Semi-structural macroeconomic models (or macrostructural models) are useful frameworks that support economic forecasts and counterfactual policy and economic scenario analysis. The macro modeling team provides macroeconomic modeling services to internal and external World Bank clients. The team comprises economists and data specialists and is charged with building and managing the Global Practice’s main macro-structural model MFMod, a system of 186 linked country models, which are used to produce the World Bank report: Macro Poverty Outlook. The team is also deeply engaged in producing stand-alone country-specific macro-structural models for internal and external clients. This work ranges from simple extensions of the core MFMod for country economist teams, to the elaboration of much more detailed models for external clients (typically Ministries of Economy or Finance and Central Banks). An increasing share of this work involves integrating climate considerations into the standard macroeconomic models. Work for external clients often involves delivering extensive training on the models.
Macro-Modeling At The World Bank
Macroeconometric Model: MFMod
-
The World Bank’s Macro-Fiscal Model (MFMod) is a structural econometric model (or macro-structural model) that reproduces the flow of funds across the whole economy by mapping out the main identities of the national accounts, balance of payments, labor markets, and fiscal sectors. Macro-structural models make a concerted effort to estimate the economic and behavioral determinants of economic variables. These structural relationships are developed to be both consistent with economic theory (in the long run) and the observed dynamics of the economy in the short term. As a result, the speed of adjustment of each country-specific model to its economically-determined long-term equilibrium depends mainly on the actual historical behavior of the economy.
In MFMod, the modeling of GDP comprises three standard measurements: GDP from the (i) expenditure side, (ii) production side, and (iii) income side. To include climate change considerations, modifications can be made to accommodate additional dimensions in each of the approaches. For example, energy can be disaggregated (into renewables and others) on both the production and consumption sides.
The climate change expansion of MFMod (MFMod-CC) was developed to incorporate the impacts of climate change. Initial versions of MFMod-CC developed for Pakistan and Jamaica drew from the existing climate literature to introduce emissions and pollution modules; damage functions from higher temperatures, pollution, and flooding on economic activity; and an adaptation module to analyze the economic benefits of adaptation investments to adjust to climate change. Since then, the approach has been substantially refined as MFMod-CC has been employed in over 40 countries as part of the World Bank's Country Climate and Development reports (CCDR) process. Combined with the standard features of the MFMod framework, the climate-augmented versions provide a vehicle for systematically evaluating the impact of climate change, and of both climate and non-climate policies. This includes climate-related outcomes (tons of carbon emitted, energy transitions, economic and health damages due to higher temperatures and pollution, the impact of climate shocks, etc.), as well as the traditional social and economic outcomes (growth, fiscal sustainability, inflation, and current account stability). The model can also quantify associated co-benefits from reduced pollution, better health and productivity outcomes, increased temperatures, and rain variability. Recent efforts (in the case of Uganda) include a mapping of aggregate economic activity to natural capital. This includes estimates of adjusted national savings, which account for resource depletion, the cost of carbon and pollution, as well as expenditures on education.
-
The World Bank’s Macro-Fiscal Model (MFMod) is a structural macro-econometric model that reproduces the flow of funds across the whole economy by mapping out the main identities of the national income accounts, balance of payments, labor markets, and fiscal accounts. The model consists of 186 linked country models. The size of an individual country model depends on data availability. A modular approach to the model build allows for the inclusion of countries with limited data availability. At the same time, countries with more extensive data coverage are equipped with a more comprehensive macroeconomic model, including modules on domestic oil sectors for oil-exporting countries, labor markets, the financial and capital accounts of the balance of payments (BoP), detailed fiscal accounts, and a more comprehensive module on greenhouse gas (GHG) emissions.
Several data sources are utilized for this purpose. For most of the high-income economies included in the MFMod, data are taken from the most recent Global Economic Prospects (GEP) report, published bi-annually by the World Bank. Global commodity prices are sourced from the most recent Commodity Markets Outlook, published on a bi-annual basis by the World Bank. Country models for oil-exporting economies include a model block on the oil sector with data being sourced from national statistical offices and the World Economic Outlook (WEO), provided by the International Monetary Fund (IMF).
Data for low-income and middle-income countries—the focus of the Macro Poverty Outlook (MPO)—are derived from several sources in close collaboration between country teams and the MPO central team. The main data source for national income accounts (NIA) data are series provided by national statistical offices. Most country models in MFMod build on expenditure accounts, as well as production accounts for GDP accounting. Countries with limited NIA data availability, around 10 percent of all countries in MFMod, include GDP production accounts only. National statistical office data are underpinned by global databases provided by the World Bank and the IMF. The World Development Indicators (WDI), provided by the World Bank, represent the secondary source for the NIA. Data provided by the WEO are utilized, too. Statistics on financial sectors, including interest rates, exchange rates, and data on monetary aggregates, are taken from national statistical offices, the WDI, the International Financial Statistics (IFS), provided by the IMF, and the WEO.
Balance of Payments (BoP) series are sourced from national statistical offices, the IMF’s Balance of Payments and International Investment Position Statistics (BoP/IIP), and the WEO. Data on remittances are taken from the World Bank’s Migration and Remittances Data. All individual country models include data on the current account balance. Around 80 percent of all country models also include data on financial and capital accounts of the BoP.
Individual country models are linked via exports and imports of goods and services and remittance flows. An export market growth variable links demand for a country’s exports of goods and services to import demand of its trading partners. Trading partners and their relative weights in the export market growth variable are derived from a trade matrix, calculated using the United Nations (UN) Comtrade database. The export market growth variable is then included in the behavioral equation for export demand. Furthermore, a remittance inflow variable is calculated based on income growth of remitter countries. A migrant stock matrix determines the weights of remitter countries for each receiving country. The matrix is derived from Ratha and Shaw (2007)1. The remittance inflow variable enters the behavioral equations for actual remittances. Finally, price deflators for exports and imports for each country model are based on weighted averages of global commodity prices of 32 internationally traded commodities. Weights are derived from a trade matrix based on UN Comtrade data. The derived deflators enter the behavioral equations for export and import prices of goods and services.
Detailed data on fiscal accounts are mostly provided by national statistical offices, the WDI, the WEO, and the IMF’s Government Finance Statistics (GFS). A small group of countries, four countries in total, only provide data on their General Government Balance. Other country models build on extensive data on government accounts, including detailed current expenditure accounts, capital expenditures, government revenues, external and domestic debt, as well as their financing for the most comprehensive country models.
Population statistics are provided by the Health, Nutrition and Population (HNP) Global Practice of the World Bank. These include historic data and forecasts for total population and working-age population. Labor market statistics are provided by the WDI, the International Labour Organization (ILO), and the Penn World Tables.
Finally, the greenhouse gas module includes emissions data for every country in the model, with a small group of three countries including data on CO2 emissions only. Data on greenhouse gas emissions are sourced from the World Resources Institute’s CAIT data tool and national statistical offices. Additional data on CO2 emissions and total energy supply are taken from the International Energy Agency (IEA) database on energy supply and indicators on CO2 emissions.
FEATURED
The World Bank Macro-Fiscal Model Technical Description: MFMod
Macro Poverty Outlook
Contact Us
Senior communications officer, World Bank
1818 H Street, NW Washington, DC 20433 USA