E Q U I TA B L E G R O W T H , F I N A N C E & I N S T I T U T I O N S N OT E S World Bank Group Macroeconomic Models for Climate Policy Analysis January 2022 © 2022 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of the staff of The World Bank with external contributions. The findings, 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 accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. 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. “World Bank Group Macroeconomic Models for Climate Policy Analysis.” EFI Note. 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 reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse 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. >>> Contents 1. Greening the WBG’s Portfolio of Macroeconomic Models 5 2. Unresolved Problems when Using a Macro Model to 8 Analyze Climate Change 2.1. The Aggregation of Local Events and Activities 8 Tends to Mute Some Effects 2.2. The Uncertainty of Global Climate Change and 9 its Country Impacts 3. Issues for Policy Makers to Consider when Using Models 9 4. Annex: Links to Model Documentation 10 >>> World Bank Group Macroeconomic Models for Climate Policy Analysis Policymakers in developing countries face multiple challenges related to climate change, among them: assessing the impact of climate change on economic growth and living standards; identifying a mix of policies and investments that can make countries more climate-resilient; and developing credible options for mitigation compatible with planetary limits and their own development objectives. Over the past decades, the World Bank Group has developed a range of macroeconomic models for projections, diagnostics, and policy impact simulations. Since 2019, the WBG has accelerated its efforts to make its modeling portfolio climate-aware. This note, prepared by staff from DEC, EFI and SD, provides an overview of the climate aspects of these macroeconomic models, as well as complementary tools to investigate the links with poverty, inequality, or financial stability. It illustrates how these tools interact with each other, and how they are applied in operations. GREENING >>> THE WBG’S PORTFOLIO OF MACROECONOMIC MODELS To provide policymakers with reliable recommendations on a variety of climate related policies, the WBG has a diverse and complementary set of models. The analytics range from evaluating the aggregate, sectoral, and welfare effects of mitigation measures to assessing country-specific adaptation needs, considering the impacts of extreme weather events as well as gradual global warming. Key indicators include macroeconomic outcomes (GDP, consumption, inflation, exchange rates, fiscal impacts, debt), sectoral indicators (energy prices, industrial, agricultural and service sector output), co-benefits (health and productivity improvements from reduced air pollution, and tax efficiency gains when carbon tax revenues allow a distorting tax to be removed), and poverty and distributional issues (e.g., by income group, geographical region and skill types). Table 1 summarizes the range of climate and development issues addressed by each model in the WBG suite, revealing both strengths and limitations of individual models, as well as the complementarity among models. WORLD BANK GROUP MACROECONOMIC MODELS FOR CLIMATE POLICY ANALYSIS <<< 5 > > > T A B L E 1 - Scope of WBG Climate Informed Modeling Approaches SHOCK WAVES/ MODELS MFMOD ENVISAGE MANAGE CPAT FSAP GIDD UNBREAKABLE Impacts of mitigation/adaptation policy on Macroeconomic indicators  ü ü ü Sectoral indicators ü ü ü ü ü ü ü Co-benefits ü ü ü Poverty/distributional impacts ü~ ü~ ü ü ü Impact of extreme weather events on Macroeconomic indicators ü ü Sectoral indicators ü~ ü Financial indicators ü~ ü Poverty/distributional impacts ü ü Impact of gradual global warming on Macroeconomic indicators ü ü ü Sectoral indicators ü~ ü ü ü ü Financial indicators ü ü Poverty/distributional issues ü~ ü~ ü ü Source: World Bank staff; Note: the combined symbol ü~ implies the issue is partially addressed in the framework. These models can be used alone, as well as in conjunction country-specific and are estimated econometrically with other tools and models. For instance, macroeconomic using historical data. MFMod is suitable for forecasting model can include reduced-form simplified versions of highly and can be used to simulate a range of climate and complex sectoral models (e.g., crop models or catastrophe policy scenarios. Models cover greenhouse gas (GHG) models) or be calibrated based on results from sectoral emissions from five sources and economic damages analysis. Similarly, outputs from macroeconomic models can from climate change derived from the literature that be used directly to inform decision-makers or be used as include physical damage from extreme weather events inputs for further analysis (e.g., on distributional impacts or and the impacts of higher temperatures and increased financial stability). For instance, climate-enhanced FSAPs rain variability on economic activity (e.g., effects on (Financial Sector Assessment Program) assess risks, including competitiveness, say of tourism, reduced labor and from climate, and opportunities to improve regulatory and agricultural productivity, and declines in health and supervision practices, using a framework combining sectoral labor supply). The framework also incorporates co- models, macroeconomic models, and financial-sector models. benefits from mitigation (e.g., reduced pollution, resulting in improved health outcomes, lower health The main features of these models are the following: spending, and increased labor productivity and supply) and interactions with other country-level externalities, • Macro-Fiscal Model (MFMod) is a customizable such as those coming from excess informality or the macro-structural country-level modelling framework, elimination of tax distortions. The standard model including expenditure and sectoral national income includes a basic adaptation module that can be accounts, financial and current accounts of the balance supplemented with country-specific data and estimates. of payments, labor markets, inflation, exchange rates MFMod is particularly handy for exploring the dynamics and monetary policy. The framework has a mix of after economic shocks (e.g., natural disasters or empirically determined short-run behaviors and a theory- material price changes). It is easy to use, build and informed long-run effects. Parameters of the model are maintain and can be paired with a microsimulation WORLD BANK GROUP MACROECONOMIC MODELS FOR 6 >>> CLIMATE POLICY ANALYSIS module such as the GIDD system described below to specialized training. CPAT can be linked to a Multi- explore distributional issues, as well as with sectoral Regional Input-Output (MRIO) model, which is a models (e.g., EPM, MESSAGEix and GTAP-BIO-W). framework for simulating the effects of climate-fiscal policy shocks on domestic and international sectoral • ENVISAGE and MANAGE are state-of-the-art outputs and employment. CPAT features a large set of dynamic-recursive Computable General Equilibrium market imperfections and is estimated using Bayesian (CGE) models. ENVISAGE is a global model, covering statistical techniques. 127 countries and aggregating the remaining into 20 regions. ENVISAGE’s strengths lie in tracing the impacts • ARIO—Adaptive Regional Input-Output model—is a of policies imposed in one country on the economies of short-term framework that can be run at daily and weekly other, and global or regional scenarios. MANAGE is a frequencies to model how firm-level dependencies highly customizable country-level modelling framework and transportation networks can propagate localized and includes a more extensive modelling of both climate economic disruptions such as floods or wind damage damages and adaptation. Both models are calibrated to other parts of the economy via the supply chain and on data at a single point in time, using elasticities drawn transportation networks, generating knock on effects from the literature and, where available, supplemented potentially much larger than the first round impacts with country-specific estimates. Their strengths lie in typically accounted for. The model has been used to their micro-foundations and sectoral detail, which in represent the impacts of hurricanes (e.g. in the US and some cases includes 50 or more. The models cover India), earthquakes (e.g., in the US, Japan, and China), most of the same physical damages and mitigation and and COVID-19 (globally, with an application to the adaptation policies described for the MFMod system. impact on GHG emissions). The greater sectoral detail means they can provide a more disaggregated accounting of losers and winners To extend the exploration of the poverty and inequality, from economic damages and climate policies than the these macroeconomic models can be complemented with MFmod system. The standard single household can bottom-up household-level approaches or their outputs can be be disaggregated into different (urban rural/ high-low connected to microsimulation models to disaggregate results skilled; deciles) groups or a model can be paired with at the household levels. Available tools include: a microsimulation mode to analyze distribution issues. Dynamics in both models are relatively simple and • Global Income Distribution Dynamics (GIDD) is a focus on equilibrium impacts after economies have microsimulation tool used to assess the distributional time to adjust; therefore, they do not capture well the impact of structural changes brought on by large shocks. out-of-equilibrium behavior of economies following The model has been used at the country, regional and a shock and are less well suited than MFMod for global levels to assess a wide range of issues, including analyzing the impacts of extreme weather events. Both the impact of climate change, demographic changes, systems can be paired with more detailed sectoral trade reforms, and the expansion of biofuels, among models to strengthen the micro basis for simulations. others. The microsimulation component takes output, Both systems are technically demanding and typically employment, wage and price results of the MANAGE/ require a specialist to operate. CGE or MFMod models to generate income distributions over time consistent with the macro forecasts. GIDD • Carbon Pricing Assessment Tool (CPAT), a joint World allows for projections of poverty and inequality and Bank - IMF product, is a reduced form model designed distributional impacts across household types, regions, to assess the first-order climate-macro implications of and vulnerable categories. environment tax reform. Policy designs can be selected at a granular level due to a high flexibility in making choices regarding carbon pricing instruments, sectoral coverage, subsidy reform, and revenue recycling. The tool covers a wide range of climate-macro measures: emissions, energy consumption, public revenues, GDP, equity, air pollution, health costs, and transport externalities. Its tractability allows its use without WORLD BANK GROUP MACROECONOMIC MODELS FOR CLIMATE POLICY ANALYSIS <<< 7 • SHOCK WAVES estimates the impacts of the environmental and social impacts of projects and policies. macroeconomic transformations and climate shocks on The Emissions Intensity and Trade Exposure (EITE) Country household incomes, including poverty and inequality Comparison tool provides a historical snapshot of the relative by modelling the poverty impacts of gradual global emissions intensity and trade exposure of key products warming and natural disasters and shocks, in a bottom- to identify the sectors facing high transition risks. Finally, up way. It creates baselines for the future evolution of the Electricity Planning Model (EPM) identifies least-cost poverty and other distributional metrics to represent electricity mixes based on climate ambitions. the impact on policy choices regarding growth, and redistribution, and estimate climate change impacts at > > > U N R E S O LV E D P R O B L E M S W H E N the household level using results from sectoral models U S I N G A M A C R O M O D E L T O A N A LY Z E for five channels: food prices, farmer incomes, health C L I M AT E C H A N G E impacts such as diarrhea, malaria and stunting, labor productivity effects, and natural disasters such as To provide the best possible guidance to client countries, storms, floods, droughts. The model is short run with the WBG is working to improve compatibility across and a time horizon at 2030 and aims to inform adaptation communication between models, building on their individual priorities. It helps to understand future socioeconomic strengths. For instance, MFMod and MANAGE results and climate evolutions with due consideration of can be used as inputs to CPAT or the Equity Policy Lab’s uncertainty. Its objective is not to forecast, but to identify microsimulation tool, and vice versa. Strengthening inter- the main vulnerabilities and the most effective policies model compatibility between sectoral models such as energy, to dampen climate change impacts. water, agricultural and financial model also ongoing and central to improving economic insights and granularity. In • UNBREAKABLE estimates the impacts of acute parallel, many open questions remain in the field of macro covariate shocks including natural disasters and layoffs climate modelling. on household consumption and welfare and provides estimates of “socioeconomic resilience” for countries or >>> THE AGGREGATION OF LOCAL EVENTS AND regions. This is a proxy for the ability of the population ACTIVITIES TENDS TO MUTE SOME EFFECTS to cope with and recover from shocks, based on individual characteristics like access to finance, but also • Disasters or policy actions (e.g., coal mine closures) government capacity like ability to finance, contingent often have impacts that are concentrated on small planning, and adapt social protection. It identifies ex ante regions (e.g., New Orleans in the US). Their local effects risk factors, tracks differential impacts, and assesses are difficult to discern in the data when aggregated to the long-term benefits of risk reduction and resilience the country level. strategies. Where data are available, the model can • In models with representative households, savings and be run at the national or regional or provincial scale, borrowing constraints of the households immediately accounting for spatial heterogeneity in natural, social affected by a disaster may be underestimated and economic conditions. and hence the speed of post-shock adjustment overestimated by lumping unaffected households with This overview is not an exhaustive stocktaking of WBG affected households. modeling approaches because it is restricted to macroeconomic • Similarly, the impact of sudden shocks that affect a and distributional models deployed across multiple country specific component of the supply chain may be under- and regional practices. Other standalone, country-specific estimated in cases where the shocked sector albeit models have been developed over time for specific research small economically, has few substitutes and is an questions related to climate change. Other adaptable analytic important input to other larger sectors. frameworks cover a wide range of regulatory policies. For • Finally while adjustment speeds in models reflect instance, the Climate Change Institutional Assessment (CCIA) empirical realities, narrow shocks that require that identifies the strengths and weaknesses of the institutional generate shifts in labor demand in terms of tasks, skills, framework for addressing climate change. The Social and spatial distribution may imply bigger disruptions Protection System Stress Test assesses the adaptiveness of than the historical averages incorporated into the social protection systems focusing on their ability to respond models. to climate and other covariate shocks. National Environmental and Social Impact Assessment (ESIA) systems categorize WORLD BANK GROUP MACROECONOMIC MODELS FOR 8 >>> CLIMATE POLICY ANALYSIS Theoretically, a fully articulated macroeconomic model policy is actually implemented. Notably, they struggle could be constructed to include the rich detail of households, to represent regulatory policies that do not rely on firms, geography and technological change. In practice such economic incentives (e.g., technology mandates) or a model would be too large and impossible to solve, and that are intended to promote technological innovation expensive to maintain. Instead work on these issues focuses or scale economies. on the coupling of macroeconomic models with more granular models at the sectoral or regional level, and the combination Given the intractable nature of these uncertainties, including of top-down and bottom-up models that represent individual the extent to which policy may affect productivity, a scenario households or firms. Initial work has focused on soft-links such approach is typically followed. And while scenarios can cover that results are coherent between the systems. Future work at a wider range of potential outcomes, they do not normally the WBG will deepen the interactions between systems. include the kind of extreme negative outcomes that might result in climate or economic tipping points. As a result, even >>> THE UNCERTAINTY OF GLOBAL CLIMATE probabilistic scenarios likely do not include the worst-case CHANGE AND ITS COUNTRY IMPACTS scenarios and therefore may, when viewed as an ensemble, be biased to the upside. Work is underway to improve our The future path of global carbon emissions, the global understanding of these uncertainties and the mechanisms and local climates, economic policy and technology are behind tipping points and endogenous technical change, but it all inherently uncertain and yet each will have potentially is likely to be some time before results are achieved and these enormous impacts on the macroeconomics of climate change. can be fully incorporated into macro models. • Climate modelers deal with the first of these uncertainties > > > I S S U E S F O R P O L I C Y M A K E R S T O through scenario analysis, providing a series of climate CONSIDER WHEN USING MODELS and societal pathways based on different assumptions about these outturns. The macro models follow suit by Model simulations are one of many inputs into the adopting one or more of these well-defined scenarios policymaking process. Some issues to bear in mind: when simulating future climate outcomes. But there are no guarantees that any of the pathways will be • Each of these models is an abstraction of reality, and achieved. none covers the full range of issues inherent to the • Technological change, notably change that affects either interaction of climate and macroeconomics. Critically the use of energy (e.g., the spread of electric vehicles important factors like the full scope of potential climate or energy-efficient lightbulbs) or the cost of producing outcomes including tipping points, or the pace of renewable energy (e.g., PV cell cost reductions or technological change under different policy regimes are hydrogen storage technologies), can completely fundamentally unknowable and can only be addressed change the relative prices upon which energy decisions through scenario analysis, or the interactions of on both the consumer and producer side are made, with unspecified economic distortions and climate change major ramifications for economic projections. Again, are not dealt with and likely bias results. in the absence of new scientific analyses about how • Model results expressed as percent of GDP or technology will evolve, this uncertainty must be dealt consumption or income reflect part of the story. Social with via scenario analysis. welfare includes issues many other criteria including • A third source of uncertainty concerns the reaction of economic stability, distributional equity, and of course economies to extreme events. While models deal with the pathway to an end result matters. Two pathways the normal day-to-day interactions of multiple shocks, may end up at the same level of GDP and carbon there is limited empirical information for dealing with emissions, but if one is smooth and associated with extreme events and tail risks. As a result, scenarios limited disruption it may well be preferred to one that including probabilistic simulation strategies will under- is bumpy and associated with extended periods of estimate the full range of possible outcomes unless unemployment and macroeconomic instability. While they are explicitly laid out. models can incorporate some sort of social welfare • Finally, while models can help us judge the likely function, ultimately the choice of pathway is an efficacy of some policies (notably price-based policies inherently political one. for which they have been designed), they are inevitably • The choice of model and approach should be driven limited in their ability to represent peculiarities in how a by the questions that policymakers want to answer. WORLD BANK GROUP MACROECONOMIC MODELS FOR CLIMATE POLICY ANALYSIS <<< 9 So, for example, the MFMod framework has limited >>> ANNEX: LINKS TO MODEL sectoral disaggregation as compared with MANAGE D O C U M E N TAT I O N but does a much better job of simulating the path of an economy out of equilibrium following a shock and the As work extends the climate features incorporated into possibilities and implications of economic instability. these models, the references below, while reflecting the most The CGE platforms are best for looking at long-term recent publicly available documentation for the various models policy issues, or physical transitions that may require necessarily may not reflect fully their current state. substantial structural changes, or when the focus is on finer dis-aggregations of economic activity. The CPAT MFMOD (climate enhanced) : https://openknowledge.world- framework is relatively easy to use and perhaps best bank.org/handle/10986/36307 suited to looking at mitigation issues over the near ENVISAGE: https://mygeohub.org/groups/gtap/File:/uploads/ term; like most models, it does not do well as forecast- ENVISAGE10.01_Documentation.pdf ing horizons lengthen. MANAGE: https://mygeohub.org/groups/gtap/File:/uploads/ MANAGERef.pdf • Finally, no single model can address all the policy-rel- GIDD: https://web.worldbank.org/archive/website01589/ evant questions that arise. In many cases, employing WEB/IMAGES/PARAL-31.PDF more than one model, or using a combination of macro and sectoral models is the best approach. WORLD BANK GROUP MACROECONOMIC MODELS FOR 10 >>> CLIMATE POLICY ANALYSIS