemala. Therefore, Fisher Decomposition for this country does not include the agricultural and mining sector. Econometric achieving reductions in energy intensity and improving sec- toral energy efficiency. 41 analysis and Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Higher energy prices are associated with lower energy intensi- ty and higher sectoral energy efficiency. Table 2.1 shows that a 1 results percent increase in energy prices is associated with a 0.16 per- cent decrease in the economy-wide energy intensity index. This coefficient can also be interpreted as the price elasticity of ener- Regression results show the importance of energy pric- gy intensity. Table 2.2 shows that a 1 percent increase in energy es and regulatory policies in explaining changes in energy prices is associated with a 0.20 percent decrease in the sectoral intensity in the region. Higher energy prices help to lower energy efficiency index. Table 2.3 shows that a 1 percent increase energy intensity owing to improvements in sectoral energy in energy prices is associated with a 0.04 percent increase in the efficiency, though some of these improvements are offset economic activity component of the energy intensity index. by a higher energy intensity of economic activity. Improve- ments in energy efficiency regulations led to improvements These findings suggest that higher energy prices produce a in energy intensity and sectoral indices of energy efficiency decline in the index as sectoral energy efficiency improves. while having no statistically significant effect on the energy However, this effect is partially offset by a “rebound effect” intensity of economic activity. When looking at the region- (such as higher energy consumption in response to improve- al (LAC) level, national planning, governance, and financing ments in energy efficiency), leading to the reallocation of some mechanisms appear as the most salient regulatory policies economic activity to the more energy-intensive industries. 16 16 For more details on the rebound effect, see Sorrell (2007). Table 2.1. Link between RISE scores, energy intensity index, and energy price (1) (2) (3) (4) (5) Energy price -0.16*** -0.16*** -0.16*** -0.15*** -0.19*** RISE EE scores -0.08*** RISE EE scores t-1 -0.07*** RISE EE scores t-3 -0.07*** RISE EE scores t-5 -0.08*** N 319 319 300 262 224 Chapter 3. Results of the Fisher decomposition analysis R-square 0.71 0.75 0.77 0.81 0.84 Country and year FE Yes Yes Yes Yes Yes Note: Control variables include levels and squared terms of the capital-labor ratio, investment to capital ratio, population growth rate, cooling degree days (CDD), heating degree days (HDD), and levels and squared terms of a time trend. All variables are in natural logarithms except for population growth rates and time trends. FE = Fixed Effects. Source: OLADE, UN data, Penn World Tables, IEA, World Bank Table 2.2. Link between RISE scores, energy efficiency, and energy price 42 (1) (2) (3) (4) (5) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Energy price -0.20*** -0.20*** -0.20*** -0.18*** -0.22*** RISE EE score -0.09*** RISE EE scores t-1 -0.08*** RISE EE scores t-3 -0.07*** RISE EE scores t-5 -0.08*** N 319 319 300 262 224 R-square 0.72 0.75 0.78 0.81 0.84 Country and year FE Yes Yes Yes Yes Yes Note: Control variables include levels and squared terms of the capital-labor ratio, investment to capital ratio, population growth rate, cooling degree days (CDD), heating degree days (HDD), and levels and squared terms of a time trend. All variables are in natural logarithms except for population growth rates and time trends. FE = Fixed Effects. Source: OLADE, UN data, Penn World Tables, IEA, World Bank. Table 2.3. Link between RISE scores, economic activity, and energy price (1) (2) (3) (4) (5) Energy price 0.04*** 0.04*** 0.04** 0.03* 0.03** RISE EE score 0.01* RISE EE scores t-1 0.01 RISE EE scores t-3 0.00 RISE EE scores t-5 -0.00 Chapter 3. Results of the Fisher decomposition analysis N 319 319 300 262 224 R-square 0.75 0.75 0.78 0.83 0.88 Country and year FE Yes Yes Yes Yes Yes Note: Control variables include levels and squared terms of the capital-labor ratio, investment to capital ratio, population growth rate, cooling degree days (CDD), heating degree days (HDD), and levels and squared terms of a time trend. All variables are in natural logarithms except for population growth rates and time trends. FE = Fixed Effects. Source: OLADE, UN data, Penn World Tables, IEA, World Bank. Energy efficiency regulatory policies reduce energy intensi- Better national planning, strong energy efficiency policy, 43 ty and improve sectoral energy efficiency. In various model regulation and implementation capacity at all govern- specifications, improvements in RISE energy efficiency scores ment levels, and improvements in financing mechanisms Realizing the Potential of Energy Efficiency in Latin America and the Caribbean consistently show reductions in the energy intensity index all help to lower energy intensity and promote sectoral and improvements in sectoral energy efficiency. A 1 percent energy efficiency. Tables 4.4 through 4.6 show that im- increase in score results in a 0.08 percent reduction in the provements in national planning scores, the scores of en- energy intensity index and sectoral index. There is no statisti- ergy efficiency entities, and financing mechanism scores cally significant effect of regulatory policies on the economic yield negative and statistically significant effects on the activity component of energy intensity scores. The effects of energy intensity and sectoral indices. A 1 percent increase regulatory policies, measured by RISE scores, remain stable in score results in a 0.01 to 0.02 percent reduction in the over time. Higher RISE scores from three or five years ago two indices. These changes do not appear to have any have a statistically significant and negative relationship with effect on the industrial composition of the economy. the economy-wide energy intensity index and sectoral index. Table 3.4. Link between national planning RISE score, energy intensity, and energy price Energy intensity Efficiency Index Activity Index Energy price -0.17*** -0.21*** 0.04*** RISE scores: national -0.02*** -0.02*** -0.00 planning N 319 319 319 R-square 0.72 0.73 0.75 Country and year FE Yes Yes Yes Note: Dependent variables are the Fisher energy intensity, energy efficiency, and activity indices, respectively. Control variables include levels and squared terms of the capital-labor ratio, investment to capital ratio, population growth rate, cooling degree days (CDD), heating degree days (HDD), and levels and squared terms of time trend. All variables are in natural logarithms except for population growth rates and time trends. FE = Fixed Effects. Source: OLADE, UN data, Penn World Tables, IEA, World Bank. Chapter 3. Results of the Fisher decomposition analysis 44 Table 3.5. Link between the RISE score of energy efficiency entities, energy intensity, and energy price Energy intensity Efficiency Index Activity Index Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Energy price -0.15*** -0.20*** 0.04*** RISE scores: EE entities -0.02*** -0.01** -0.00 N 319 319 319 R-square 0.73 0.73 0.75 Country and year FE Yes Yes Yes Note: Dependent variables are Fisher energy intensity, energy efficiency, and activity indices, respectively. Control variables include levels and squared terms of the capital-labor ratio, investment to capital ratio, population growth rate, cooling degree days (CDD), heating degree days (HDD), and levels and squared terms of a time trend. All variables are in natural logarithms except for population growth rates and time trends. FE = Fixed Effects. Source: OLADE, UN data, Penn World Tables, IEA, World Bank. Table 3.6. Link between financing mechanism RISE scores, energy intensity, and energy price Energy intensity Efficiency Index Activity Index Energy price -0.15*** -0.19*** 0.04*** RISE scores: financial -0.02*** -0.02*** 0.00 mechanism N 319 319 319 Chapter 3. Results of the Fisher decomposition analysis R-square 0.72 0.73 0.75 Country and year FE Yes Yes Yes Note: Dependent variables are Fisher energy intensity, energy efficiency, and activity indices, respectively. Control variables include levels and squared terms of capital-labor ratio, investment to capital ratio, population growth rate, cooling degree days (CDD), heating degree days (HDD), and levels and squared terms of a time trend. All variables are in natural logarithms except for population growth rates and time trends. FE = Fixed Effects. Source: OLADE, UN data, Penn World Tables, IEA, World Bank. Counterfactual ness of the investments, as their value relative to alterna- tive investments drops. Counterfactual simulations reveal 45 simulations that countries with large energy subsidies, such as Argenti- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean na, Bolivia, Ecuador, and El Salvador, can achieve a 1.7 to 2.5 percent reduction in their energy intensity index and 2 to and results 3 percent improvements in their sectoral energy efficiency. Phasing out fossil fuel subsidies in countries with moderate subsidization, such as Chile, Colombia, and Mexico, can re- Eliminating energy subsidies has a large potential for reduc- sult in a 0.4 to 1 percent decline in the energy intensity index ing energy intensity in LAC countries (figures 3.3 and 3.4). and 0.5 to 1.3 percent improvements in their sectoral ener- From the perspective of investors, energy subsidies lower gy efficiency. The average gains resulting from eliminating the commercial feasibility of energy efficiency projects by subsidies in the entire region are estimated at 0.5 for energy increasing their payback period. This reduces the attractive- intensity and 0.7 percent for sectoral energy efficiency. Figure 3.3 Projected reductions in energy intensity from eliminating subsidies 3,0% 2,5% Argentina El Salvador 2,0% Reduction in energy intensity Ecuador Bolivia 1,5% 1,0% Chile 0,5% LAC Mexico Colombia Chapter 2. Quantitative analysis 0,0% 0% 5% 10% 15% 20% 25% Average energy subsidy rate Source: WBG, based on IEA Fossil Fuel Subsidies Database, IMF Getting Energy Prices Right Database, and OLADE. 46 Figure 3.4 Projected reductions in energy intensity from improved energy Realizing the Potential of Energy Efficiency in Latin America and the Caribbean efficiency and elimination of subsidies 3,5% Argentina El Salvador 3,0% 2,5% Bolivia Reduction in energy intensity 2,0% Ecuador 1,5% 1,0% Chile 0,5% LAC Mexico Colombia 0,0% 0% 5% 10% 15% 20% 25% Average energy subsidy rate Source: Authors’ estimates based on IEA Fossil Fuel Subsidies Database, IMF Getting Energy Prices Right Database, and OLADE. Strengthening energy efficiency regulations, particularly additional 6.5 percent to 9.5 percent improvements in their with respect to financing, planning, and energy efficiency sectoral energy efficiency. entities, can significantly improve the prospect of reducing energy intensity and raising sectoral energy efficacy (figures Raising RISE scores to Mexico’s levels in countries with mod- Chapter 3. Results of the Fisher decomposition analysis 3.6 and 3.7). erate energy efficiency regulatory policies, such as Brazil, Chile, Colombia, and Ecuador, would bring additional declines of 2 to Countries with weak regulatory policies would have achieved 3.5 percent in their energy intensity index and 2.5 to 4 percent much better energy efficiency if they had implemented additional improvements in sectoral energy efficiency. more pertinent policies and boosted their RISE score to that of the leading country (such as Mexico in 2019)17. With The average energy intensity and sectoral gains in energy the right mix of regulatory policies, Argentina, Dominican efficiency resulting from strengthening regulatory policies Republic, Guatemala, and Peru could achieve additional de- across the entire region are estimated at 2.3 percent and clines of 6 to 8 percent in their energy intensity index and 2.7 percent, respectively. 17 Mexico has a high RISE score in the three policy areas identified as of higher potential (financing, planning, and energy efficiency) due to the fact that it has put I place a large number of such policies, but this does not imply that that implementation of these policies is fully effective, nor that there is no additional space for further improvements. Figure 3.6 Δ Energy efficiency 47 Projected percentage changes in energy intensity Additional Δ: RISE Realizing the Potential of Energy Efficiency in Latin America and the Caribbean with improved RISE scores 20% 10% 11,3% 0,4% 0% -28,6% -24,7% -21,6% -20,6% -19,8% -11,3% -10,3% -4,4% -6,6% -3,5% -3,4% -10% -2,1% -2,3% -67,3% 0,0% -20% -6,9% -7,9% -3,2% -30% -6,0% -2,7% -40% Dominican Colombia Guatemala Peru Chile Argentina Mexico Brazil Costa Rica Ecuador LAC Republic Source: Authors’ estimates based on OLADE, UN data, Penn World Tables, IEA. Note: Yellow bars show additional energy intensity reductions if each country had raised their RISE score to that of the leading country. Blue bars show additional energy intensity reductions if each country had raised their RISE score to that of the leading country for 2019 RISE EE scores. RISE scores have been periodically updated, and we used the version available at the time of the analysis. Figure 3.7 Δ Energy efficiency Projected percentage changes in energy efficiency Additional Δ: RISE with improved RISE scores 10% 5% 1,5% 0,6% 4,8% -2,2% 0% -29,4% -24,6% -19,9% -15,4% -14,7% -8,1% -6,0% -3,9% -3,6% -5% -2,5% -8,6% -2,7% -10% -0,0% -15% -3,8% -20% -9,5% -3,2% Chapter 3. Results of the Fisher decomposition analysis -25% -7,4% -30% -6,6% -35% -40% Dominican Guatemala Colombia Chile Peru Mexico Argentina Costa Rica Brazil Ecuador LAC Republic Source: Authors’ estimates based on OLADE, UN data, Penn World Tables, IEA. Note: Yellow bars show additional energy intensity reductions if each country had raised their RISE score to that of the leading country. Blue bars show additional energy intensity reductions if each country had raised their RISE score to that of the leading country for 2019 RISE EE scores. RISE scores have been periodically updated, and we used the version available at the time of the analysis. Energy efficiency improvements from energy subsidies and to 13.38 percent of total energy consumption in Haiti, 9.93 48 regulatory reforms offer significant energy savings (figure 3.8). percent in El Salvador, and 8.5 percent in Honduras. Argenti- Combined, these savings vary from US$61 million in Panama na and Bolivia would also see significant energy savings from Realizing the Potential of Energy Efficiency in Latin America and the Caribbean to US$6.3 billion in Brazil, with smaller and poorer economies energy reforms (9.73 percent and 9.45 percent, respectively). of Central America and the Caribbean experiencing the larg- These savings would be even larger if energy security and en- est relative savings. For example, energy reforms could cut up vironmental benefits were accounted for. Figure 3.8 RISE Regulatory Reform score Projected reductions in the EE component of Energy Subsidy Reform energy intensity, due to elimination of subsidies Percentage of total energy consumption 7 14% 6 12% Percentage of total energy consumption 5 10% 4 8% US$ Billions 3 6% 2 4% 1 2% Chapter 3. Results of the Fisher decomposition analysis 0 0% Ecuador El Salvador Paraguay Uruguay Argentina Bolivia Brazil Chile Colombia Costa Rica Dominican Republic Guatemala Haiti Honduras Jamaica Mexico Nicaragua Panama Peru Source: Authors’ estimates based on IEA Fossil Fuel Subsidies Database, IMF Getting Energy Prices Right Database, OLADE, UN data, Penn World Tables, and IEA. Note: Yellow bars show simulated changes in the value of energy consumption resulting from elimination of energy subsidies, blue bars show additional energy savings if each country had raised their RISE score to that of the leading country, and purple markers show total energy savings as a percentage of total baseline energy consumption. 49 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Chapter 3. Results of the Fisher decomposition analysis 50 Chapter 4. How can the region improve its energy efficiency? Some recommendationsanalysis Realizing the Potential of Energy Efficiency in Latin America and the Caribbean 4 its energy How can the region improve efficiency? Some recommendations Exploiting synergies between sustainable programs, tech- incentives are flexible (that is, they can be adapted to vari- 51 nology transfer, financing, and adequate energy pricing will ous purposes and sectors) and easy to implement. They are be essential to realize the region’s EE potential. Based on usually effective in increasing the likelihood of firms invest- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean the IEA’s bottom-up scenarios (IEA 2021a), considerable ing in energy efficiency projects or carrying out energy au- improvements in the region’s energy intensity are possible, dits to better understand their current energy consumption from 1.1 percent to 2.3 percent annual reductions through (Brutscher and Ravillard 2019). Outside the region, Italy’s 2040. These scenarios mark a clear upside compared with “Superbonus” program provides an example: it finances res- the region’s paltry improvement trend over the past 20 idential energy efficiency investments with a tax credit of years (around 0.5 percent per year). 110 percent the value of the investment. The measure was launched in 2020 and has already resulted in investments There is an urgent need for sustainable and well-financed for more than €20 billion (IEA 2022). 18 In the region, actions policies and programs that focus on efficient technology linked to financial incentives have correlated with improve- integration complemented by a just phase-out of ener- ments and increased private investments in energy efficien- gy subsidies. Energy efficiency policies in the region have cy (Anderson and Newell 2004; Blok 2004). shown mixed results depending on the country and sector, with more policies and measures often not translating into Wider access to financing is effective—but only under cer- consistent improvements in energy efficiency and final en- tain conditions. For example, energy efficiency investments ergy intensity. National and local governments should take require an adequate enabling environment characterized by advantage of existing programs and improved capacities clear information, trusted parties that can provide needed to set up long-term programs and scale up the measures services, and essential equipment and material, all of which already implemented. Their focus should be on enhancing can be hard to come by. To ensure their presence, national technology transfer, improving the sustainability of pro- governments must implement clear long-term plans and grams, improving access to financing, and reducing energy streamlined procedures to support energy efficiency invest- subsidies while ensuring that vulnerable social groups are ments (IEA 2022). protected. Using Mexico’s RISE scores as a benchmark, counterfactual simulations of the effect of improved energy LAC countries must continue to reduce energy subsidies efficiency governance in each country demonstrate notable while protecting vulnerable populations. Energy subsidies reductions in their energy intensity of between 2.5 percent directly affect how energy is used and the choice to acquire and 7.9 percent, depending on the country. efficient technologies. Although the currently high energy prices makes subsidy reduction difficult, they also offer an Incentivizing the integration of the most efficient technolo- opportunity to establish frameworks that will allow subsi- gies in various sectors can be done by establishing special tax dies to be phased out automatically as soon as prices drop. schemes to promote efficient technology uptake (either in the Counterfactual simulations show that some LAC coun- form of tariffs or taxes that are removed or lowered for efficient tries can leverage the elimination of fossil fuel subsidies Chapter 4. How can the region improve its energy efficiency? Some recommendationsanalysis technology or, conversely, in the form of increased tariffs or to achieve significant reductions in their energy intensity taxes for inefficient technology). Other drivers include better (between 0.5 percent and 2.5 percent depending on the access to information on the impact of improved technologies country). As an example, Argentina showed a 37.7 percent and offers of concessional financing for select investments. increase in residential energy intensity between 2000 and Particularly for the industrial and commercial sectors, pro- 2016, fueled by a combination of greater numbers of (nonef- grams should be designed so that their sustainability and scal- ficient) household appliances and highly subsidized residen- ing are backed by co-financing from the private sector. tial tariffs.19 However, a 44 percent reduction in subsidies between 2015 and 2016 (ECLAC 2021) brought a 16 percent Better information on the impacts of existing energy ef- drop in residential energy intensity between 2016 and 2018. ficiency measures would help accelerate the uptake of technology by the private sector. Improved frameworks for Finally, the importance of coordinated government actions monitoring and reporting of results and impacts of energy and a comprehensive and integrated approach to address efficiency initiatives would enable the private sector to im- the multiple challenges to EE implementation cannot be plement successful measures without public support. understated. Success with energy efficiency depends on the actions of policymakers responsible for energy, industry, Financing and incentives facilitate investments in more housing, transport, and finance, as well as the equivalent efficient technology. When adequately designed, financial actors at the subnational and local levels. 18 https://www.governo.it/it/superbonus. 19 Per capita energy consumption in the residential sector is high for Argentina, likely due to energy consumption for heating due to colder weather. 52 References Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Anderson, S., and R. Newell. 2004. “Information Programs IEA (International Energy Agency). 2020a. Recommenda- for Technology Adoption: The Case of Energy-Efficien- tions of the Global Commission for Urgent Action on cy Audits.” Resource and Energy Economics 26 (1): 27­ Energy Efficiency. Paris: IEA. https://www.iea.org/ –50. 10.1016/j.reseneeco.2003.07.001. reports/recommendations-of-the-global-commis- sion-for-urgent-action-on-energy-efficiency. Bagaini, A., F. Colelli, E. Croci, and T. Molteni. 2020. “Assess- ing the Relevance of Barriers to Energy Efficiency Im- IEA. 2020b. Energy Efficiency Indicators: December 2020 Edi- plementation in the Building and Transport Sectors in tion: Database Documentation. Paris: IEA. Eight European Countries.” The Electricity Journal 33 (8): 106820. https://doi.org/10.1016/j.tej.2020.106820. IEA. 2021a. World Energy Outlook 2020. Paris: IEA. Blok, K. 2004. “Improving Energy Efficiency by Five Percent IEA. 2021b. “Policies Database.” IEA, Paris. https://www.iea. or More per Year?” Journal of Industrial Ecology 8 (7): org/policies?topic=Energy%20Efficiency. 87–99. https://doi.org/10.1162/1088198043630478. IEA. 2022. “Accelerating Energy Efficiency: What Boyd, G. A., and J. M. Roop. 2004. “A Note on the Fisher Ideal Governments Can Do Now to Deliver Ener- Index Decomposition for Structural Change in Energy gy Savings.” IEA, Paris. https://www.iea.org/ Intensity.” The Energy Journal 25 (1): 87–102. commentaries/accelerating-energy-efficien- cy-what-governments-can-do-now-to-deliver-ener- Brutscher, P-B, and P. Ravillard. 2019. “Promoting Energy Au- gy-savings. dits: Results from an Experiment,” EIB Working Paper 2019/06, European Investment Bank, Luxembourg. Loureiro, T., C. Pozza, F. D. Mexis, S. Olivero, C. D. Csiky, and A. https://ideas.repec.org/p/zbw/eibwps/201906.html. Bogi. 2021. “Integration of Finance in Energy Efficien- cy.” Environmental Science Proceedings 11 (1): 7. https:// CONUEE (Comisión Nacional para el Uso Eficiente de la En- doi.org/10.3390/environsciproc2021011007. ergía). n.d. https://www.gob.mx/conuee. Recalde, M., and C. Guzowski. 2012. “Boundaries in Promot- ECLAC. 2021. “Policy and Measures.” Base de Información ing Energy Efficiency: Lessons from the Argentinean de Eficiencia Energética (BIEE). https://biee-cepal.en- Case.” International Journal of Hydrogen Energy 37 (19): erdata.net/en/measures/search. 14725–29. ECLAC and GTZ (Deutsche Gesellschaft für Technische Sorrell, S. 2007. The Rebound Effect: An Assessment of the Zusammenarbeit). 2010. Indicadores de Políticas Públi- Evidence for Economy-Wide Energy Savings from Im- cas en Materia de Eficiencia Energética en América Lati- proved Energy Efficiency. London: UK Energy Research na y el Caribe. Santiago, Chile: ECLAC. Centre (UKERC). ESMAP. 2020. Regulatory Indicators for Sustainable Energy: World Bank. 2021. “World Bank Data.” https://data.world- Sustaining the Momentum. Washington, DC: World Bank. bank.org. IDB (Inter-American Development Bank). 2019. Towards World Bank. 2021. “RISE—Regulatory Indicators for Sus- Greater Energy Efficiency in Latin America and the Ca- tainable Energy.” https://rise.esmap.org/. ribbean: Progress and Policies. New York, NY: IDB. https://publications.iadb.org/publications/english/ References document/Towards_Greater_Energy_Efficiency_in_ Latin_America_and_the_Caribbean_Progress_and_ Policies.pdf. Appendix A 53 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean A.1. Note on data sources sumption comes from the World Bank’s World Develop- ment Indicators (WDIs). The findings of this Report are based on original research by authors of the Report using a dataset of 56 countries (23 LAC Macroeconomic variables: Data on key macroeconomic countries, and 33 benchmark countries/ regions), 20 years, variables such as GDP, household consumption, population, and 1,239 observations, coming from nine original sources. capital stock, investment, and exchange rates are obtained from the Penn World Table (v10.0) for the period 1950–2019. Data on final energy consumption by industry across 23 Latin American and Caribbean countries are provided by Weather: Yearly weather variables, such as heating degree the Latin American Energy Organization (OLADE). Data on days (HDDs) and cooling degree days (CDDs) for the years sector-specific energy consumption in benchmark coun- 2000–21, are obtained from the Weather for Energy Tracker tries and regions—the European Union, the United States, database maintained by the IEA and the Euro-Mediterra- China, India, and the Russian Federation—are obtained nean Center on Climate Change (CMCC). from the International Energy Agency (IEA). The time pe- riod is wide: 2000–19. Time frames and data availability Energy prices and consumption by source, including for vary by country. electricity and hydrocarbons such as gasoline, diesel, fuel oil, kerosene, liquefied petroleum gas, and natural gas, are Economic sectors in Latin America and the Caribbean obtained from OLADE. OLADE provides two energy price are defined in accordance with OLADE. The economy is thus series, one for 2000–19 with data gaps, and another, updat- divided into four ISIC 2-digit industries (ISIC Rev. 3): ed, series for 2014–20. The two series were combined and harmonized to create an energy price series covering the • Agriculture and mining (ISIC A–C) 2000–19 period with the fewest data gaps possible. Energy prices were then extrapolated to fill the series. Finally, we • Manufacturing (ISIC D) calculated a consumption weighted average energy price for each country by incorporating the energy consumption • Services (ISIC 41, 50–93) data series provided by OLADE. • Residential (rural and urban households)20 Energy subsidies are obtained from two data sources: the IEA Fossil Fuel Subsidies Database and the International A fifth, residual, sector (called Other Sectors) absorbs the re- Monetary Fund (IMF) Fossil Fuel Subsidies by Country and mainder of final energy consumption and economic activities. Fuel Database. The degree of data availability differs across Note that statistics on final energy consumption are limited the two sources. The magnitude of energy subsidies also by the small number of categories used in this method. differs. We use the IEA data whenever possible as they pro- vide more conservative estimates of energy subsidies. The Value added by industries: For the agriculture and mining, IEA data are available for 2010–20 for seven countries in our manufacturing, and services industries, we measure the study: Argentina, Bolivia, Colombia, Ecuador, El Salvador, size of economic activity by industry based on value-add- Mexico, and Trinidad and Tobago. For the remaining coun- ed data, obtained from national accounts published by the tries of Latin America and the Caribbean (LAC), we use IMF United Nations. For the residential sector, we measure eco- data, which are available for 2015–25. The drastically different nomic size using household consumption. For benchmark methodologies used by the IEA and IMF in estimating energy countries, the value added by industry and household con- subsidies make it difficult to compare the two data sets. 20 The “Residential” sector is not an ISIC industry. According to OLADE’s Energy Statistics Manual 2017: “The end-use Appendix sectors are classified according to the traditional division of economic sectors and the ISIC (International Standard Industrial Classification), version 3. It also includes the residential sector, which is not an economic activity.” Table A.1. Data sources 54 Source Data type Details Realizing the Potential of Energy Efficiency in Latin America and the Caribbean –19 (OECD), and on benchmark Final energy consumption Data on 23 LAC countries in 2000­ OECD and IEA by industry countries and regions (IEA). Definition of economic Four sectors of focus: agriculture and mining, manufacturing, services, sectors in LAC countries and residential. OLADE Energy prices and Data on major energy sources—electricity, gasoline, diesel, fuel oil, consumption by source kerosene, liquefied petroleum gas, natural gas—in 2000–19. IEA Estimates of energy subsidies in 2010–20 in LAC are obtained from two Energy subsidies data sources using very different methods to estimate subsidies, making it difficult to compare data across them. IMF Data on key macroeconomic variables in 1950–2019, such as GDP, Penn World Macroeconomic variables household consumption, population, capital, investment, and exchange Tables rates, are obtained from the Penn World Table (v10.0). RISE energy efficiency RISE Database Time period, 2010–19. scores UN data Value added by sector Time period varies by country. World Bank WDI Yearly weather variables in 2000–21, such as heating degree days and IEA/CMCC Weather cooling degree days, are obtained from the Weather for Energy Tracker database Note: CMCC = Euro-Mediterranean Center on Climate Change; GDP = gross domestic product; IEA = International Energy Agency; IMF = International Monetary Fund; LAC = Latin America and the Caribbean; OECD = Organisation for Appendix Economic Co-operation and Development; OLADE = Latin-American Energy Organization; RISE = Regulatory Indicators for Sustainable Energy; UN = United Nations; WDIs = World Bank World Development Indicators. Table A.2. Selected countries and regions 55 Energy Economic Energy Country/region Time period Realizing the Potential of Energy Efficiency in Latin America and the Caribbean intensity (%) activity (%) efficiency (%) EU12 2000–18 -25.2 -6.6 -19.9 EU28 2000–18 -24.7 -5.7 -20.2 United States 2000–18 -27.6 23.1 -41.1 China 2000–18 -45.9 -15.5 -36.0 India 2000–18 -40.6 -10.1 -33.9 Russia Federation 2000–18 -32.8 4.1 -35.5 Latin America and the Caribbean 2000–18 -9.5 -0.7 -8.8 Guyana 2006–18 -42.8 -25.2 -23.6 Belize 2001–18 -35.9 -9.2 -29.4 Dominican Republic 2000–13 -28.6 1.1 -29.4 Honduras* 2000–19 -27.9 6.2 -32.1 Colombia 2000–16 -24.7 -6.0 -19.9 Guatemala* 2001–18 -21.6 4.0 -24.6 Peru 2000–11 -20.6 -6.9 -14.7 Chile 2003–15 -19.8 -5.3 -15.4 Panama 2007–19 -19.8 -11.5 -9.4 Nicaragua 2006–19 -12.9 -7.1 -6.2 Argentina 2004–19 -11.3 -5.6 -6.0 Mexico 2000–19 -10.3 -2.3 -8.1 Haiti* 2005–18 -9.5 -11.5 2.3 El Salvador 2000–16 -5.4 -1.0 -4.4 Paraguay* 2000–17 -4.8 -11.3 7.3 Brazil 2000–19 -4.4 -5.8 1.5 Grenada 2000–18 -4.2 0.0 -4.2 Trinidad and Tobago* 2000–15 -3.3 -14.1 12.6 Suriname 2006–18 -1.0 0.4 -1.4 Costa Rica 2000–14 0.4 -0.2 0.6 Barbados* 2006–19 4.6 -18.9 28.9 Bolivia 2000–18 4.7 -9.3 15.4 Ecuador 2000–19 11.3 6.2 4.8 Jamaica 2000–18 17.6 -4.7 23.4 Uruguay 2000–19 27.5 -8.0 38.6 Note (*): The OLADE does not report energy consumption data of the agricultural and mining sector (ISIC A–C) from Appendix Honduras, Guatemala, Haiti, Paraguay, Trinidad and Tobago, and Barbados. Therefore, Fisher Decomposition for these countries does not include the agricultural and mining sector. 56 A.2. Results Result 1: Fisher decomposition analysis Realizing the Potential of Energy Efficiency in Latin America and the Caribbean In application to energy, the Fisher Ideal index decompos- es changes in aggregate energy intensity into an activity (or The report applies the Fisher decomposition method (Boyd composition) component, measuring the shifting of eco- and Roop 2004) to disaggregate broad trends in energy nomic activities across sectors (e.g., from industry to ser- trends into discrete efficiency improvements and changes vices), and an efficiency component, which measures the in economic activity. changes in energy efficiency within each sector. The Fisher Ideal Index is a perfect decomposition with no resid- Technically, the Fisher Ideal Index is calculated as the geo- ual term, which means that it can perfectly explain the chang- metric mean of the Laspeyres Index and the Paasche Index. es in energy intensity compared to a base period. In addition, The former hold constant economic activities (or compo- according to Boyd and Roop (2004), the Fisher Ideal index sat- sition) or efficiency at the base period (t = 0) levels, and the isfies all four axioms of the index number theory, namely, factor latter hold constant economic activities or efficiency at the reversal, positivity, time reversal, and quantity reversal. end-period (t = T) period levels. Appendix LAC region 57 In the LAC region, the economic weight of the industrial sector is clearly shifting over time to both the services and residen- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean tial sectors. Energy efficiency (EE) is not improving in the industrial and services sectors but is so in the residential sector, backed by successful EE policies. Figure A.1 Overall Intensity Fisher indices, sectoral composition, and Activity Component energy efficiency - LAC region Efficiency Component Energy Indices - LAC 1,2 1,0 0,8 0,6 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - LAC Industry Services Residential 0,8 0,6 Percentage 0,4 0,2 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Industry corresponds to (ISIC C-F), Services corresponds to (ISIC G-Q). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - LAC Industry Services Residential 9,0 MJ/million 2010 US$ 6,0 3,0 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Industry (ISIC C-F), Services (ISIC G-Q) LAC countries 58 Argentina’s improvement in energy intensity can be most- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean ly attributed to changes in economic composition, a shift away from manufacturing, and increased efficiency in the services sector. Figure A.2 Overall Intensity Fisher indices, sectoral composition, Activity Component and energy efficiency - Argentina Efficiency Component Energy Indices - Argentina 1,05 1,00 0,95 0,90 0,85 0,80 0,75 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Real value added, percentage of total - Argentina Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Argentina Agriculture, Fishing and Mining Manufacturing Services All other Residential MJ/million 2010 US$ 12,0 10,0 8,0 6,0 4,0 2,0 0,0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Figure A.3 Overall Intensity 59 Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Belize Efficiency Component Energy Indices - Belize 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - Belize Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Belize Agriculture, Fishing and Mining Manufacturing Services All other Residential 70,00 MJ/million 2010 US$ 60,00 50,00 40,00 30,00 20,00 10,00 0,00 -10,00 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) 60 Figure A.4 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Bolivia Efficiency Component Energy Indices - Bolivia 1,60 1,20 0,80 0,40 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - Bolivia Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Bolivia Agriculture, Fishing and Mining Manufacturing Services All other Residential MJ/million 2010 US$ 25,00 20,00 15,00 10,00 5,00 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) In the case of Bolivia, data on the energy consumption of the construction sector (ISIC rev.3, F) were also available. Thus, the sectors used in the Fisher decomposition include agriculture and mining, manufacturing, construction, Appendix services, and residential. Figure A.5 Overall Intensity 61 Fisher indices, sectoral composition, and energy efficiency - Brazil Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Efficiency Component Energy Indices - Brazil 1,10 1,05 1,00 0,95 0,90 0,85 0,80 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Real value added, percentage of total - Brazil Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Brazil Agriculture, Fishing and Mining Manufacturing Services All other Residential MJ/million 2010 US$ 16,00 12,00 8,00 4,00 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) In Brazil the reduction in energy intensity due to the reduced relative economic weight of the manufacturing sector has been offset by that sector’s decreased energy efficiency, alongside the decreased efficiency of the agriculture and Appendix mining sector. 62 Figure A.6 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Chile Efficiency Component Energy Indices - Chile 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real value added, percentage of total - Chile Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Chile Agriculture, Fishing and Mining Manufacturing Services All other Residential 12,0 MJ/million 2010 US$ 10,0 8,0 6,0 4,0 2,0 0,0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Figure A.7 Overall Intensity 63 Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Colombia Efficiency Component Energy Indices - Colombia 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Real value added, percentage of total - Colombia Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Colombia Agriculture, Fishing and Mining Manufacturing Services All other Residential MJ/million 2010 US$ 16,00 12,00 8,00 4,00 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Colombia’s energy efficiency improvements can be largely attributed to the manufacturing, residential, agriculture, and mining sectors, which saw considerable reductions in energy intensity. In small part, they can also be attributed Appendix to the reduced weight of the manufacturing sector relative to the rest of the economy. 64 Figure A.8 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Costa Rica Efficiency Component Energy Indices - Costa Rica 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Real value added, percentage of total - Costa Rica Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Costa Rica Agriculture, Fishing and Mining Manufacturing Services All other Residential 8,00 MJ/million 2010 US$ 7,00 6,00 5,00 4,00 3,00 2,00 1,00 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Figure A.9 Overall Intensity 65 Fisher indices, sectoral composition, and energy Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean efficiency - Dominican Republic Efficiency Component Energy Indices - Dominican Republic 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Real value added, percentage of total - Dominican Republic Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Dominican Republic Agriculture, Fishing and Mining Manufacturing Services All other Residential 7,00 MJ/million 2010 US$ 6,00 5,00 4,00 3,00 2,00 1,00 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) 66 Figure A.10 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Ecuador Efficiency Component Energy Indices - Ecuador 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Real value added, percentage of total - Ecuador Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Ecuador Agriculture, Fishing and Mining Manufacturing Services All other Residential 9,00 MJ/million 2010 US$ 8,00 7,00 6,00 5,00 4,00 3,00 2,00 1,00 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Figure A.11 Overall Intensity 67 Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - El Salvador Efficiency Component Energy Indices - El Salvador 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Real value added, percentage of total - El Salvador Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - El Salvador Agriculture, Fishing and Mining Manufacturing Services All other Residential 8,00 MJ/million 2010 US$ 7,00 6,00 5,00 4,00 3,00 2,00 1,00 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) 68 Figure A.12 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Grenada Efficiency Component Energy Indices - Grenada 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - Grenada Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Grenada Agriculture, Fishing and Mining Manufacturing Services All other Residential 7,00 MJ/million 2010 US$ 6,00 5,00 4,00 3,00 2,00 1,00 0,00 -1,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Figure A.13 Overall Intensity 69 Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Guyana Efficiency Component Energy Indices - Guyana 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - Guyana Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Guyana Agriculture, Fishing and Mining Manufacturing Services All other Residential 80,00 MJ/million 2010 US$ 70,00 60,00 50,00 40,00 30,00 20,00 10,00 0,00 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) 70 Figure A.14 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Jamaica Efficiency Component Energy Indices - Jamaica 1,80 1,60 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - Jamaica Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Jamaica Agriculture, Fishing and Mining Manufacturing Services All other Residential 70,00 MJ/million 2010 US$ 60,00 50,00 40,00 30,00 20,00 10,00 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Figure A.15 Overall Intensity 71 Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Mexico Efficiency Component Energy Indices - Mexico 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Real value added, percentage of total - Mexico Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Mexico Agriculture, Fishing and Mining Manufacturing Services All other Residential 2,50 MJ/million 2010 US$ 2,00 1,50 1,00 0,50 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) 72 Figure A.16 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Nicaragua Efficiency Component Energy Indices - Nicaragua 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Real value added, percentage of total - Nicaragua Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Nicaragua Agriculture, Fishing and Mining Manufacturing Services All other Residential 12,00 MJ/million 2010 US$ 10,00 8,00 6,00 4,00 2,00 0,00 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Figure A.17 Overall Intensity 73 Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Panama Efficiency Component Energy Indices - Panama 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Real value added, percentage of total - Panama Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Panama Agriculture, Fishing and Mining Manufacturing Services All other Residential 20,00 MJ/million 2010 US$ 18,00 16,00 14,00 12,00 10,00 8,00 6,00 4,00 2,00 0,00 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) 74 Figure A.18 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Peru Efficiency Component Energy Indices - Peru 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Real value added, percentage of total - Peru Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Peru Agriculture, Fishing and Mining Manufacturing Services All other Residential 12,00 MJ/million 2010 US$ 10,00 8,00 6,00 4,00 2,00 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Figure A.19 Overall Intensity 75 Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Suriname Efficiency Component Energy Indices - Suriname 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - Suriname Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Suriname Agriculture, Fishing and Mining Manufacturing Services All other Residential 12,00 MJ/million 2010 US$ 10,00 8,00 6,00 4,00 2,00 0,00 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) 76 Figure A.20 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - Uruguay Efficiency Component Energy Indices - Uruguay 1,60 1,40 1,20 1,00 0,80 0,60 0,40 0,20 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Real value added, percentage of total - Uruguay Agriculture, Fishing and Mining Manufacturing Services All other Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Agriculture, Fishing and Mining correspond to (ISIC Rev.3, A-C); Services corresponds to (ISIC Rev.3, G-O and 41); Manufacturing corresponds to (ISIC Rev.3, D). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Urugay Agriculture, Fishing and Mining Manufacturing Services All other Residential 18,00 MJ/million 2010 US$ 16,00 14,00 12,00 10,00 8,00 6,00 4,00 2,00 0,00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Appendix Note : Agriculture, Fishing and Mining (ISIC Rev 3, A-C); Manufacturing (ISIC Rev 3, D), Services (ISIC Rev 3, G-O and 41) Benchmark countries as a lack of value-added data on manufacturing from the 77 WDI, the exercise focused on four sectors: industry, ser- Results of a Fisher decomposition of benchmark countries are vices, residential, and a residual category called “all others.” Realizing the Potential of Energy Efficiency in Latin America and the Caribbean shown in this section. Countries and regions selected as bench- marks in this exercise include the United States, the European For the sake of comparison, a Fisher decomposition of 19 LAC Union (EU28 and EU12), China, India, and the Russian Federation. countries is also presented. These countries are Argentina, Bo- livia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Because of a lack of data on the energy consumption of the Ecuador, El Salvador, Guatemala, Haiti, Honduras, Jamaica, agriculture and fishing sectors in the IEA database, as well Mexico, Nicaragua, Panama, Paraguay, Peru, and Uruguay. Figure A.21 Activity Component Efficiency Component Fisher indices, sectoral composition, and energy Overall Intensity efficiency - China Energy Indices - China 1,4 1,2 1,0 0,8 0,6 0,4 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - China Industry Services Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Industry corresponds to (ISIC 2018 C-F), Services corresponds to (ISIC G-Q). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - China Industry Services Residential MJ/million 2010 US$ 20,0 16,0 12,0 8,0 4,0 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Industry (ISIC C-F), Services (ISIC G-Q) 78 Figure A.22 Overall Intensity Fisher indices, sectoral composition, Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean and energy efficiency - India Efficiency Component Energy Indices - India 1,4 1,2 1,0 0,8 0,6 0,4 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - India Industry Services Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Industry corresponds to (ISIC C-F), Services corresponds to (ISIC G-Q). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - India Industry Services Residential 20,0 MJ/million 2010 US$ 16,0 12,0 8,0 4,0 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Industry (ISIC C-F), Services (ISIC G-Q) Figure A.23 Overall Intensity 79 Fisher indices, sectoral composition, and Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean energy efficiency - Russian Federation Efficiency Component Energy Indices - Russian Federation 1,4 1,2 1,0 0,8 0,6 0,4 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - Russian Federation Industry Services Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Industry corresponds to (ISIC C-F), Services corresponds to (ISIC G-Q). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - Russian Federation Industry Services Residential 20,0 MJ/million 2010 US$ 16,0 12,0 8,0 4,0 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Industry (ISIC C-F), Services (ISIC G-Q) 80 Figure A.24 Overall Intensity Fisher indices, sectoral composition, and energy Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean efficiency - United States Efficiency Component Energy Indices - United States 1,4 1,2 1,0 0,8 0,6 0,4 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - United States Industry Services Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Industry corresponds to (ISIC C-F), Services corresponds to (ISIC G-Q). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - United States Industry Services Residential 10,0 MJ/million 2010 US$ 8,0 6,0 4,0 2,0 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Industry (ISIC C-F), Services (ISIC G-Q) Figure A.25 Overall Intensity 81 Fisher indices, sectoral composition, and energy Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean efficiency - EU28 Efficiency Component Energy Indices - EU28 1,4 1,2 1,0 0,8 0,6 0,4 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - EU28 Industry Services Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Industry corresponds to (ISIC C-F), Services corresponds to (ISIC G-Q). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - EU28 Industry Services Residential 10,0 MJ/million 2010 US$ 8,0 6,0 4,0 2,0 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Industry (ISIC C-F), Services (ISIC G-Q) 82 Figure A.26 Overall Intensity Fisher indices, sectoral composition, and Activity Component Realizing the Potential of Energy Efficiency in Latin America and the Caribbean energy efficiency - EU12 Efficiency Component Energy Indices - EU12 1,4 1,2 1,0 0,8 0,6 0,4 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real value added, percentage of total - EU12 Industry Services Residential 1,0 0,8 Percentage 0,6 0,4 0,2 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Note: 1) Residential sector is measured by the share of real private consumption in GDP following Metcalf (2008); Industry corresponds to (ISIC C-F), Services corresponds to (ISIC G-Q). 2) Shares do not sum up to 100% due to different value measures per sector. GDP = gross domestic product; MJ = megajoules. Energy Intensity by Sector - EU12 Industry Services Residential 10,0 MJ/million 2010 US$ 8,0 6,0 4,0 2,0 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Appendix Note : Industry (ISIC C-F), Services (ISIC G-Q) Result 2: Econometric analysis 83 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Table A.3. Effect of removing energy subsidies on energy intensity Country Years Δ Energy intensity (%) Δ Simulated (%) Avg. subsidy (%) Argentina 2004–19 -11.3 -13.7 19.1 Bolivia 2000–18 4.7 3.0 10.8 Brazil 2000–19 -4.4 -4.5 0.4 Chile 2003–15 -19.8 -20.8 7.8 Colombia 2000–16 -24.7 -25.1 3.3 Costa Rica 2000–14 0.4 0.4 0.1 Dominican Republic 2000–13 -28.6 -28.6 0.0 Ecuador 2000–19 11.3 9.5 11.2 El Salvador 2000–16 -5.4 -7.8 18.1 Guatemala 2001–18 -21.6 -21.7 0.8 Haiti 2005–18 -9.5 -9.5 0.0 Honduras 2000–19 -27.9 -27.9 0.0 Jamaica 2000–18 17.6 17.6 0.0 Mexico 2000–19 -10.3 -10.7 3.3 Nicaragua 2006–19 -12.9 -12.9 0.0 Panama 2007–19 -19.8 -19.8 0.0 Paraguay 2000–17 -4.8 -4.8 0.0 Peru 2000–11 -20.6 -20.6 0.0 Appendix Uruguay 2000–19 27.5 27.5 0.0 Table A.4. Effect of removing energy subsidies on energy efficiency 84 Country Years Δ Energy efficiency (%) Δ Simulated (%) Avg. subsidy (%) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Argentina 2004–19 -6.0 -9.3 19.1 Bolivia 2000–18 15.4 13.0 10.8 Brazil 2000–19 1.5 1.4 0.4 Chile 2003–15 -15.4 -16.6 7.8 Colombia 2000–16 -19.9 -20.4 3.3 Costa Rica 2000–14 0.6 0.6 0.1 Dominican Republic 2000–13 -29.4 -29.4 0.0 Ecuador 2000–19 4.8 2.6 11.2 El Salvador 2000–16 -4.4 -7.5 18.1 Guatemala 2001–18 -24.6 -24.7 0.8 Haiti 2005–18 2.3 2.3 0.0 Honduras 2000–19 -32.1 -32.1 0.0 Jamaica 2000–18 23.4 23.4 0.0 Mexico 2000–19 -8.1 -8.7 3.3 Nicaragua 2006–19 -6.2 -6.2 0.0 Panama 2007–19 -9.4 -9.4 0.0 Paraguay 2000–17 7.3 7.3 0.0 Peru 2000–11 -14.7 -14.7 0.0 Uruguay Appendix 2000–19 38.6 38.6 0.0 Table A.5. Effects of policies on energy intensity 85 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean RISE: Δ Energy RISE: RISE: EE RISE: Country Years National intensity (%) Aggregate (%) entities (%) Financing (%) planning (%) Argentina 2004–19 -11.3 -18.6 -12.8 -12.7 -12.9 Bolivia 2000–18 4.7 -3.1 3.1 3.1 2.9 Brazil 2000–19 -4.4 -6.6 -4.8 -4.8 -4.9 Chile 2003–15 -19.8 -23.0 -20.5 -20.5 -20.5 Colombia 2000–16 -24.7 -27.4 -25.3 -25.2 -25.3 Costa Rica 2000–14 0.4 -3.0 -0.3 -0.2 -0.3 Dominican Republic 2000–13 -28.6 -34.6 -29.8 -29.8 -29.9 Ecuador 2000–19 11.3 7.9 10.6 10.6 10.6 El Salvador 2000–16 -5.4 -12.9 -6.9 -6.9 -7.1 Guatemala 2001–18 -21.6 -28.5 -23.0 -22.9 -23.1 Haiti 2005–18 -9.5 -22.9 -12.3 -12.3 -12.6 Honduras 2000–19 -27.9 -36.5 -29.7 -29.7 -29.9 Jamaica 2000–18 17.6 12.6 16.6 16.6 16.5 Mexico 2000–19 -10.3 -10.3 -10.3 -10.3 -10.3 Nicaragua 2006–19 -12.9 -19.4 -14.2 -14.2 -14.4 Panama 2007–19 -19.8 -21.4 -20.1 -20.1 -20.1 Paraguay 2000–17 -4.8 -13.0 -6.5 -6.5 -6.7 Peru 2000–11 -20.6 -28.5 -22.2 -22.1 -22.4 Uruguay 2000–19 27.5 22.2 25.1 25.1 25.0 Appendix EE = energy efficiency; RISE = Regulatory Indicators for Sustainable Energy. Table A.6. Effects of policies on energy efficiency 86 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean RISE: Δ Energy RISE: RISE: EE RISE: Country Years National Efficiency (%) Aggregate entities Financing planning Argentina 2004–19 -6.0 -0.1470 -0.0841 -0.0744 -0.0884 Bolivia 2000–18 15.4 0.0581 0.1248 0.1492 0.1490 Brazil 2000–19 1.5 -0.0104 0.0075 0.0149 0.0119 Chile 2003–15 -15.4 -0.1912 -0.1848 -0.1605 -0.1585 Colombia 2000–16 -19.9 -0.2306 -0.2019 -0.2011 -0.1989 Costa Rica 2000–14 0.6 -0.0330 -0.0041 -0.0024 -0.0082 Dominican Republic 2000–13 -29.4 -0.3601 -0.2937 -0.3097 -0.2995 Ecuador 2000–19 4.8 0.0121 0.0479 0.0380 0.0425 El Salvador 2000–16 -4.4 -0.1283 -0.0535 -0.0606 -0.0574 Guatemala 2001–18 -24.6 -0.3200 -0.2514 -0.2522 -0.2564 Haiti 2005–18 2.3 -0.1451 -0.0437 0.0039 0.0135 Honduras 2000–19 -32.1 -0.4108 -0.3385 -0.3295 -0.3826 Jamaica 2000–18 23.4 0.1748 0.2179 0.2300 0.2341 Mexico 2000–19 -8.1 -0.0815 -0.0824 -0.0826 -0.0828 Nicaragua 2006–19 -6.2 -0.1401 -0.0768 -0.0733 -0.0692 Panama 2007–19 -9.4 -0.1141 -0.0959 -0.0935 -0.0935 Paraguay 2000–17 7.3 -0.0299 0.0665 0.0649 -0.0246 Peru 2000–11 -14.7 -0.2422 -0.1686 -0.1534 -0.2232 Uruguay 2000–19 38.6 0.3108 0.3546 0.3512 0.3435 Appendix EE = energy efficiency; RISE = Regulatory Indicators for Sustainable Energy. Energy price In addition, we also try to capture the vintage of a country’s 87 capital by introducing the investment-capital ratio into our The energy price is statistically significant and negatively empirical models as a control variable. Under the assump- Realizing the Potential of Energy Efficiency in Latin America and the Caribbean correlated with energy intensity. The coefficients on energy tion that newer capital is more energy efficient, we expect to price in columns (1)–(5) of table A.7 are stable and suggest see a negative relationship between the investment-capital that a 1 percent increase in energy price is associated with ratio and energy intensity and efficiency. a 16 percent drop in energy intensity. An increase in energy price could be achieved through the elimination of various However, we do not find a statistically significant relation- types of supply- or demand-side energy subsidies. ship between the capital-labor ratio, the investment-capital ratio, and Fisher indices for energy intensity, efficiency, and Further regression analyses of the two components of ener- economic activity, as is shown in tables A.7–A.9. gy intensity—energy efficiency and economic activity—re- veal that the effect of energy price on energy intensity main- Weather ly acts through reductions in energy efficiency. As table A.8 shows, a 1 percent increase in energy price is correlated with To control for variations in weather patterns by country a 20 percent decrease in the Fisher Efficiency Index, indi- and over years, we also include CDDs and HDDs in our re- cating a 20 percent improvement in energy efficiency when gression models. As is typical in studies of energy demand, holding economic activity constant. changes in local weather, as captured by CDDs and HDDs, may drive seasonal changes in energy demand, as hotter Surprisingly, energy price is positively correlated with eco- temperatures may prompt people to increase their use of nomic activity. Table A.9 shows that a 1 percent increase air conditioning, electric fans, and other cooling equipment. in energy price is associated with a 3–4 percent increase Similarly, cold days may increase households’ demand for in the Fisher Activity Index. This positive correlation indi- heating. In our analyses, however, we find that the relation- cates that as energy prices (inclusive of government taxes ships among CDD, HDD, and energy intensity, efficiency, or and subsidies) increase, economic activities become more activity are not statistically significant. This result could concentrated in energy-intensive industries such as man- be explained by the fact that temperatures vary less in the ufacturing, and vice versa. This relationship could be due to tropical climates typical of the majority of LAC countries in more energy-intensive economic activities creating more our sample, compared to temperate zones. demand for energy and thus raising energy prices. Exogenous technological advances Population growth To capture any exogenous technological advances common As tables A.7–A.9 show, population growth is statistically to all countries in the LAC region, we also include linear and significant and positively correlated with energy intensity nonlinear time trends in our empirical models. These trends and energy efficiency. Overall, a 1 percentage point increase not only capture technological innovations and improve- in the population growth rate is associated with a roughly ments available for all LAC countries, but also any shocks 13–15 percent increase in energy intensity and a 15–17 per- to energy supply or demand and macroeconomic shocks cent increase in the Fisher Energy Efficiency Index (indicating experienced by the whole region. Comparing tables A.7–A.9, a worsening of energy efficiency). As Metcalf (2008) argues, we find that exogenous technological progress plays an im- LAC countries with fast-growing populations could suffer portant role in explaining the reductions in energy intensity from congestion-induced energy costs by being overly reliant and the shift of economic activities toward less-energy-in- on existing energy generation technologies that are relatively tensive industries such as services. inefficient. In addition, countries with fast population growth rates could also attract more energy-intensive activities. On average, the energy intensity of the LAC countries decreases by about 1 percent per year and most of this Thus, without investments in more efficient energy produc- decline can be attributed to the structural change of the tion and consumer technologies, these factors could lead to economies toward less-energy-intensive industries. This increased energy intensity. could be due to new and better technologies being adopt- ed, the increased efficiency of energy and other resource Capital-labor ratio and investment-capital ratio (vintage use, improvements in productivity, and other macroeco- of capital) nomic factors that enhance energy efficiency across the entire region. We measure the overall capital intensity of LAC economies using the capital-labor ratio, and also allow for a potentially It is also worth noting that the average annual decline in en- nonlinear relationship between the capital-labor ratio and ergy intensity masks the decreasing return effect. The sta- energy intensity, energy efficiency, and economic activity. tistically significant and positive coefficients on the square Appendix We guess that a more capital-intensive energy is likely to be terms of time trends suggest that the decline in energy in- more reliant on energy consumption. tensity is slowing across the LAC region. Policy (RISE) changes have led to the observed changes in energy in- 88 tensity and efficiency by appealing to Granger causality. A highlight of our analyses is our ability to measure changes As is shown in the last three rows of tables A.7 and A.8, Realizing the Potential of Energy Efficiency in Latin America and the Caribbean in EE policies, albeit imperfectly. We rely on the Regulatory changes in the RISE index from one, three, and five years Indicators for Sustainable Energy (RISE) index to capture ago are significantly associated with decreases in Fisher changes in EE policies in 19 LAC countries over time. energy intensity and EE indices. Moreover, this relation- ship is remarkable stable across the models using differ- Tables A.7–A.9 reveal two patterns concerning the role ent lagged RISE scores. of government policies in reducing energy intensity and improving energy efficiency. First, more government EE Furthermore, we reestimate our models using disaggregat- policies are negatively correlated with energy intensi- ed categories of RISE scores and present the results of this ty and energy efficiency, suggesting that policies may exercise in tables A.10–A.12. We identify three sets of gov- have played a role in explaining the reduction in energy ernment policies, national plans, energy efficiency entities, intensity and improvements in energy efficiency in the and financing mechanisms, to be significantly correlated LAC countries. Second, we examine whether past policy with reductions in energy intensity and efficiency. Appendix Table A.7. Relationship between energy prices, regulatory policies, and energy intensity index 89 (1) (2) (3) (4) (5) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean -0.16*** -0.16*** -0.16*** -0.15*** -0.19*** Log(energy price) (0.042) (0.041) (0.044) (0.049) (0.054) 0.19 0.11 0.045 -0.11 -0.31 Log(K/L) (0.22) (0.19) (0.20) (0.25) (0.36) -0.0099 -0.0055 -0.0038 0.0019 0.0097 Log(K/L)^2 (0.010) (0.0090) (0.0093) (0.011) (0.016) -0.023 -0.0099 -0.023 0.0023 0.012 Log(I/K) (0.033) (0.032) (0.032) (0.034) (0.041) 0.15*** 0.13*** 0.15*** 0.14** 0.11 Population growth rate (5.00) (4.89) (5.10) (6.12) (8.17) -0.094 -0.10 -0.11 -0.10 -0.12 Log(CDD18) (0.082) (0.079) (0.079) (0.082) (0.091) -0.037 -0.042 -0.034 -0.019 -0.041 Log(HDD16) (0.031) (0.029) (0.030) (0.030) (0.033) -0.013*** -0.0087* -0.0075 -0.011 -0.0063 Time trend (0.0049) (0.0049) (0.0055) (0.0072) (0.010) 0.00093*** 0.0011*** 0.00100*** 0.0011*** 0.00078** (Time trend)^2 (0.00023) (0.00022) (0.00024) (0.00029) (0.00038) -0.080*** Log(RISE EE Scores) (0.016) -0.072*** Log(RISE EE Scores)t-1 (0.017) -0.069*** Log(RISE EE Scores)t-3 (0.020) -0.078*** Log(RISE EE Scores)t-5 (0.027) N 319 319 300 262 224 R-Square 0.7140 0.7467 0.7680 0.8073 0.8388 Country FE Yes Yes Yes Yes Yes Year FE Yes Yes Yes Yes Yes Dependent Var. = Log Energy Intensity Note: CDD = cooling degree day; CDD18 uses 18 degrees Celsius as the temperature threshold for counting degree days. For example, a day with average temperature of 19 degrees Celsius counts as one degree day, a day with average temperature of 20 degrees Celsius counts as 2 degree days, and a day with average temperature below 18 degrees Celsius counts as zero degree days. HDD = heating degree day; HDD16 uses 16 degrees Celsius as the temperature Appendix threshold. EE = energy efficiency; FE = Fixed Effects; I/K = investment-capital ratio; K/L = capital-labor ratio; RISE = Regulatory Indicators for Sustainable Energy. Table A.8. Relationship between energy prices, regulatory policies, and energy efficiency index 90 (1) (2) (3) (4) (5) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean -0.20*** -0.20*** -0.20*** -0.18*** -0.22*** Log(energy price) (0.043) (0.042) (0.046) (0.051) (0.058) 0.073 -0.0086 -0.12 -0.42 -0.84** Log(K/L) (0.26) (0.23) (0.24) (0.30) (0.41) -0.0044 0.00057 0.0045 0.016 0.033* Log(K/L)^2 (0.012) (0.010) (0.011) (0.014) (0.019) 0.030 0.045 0.028 0.022 -0.0058 Log(I/K) (0.037) (0.038) (0.037) (0.040) (0.048) 0.16*** 0.15*** 0.17*** 0.16** 0.11 Population growth rate (5.43) (5.32) (5.53) (6.58) (8.59) -0.13 -0.14 -0.100 -0.075 -0.041 Log(CDD18) (0.097) (0.096) (0.095) (0.11) (0.11) -0.041 -0.046 -0.035 -0.022 -0.038 Log(HDD16) (0.034) (0.032) (0.033) (0.033) (0.034) -0.00037 0.0043 0.0058 0.0048 0.016 Time trend (0.0054) (0.0055) (0.0059) (0.0077) (0.012) 0.00053** 0.00067*** 0.00060** 0.00050 -0.000073 (Time trend)^2 (0.00025) (0.00025) (0.00025) (0.00032) (0.00045) -0.090*** Log(RISE EE Score) (0.017) -0.081*** Log(RISE EE Scores)t-1 -0.019 -0.070*** Log(RISE EE Scores)t-3 -0.022 -0.075*** Log(RISE EE Scores)t-5 -0.028 N 319 319 300 262 224 R-Square 0.7202 0.7533 0.7759 0.8082 0.8441 Country FE Yes Yes Yes Yes Yes Year FE Yes Yes Yes Yes Yes Dependent Var. = Log Efficiency Index Note: CDD = cooling degree day; CDD18 uses 18 degrees Celsius as the temperature threshold for counting degree days. For example, a day with average temperature of 19 degrees Celsius counts as one degree day, a day with average temperature of 20 degrees Celsius counts as 2 degree days, and a day with average temperature below 18 degrees Celsius counts as zero degree days. HDD = heating degree day; HDD16 uses 16 degrees Celsius as the temperature Appendix threshold. EE = energy efficiency; FE = Fixed Effects; I/K = investment-capital ratio; K/L = capital-labor ratio; RISE = Regulatory Indicators for Sustainable Energy. Table A.9. Relationship between energy prices, regulatory policies, and economic activity index 91 (1) (2) (3) (4) (5) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean 0.043*** 0.043*** 0.037** 0.027* 0.028** Log(energy price) (0.015) (0.015) (0.014) (0.015) (0.014) 0.11 0.12 0.17* 0.31** 0.53*** Log(K/L) (0.088) (0.086) (0.098) (0.12) (0.15) -0.0055 -0.0061 -0.0083* -0.014** -0.023*** Log(K/L)^2 (0.0041) (0.0040) (0.0045) (0.0056) (0.0070) -0.053*** -0.055*** -0.052*** -0.020 0.018 Log(I/K) (0.017) (0.018) (0.018) (0.018) (0.018) -0.02 -0.02 -0.02 -0.02 -0.00 Population growth rate (1.62) (1.63) (1.62) (1.62) (1.76) 0.039 0.040 -0.0072 -0.030 -0.083** Log(CDD18) (0.046) (0.047) (0.045) (0.046) (0.037) 0.0040 0.0045 0.0017 0.0026 -0.0026 Log(HDD16) (0.0094) (0.0095) (0.0089) (0.0086) (0.0069) -0.013*** -0.013*** -0.013*** -0.016*** -0.023*** Time trend (0.0022) (0.0021) (0.0027) (0.0038) (0.0052) 0.00041*** 0.00039*** 0.00039*** 0.00055*** 0.00085*** (Time trend)^2 (0.000094) (0.000097) (0.00012) (0.00016) (0.00021) 0.0097* Log(RISE EE Score) (0.0056) 0.0089 Log(RISE EE Scores)t-1 (0.0056) 0.00055 Log(RISE EE Scores)t-3 (0.0076) -0.0027 Log(RISE EE Scores)t-5 (0.0089) N 319 319 300 262 224 R-Square 0.7483 0.7505 0.7806 0.8286 0.8800 Country FE Yes Yes Yes Yes Yes Year FE Yes Yes Yes Yes Yes Dependent Var. = Log Activity Index Note: CDD = cooling degree day; CDD18 uses 18 degrees Celsius as the temperature threshold for counting degree days. For example, a day with average temperature of 19 degrees Celsius counts as one degree day, a day with average temperature of 20 degrees Celsius counts as 2 degree days, and a day with average temperature below 18 degrees Celsius counts as zero degree days. HDD = heating degree day; HDD16 uses 16 degrees Celsius as the temperature Appendix threshold. EE = energy efficiency; FE = Fixed Effects; I/K = investment-capital ratio; K/L = capital-labor ratio; RISE = Regulatory Indicators for Sustainable Energy. Table A.10. The role of national planning (RISE score) 92 Log(Energy intensity) Log(Efficiency index) Log(Composition index) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean -0.17*** -0.21*** 0.043*** Log(price) (0.041) (0.043) (0.015) -0.029 -0.12 0.096 Log(K/L) (0.24) (0.28) (0.092) 0.0011 0.0058 -0.0047 Log(K/L)^2 (0.011) (0.013) (0.0043) -0.016 0.037 -0.053*** Log(I/K) (0.031) (0.036) (0.018) 15.9*** 17.8*** -1.85 Population growth rate (5.06) (5.47) (1.63) -0.12 -0.16 0.037 Log(CDD18) (0.082) (0.098) (0.047) -0.047 -0.051 0.0032 Log(HDD16) (0.032) (0.035) (0.0096) -0.015*** -0.0025 -0.013*** Time trend (0.0049) (0.0055) (0.0022) 0.0012*** 0.00077*** 0.00043*** (Time trend)^2 (0.00025) (0.00028) (0.000099) ln(National planning -0.016*** -0.015*** -0.0013 scores) (0.0051) (0.0056) (0.0019) N 319 319 319 R-Square 0.7240 0.7271 0.7486 Country FE Yes Yes Yes Year FE Yes Yes Yes Note: CDD = cooling degree day; CDD18 uses 18 degrees Celsius as the temperature threshold for counting degree days. For example, a day with average temperature of 19 degrees Celsius counts as one degree day, a day with average temperature of 20 degrees Celsius counts as 2 degree days, and a day with average temperature below 18 degrees Celsius counts as zero degree days. HDD = heating degree day; HDD16 uses 16 degrees Celsius as the temperature threshold. EE = energy efficiency; FE = Fixed Effects; I/K = investment-capital ratio; K/L = capital-labor ratio; RISE = Appendix Regulatory Indicators for Sustainable Energy. Table A.11. The role of entities responsible for promoting energy efficiency (RISE score) 93 Log(Energy intensity) Log(Efficiency index) Log(Composition index) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean -0.15*** -0.20*** 0.044*** Log(energy price) (0.041) (0.042) (0.015) 0.16 0.050 0.11 Log(K/L) (0.22) (0.26) (0.087) -0.0077 -0.0024 -0.0053 Log(K/L)^2 (0.0100) (0.012) (0.0040) -0.0046 0.047 -0.052*** Log(I/K) (0.031) (0.037) (0.018) 15.8*** 17.6*** -1.83 Population growth rate (4.96) (5.42) (1.63) -0.11 -0.15 0.037 Log(CDD18) (0.080) (0.098) (0.046) -0.038 -0.042 0.0039 Log(HDD16) (0.030) (0.033) (0.0095) -0.0095* 0.0027 -0.012*** Time trend (0.0049) (0.0055) (0.0022) 0.00088*** 0.00048* 0.00040*** (Time trend)^2 (0.00023) (0.00025) (0.000093) -0.015*** -0.014** -0.0016 ln(EE entities scores) (0.0051) (0.0055) (0.0019) N 319 319 319 R-Square 0.7248 0.7272 0.7489 Country FE Yes Yes Yes Year FE Yes Yes Yes Note: CDD = cooling degree day; CDD18 uses 18 degrees Celsius as the temperature threshold for counting degree days. For example, a day with average temperature of 19 degrees Celsius counts as one degree day, a day with average temperature of 20 degrees Celsius counts as 2 degree days, and a day with average temperature below 18 degrees Celsius counts as zero degree days. HDD = heating degree day; HDD16 uses 16 degrees Celsius as the temperature Appendix threshold. EE = energy efficiency; FE = Fixed Effects; I/K = investment-capital ratio; K/L = capital-labor ratio; RISE = Regulatory Indicators for Sustainable Energy. Table A.12. The role of financing mechanisms (RISE score) 94 Log(Energy Intensity) Log(Efficiency index) Log(Composition index) Realizing the Potential of Energy Efficiency in Latin America and the Caribbean -0.15*** -0.19*** 0.042*** Log(energy price) (0.043) (0.044) (0.015) 0.35 0.26 0.086 Log(K/L) (0.22) (0.25) (0.089) -0.017* -0.013 -0.0043 Log(K/L)^2 (0.0099) (0.011) (0.0041) -0.045 0.0048 -0.050*** Log(I/K) (0.033) (0.037) (0.018) 14.9*** 17.0*** -2.04 Population growth rate (4.95) (5.41) (1.63) -0.089 -0.13 0.038 Log(CDD18) (0.084) (0.099) (0.047) -0.034 -0.037 0.0033 Log(HDD16) (0.032) (0.035) (0.0095) -0.010** 0.0027 -0.013*** Time trend (0.0050) (0.0055) (0.0021) 0.00084*** 0.00041 0.00042*** (Time trend)^2 (0.00024) (0.00026) (0.000091) Log(Financial -0.018*** -0.021*** 0.0031 mechanism scores) (0.0063) (0.0068) (0.0027) N 319 319 319 R-Square 0.7227 0.7299 0.7495 Country FE Yes Yes Yes Year FE Yes Yes Yes Note: CDD = cooling degree day; CDD18 uses 18 degrees Celsius as the temperature threshold for counting degree days. For example, a day with average temperature of 19 degrees Celsius counts as one degree day, a day with average temperature of 20 degrees Celsius counts as 2 degree days, and a day with average temperature below 18 degrees Celsius counts as zero degree days. HDD = heating degree day; HDD16 uses 16 degrees Celsius as the temperature threshold. EE = energy efficiency; FE = Fixed Effects; I/K = investment-capital ratio; K/L = capital-labor ratio; RISE = Appendix Regulatory Indicators for Sustainable Energy. A.3. Key questions informing EE3: Incentives and mandates—industrial and 95 RISE energy efficiency scores commercial end users used for this report21 Realizing the Potential of Energy Efficiency in Latin America and the Caribbean 1. Are there any of the following EE mandates for large en- ergy users? EE1: National energy efficiency planning • Targets (e.g., kilowatt-hour savings or lower energy in- 1. Is there legislation or a national action plan that aims to tensity or carbon dioxide reductions, etc.) increase energy efficiency? • Mandatory audits • Energy management system (computer technologies 2. Is there an EE goal or target at the national level? to optimize energy use) • Energy manager in the facility 3. Are there targets defined for any of the following sectors? 2. Are there penalties in place for noncompliance with EE • Residential programs for large energy users? • Commercial services • Transport 3. Is there a requirement for periodic reporting of energy • Industrial consumption in order to enforce and/or track progress of • Power energy efficiency in large consumers’ facilities? 4. Are targets derived from detailed analysis that is publicly 4. Is there a measurement and verification program in available? place? 5. Is there a requirement for periodic progress reports 5. Is there a program to publicly recognize end users who tracking data related to the efficiency target(s)? have achieved significant energy savings measures? 6. Are there awareness programs or publicized case study EE2: Energy efficiency entities examples of significant energy savings measures? 1. Are there governmental and/or independent bodies that 7. Does the program offer technical assistance (from a gov- carry out the roles listed below ernment or independent entity) to end users to identify energy savings investment opportunities? • Setting EE strategy • Setting EE standards 8. Is there an EE mandate or incentive program for small • Regulating EE activities of energy consumers and medium enterprises? • Certifying compliance with equipment EE standards • Certifying compliance with building EE standards • Selecting and/or approving third-party auditors EE4: Incentives and mandates—public sector tasked with certifying EE standards 1. Are there binding energy savings obligations for public 2. Are EE programs developed based on market analysis buildings and/or other public facilities (may include wa- with plans open to public consultation and periodic eval- ter supply, wastewater services, municipal solid waste, uation? street lighting, transportation, and heat supply)? 3. Are there professional certification/accreditation pro- 2. Is there a reporting mechanism to track and enforce en- grams mandated for EE activities? Select all that apply: ergy savings in public sector facilities (either in-house or by a third party)? • Energy auditing/energy management • Monitoring and verification of energy consumption/ 3. Are there specific policies or mandated guidelines for savings public procurement of energy-efficient products and ser- • Building EE construction/design vices at the following levels? • Other • National • Region/state/province • Municipal/city/county Appendix 21 https://rise.esmap.org/indicators. 4. Are there guidelines or tools to help identify energy-effi- 7. Do customers receive a bill or report that compares 96 cient options for procurement (e.g., EE calculators, tech- them to other users in the same region and/or usage nical specifications, product rating catalogs)? class? Tick all that apply: Realizing the Potential of Energy Efficiency in Latin America and the Caribbean 5. Do public budgeting regulations and practices allow pub- • Residential lic entities to retain energy savings at the following lev- • Commercial els? Tick all applicable levels: • Industrial • National 8. Do customers receive a bill or report that shows their • Region/state/province energy usage compared to previous bills or reports over • Municipal/city/county time? Tick all that apply: • Residential EE5: Incentives and mandates—utilities • Commercial • Industrial 1. Generation 9. Which of the following charges do electricity customers • Are utilities required to carry out EE activities in this pay in the commercial services sector and in the indus- area? trial sector? • Are there penalties in place for noncompliance with EE requirements? • Commercial services sector • Demand (kilowatt) 2. Transmission and distribution networks • Reactive power (kVAr) • Industrial sector • Are utilities required to carry out EE activities in this • Demand (kilowatt) area? • Reactive power (kVAr) • Are there penalties in place for noncompliance with EE requirements? EE6: Financing mechanisms for energy 3. Demand-side management/demand-response efficiency • Are utilities required to carry out EE activities in this area? 1. Are any of the following financing mechanisms for EE ac- • Are there penalties in place for noncompliance with tivities available in the (R) residential sector, (C) commer- EE requirements? cial services sector, or (I) industrial sector? 4. Are any of the following mechanisms available for util- • Discounted “green” mortgages ities to recover costs associated with or revenue lost • On-bill financing/repayment from mandated EE activities: • Credit lines and/or revolving funds with banks for EE activities • Public budget financing • Energy services agreements (pay-for-performance • Consumer surcharge contracts) • Decoupling • Green or EE bonds • Vendor credit and/or leasing for EE activities 5. Are electricity tariffs cost reflective? • Partial risk guarantees • Other 6. Are any of the following time-of-use rate structures ap- plied to the residential, commercial services, or industrial 2. How many financial and/or nonfinancial institutions offer sectors? financial products for EE investments in each sector? • Real-time pricing • Residential • Variable peak pricing • Commercial • Critical peak pricing • Industrial • Seasonal rate • Peak-time rebates and/or time of day tariffs EE7: Minimum energy efficiency performance Appendix standards 1. Have minimum energy performance standards been ad- 2. Compliance system 97 opted for: • Is commission testing for energy efficiency required Realizing the Potential of Energy Efficiency in Latin America and the Caribbean • Refrigerators for final building acceptance documentation? • Heating, ventilation and/or air conditioning (HVAC) • Is there a requirement for periodic reporting to verify • Lighting equipment compliance with building EE requirements? • Industrial electric motors • Is verification carried out by a third party? • Other industrial equipment and/or domestic appli- ances 3. Renovated buildings • Light vehicles • Are renovated buildings required to meet a building 2. Verification and penalties for noncompliance: energy code, in residential and commercial sectors? • Residential sector • Are the standards mandatory? • Commercial sector • Is there a requirement for periodic reporting to verify • Are the building EE standards required to be updated compliance with standards? on a regular basis to reflect technological advances • Is the verification of compliance with standards car- and changes in best practices for building energy ef- ried out by a third party? ficiency? • Is there a penalty for noncompliance with EE stan- • Residential sector dards? • Commercial sector • Is there a periodic update of standards to reflect tech- nological advances and changes in best practices for 4. Building energy information EE standards? • Is there a mandatory standardized rating or labeling system for the energy performance of existing build- EE8: Energy labeling systems ings? • Are commercial and residential buildings required to 1. Have EE labeling schemes been adopted for? disclose property energy usage at the point of sale or when leased? • Refrigerators • Are large commercial and residential buildings re- • HVAC quired to disclose property energy usage annually? • Lighting equipment • Industrial electric motors 5. Building EE incentives • Other industrial equipment and/or domestic appli- ances • Are there mandates or targets for new buildings to • Transport vehicles achieve high-quality EE certifications, such as Lead- ership in Energy & Environmental Design (LEED) (e.g., 2. Mandatory vs voluntary labeling system percentage of new building stock that must be LEED certified)? • Are any of the above labeling schemes mandatory? • Is there a periodic update of standards to reflect tech- nological advances and changes in best practices for EE labels? EE9: Building energy codes 1. New residential and commercial buildings • Are there EE codes for new residential buildings? • Are there EE codes for new commercial buildings? • Are the building EE standards required to be updated on a regular basis to reflect technological advances and changes in best practices for building energy ef- ficiency? • Residential sector Appendix • Commercial sector 98 A.4. Types of energy efficiency policies and regulations in LAC Realizing the Potential of Energy Efficiency in Latin America and the Caribbean Figure A.27 Building energy codes Incentives & mandates: Industrial and Commercial End users Energy efficiency policies Energy labeling systems and MEPS National energy efficiency planning and regulations of various Entities Incentives & mandates: Public sector types across the LAC region Financing mechanisms Others/ information / education Mexico Brazil Argentina Chile Peru Colombia Rest of Central America Rest of South America Caribbean Source: IEA, BIEE, and RISE databases. Appendix MEPS = minimum energy performance standard. Realizing the Potential of Energy Efficiency in Latin America and the Caribbean responds to the urgent need to relaunch the energy efficiency agenda on the continent in the context of post-COVID recovery, the challenges of climate change mitigation, and high energy prices. The report assesses the current state of energy efficiency policies and measures in the region, identifies key challenges and drivers for improve- ment, and proposes ways forward. Starting with a high-level review of how energy efficiency policies have evolved in the region over the past two decades, the report proceeds to disaggregate broad trends into discrete efficiency improve- ments and changes in economic activity. It then identifies key drivers of energy intensity reductions at the sector level. Finally, it provides recommendations on policy instruments to support energy efficiency improvements.