TopicsTopics tab icon

Stay Connected

Follow ERS on Twitter
Subscribe to RSS feeds
Subscribe to ERS e-Newsletters.aspx
Listen to ERS podcasts
Read ERS blogs at USDA

Documentation

The International Macroeconomic Data Set

This data set includes historical and projected annual data for real gross domestic product (GDP), population, real exchange rates, consumer price indices (CPIs) and GDP deflators for 189 countries, 37 regional aggregates, and 12 income-based aggregates of the world economy.  The data are all measured in or centered on real 2010 dollar values.  The data are organized by region in spreadsheets that are identical except for the variable name.  The historical data and projections are updated annually.

Characteristics of the Data

The ERS International Macroeconomic Data Set is the only publically available source of international time series information on real GDP, GDP deflators, CPI indices, and real exchange rates that is consistent and complete for all countries and variables included. 

Variables measured in constant 2010 dollars. The central variable in the data set is real GDP, measured in constant 2010 dollars, which is an appropriate measure for understanding differences across countries and changes over time in national purchasing capacity.  Measuring GDP in constant (real) value terms removes the effect of price changes from the series so that changes that occur are more likely to reflect changes in capacity to purchase given quantities of goods or services. Measuring GDP in a common currency (dollars), as opposed to local currencies, allows for meaningful cross-country comparisons.  Researchers typically use one of two methods to convert a local currency series into a common currency series.  In this data set, a common base year of market exchange rates is used to convert each country’s GDP into a common currency.  An alternative method uses purchasing power parity exchange rates, which adjust market exchange rates for domestic purchasing power.  This latter approach is more appropriate for comparing living standards around the world.  However, if one is trying to evaluate international purchasing capacity for trade, use of market exchange rates is viewed as a better alternative (Callen, 2007, OECD, 2007).

Treatment of missing values. The variables included in this data set are the most prominent and readily available macroeconomic data on countries around the world.  Still, sources of data for many of the countries included in the data set have substantial gaps.  For most statistical analyses, missing values can introduce discontinuities into the data, lead to biased results, and violate standard rules for statistical analysis.  The objective in filling in missing values in this data set is to retain series where estimation of central parameters is unchanged.  The rule applied in filling in data is to extrapolate missing values based on trends observed in the available data. This smoothing process reduces measures of variance from what they might otherwise be if complete data were available, but it permits development of a complete data set with which to estimate regional and world aggregates.1   Thus, the data for each country are complete: published without missing values for any year. 

How the Variables Are Derived

Real U.S. dollar GDP. Real GDP measures the size of an economy and the degree to which it evolves over time, holding prices constant.  Real GDP converted into a common currency, in this case the U.S. dollar, allows for cross-country comparison.  To derive real GDPs, current (nominal) GDP in the local currency is converted to constant GDP using the local currency GDP deflator: for countries  and over time , where  is the base year.  For the current series, 2010 is the base year,

Equation 1

 

Finally, the GDPs are converted to real dollar GDP using the base year exchange rate (XRib).

 

Equation 2

 

Deriving real GDP in this way, rather than simply adjusting dollar values of current GDP by the U.S. GDP deflator, has two advantages.  First, it does not implicitly assume that the United States and the foreign country have identical rates of inflation in every year.  Second, it holds the exchange rate used to convert to the dollar series constant so that the resulting series is in constant terms and does not reflect changes in exchange rates.  The difference between the “real $2010” series used in this data set and a series created using an “adjusted nominal” approach is illustrated in the figures below.  The adjusted nominal approach first converts nominal GDP in local currency to U.S. dollars and then uses the U.S. GDP deflator to adjust for changes in prices over time.



1 The regional and world totals are still limited to be summations of included countries.  While there may be as many as 30 excluded countries and territories due to data limitation, these omitted countries/territories account for less than 1 percent of world GDP.

 

 

Mexico's real and adjusted nominal GDP

 

Thailand's real and adjusted nominal GDP

 

 

The difference between the real GDP (real $2010) and the adjusted nominal GDP (adjusted nominal) is substantial for Mexico but much less so for Thailand.  Inflation in Mexico has historically been much higher than inflation in the United States, so using the U.S. GDP deflator instead of the Mexican GDP deflator results in a series that is very different from the real series based on the Mexico GDP deflator.  In contrast, the GDP deflator for Thailand is close to that of the United States, resulting in the two series being more similar. 

Price indices. Consumer price indices and GDP deflators indicate how prices change over time and are used as measures of inflation.  Both measures are derived from geometric weighted averages of price changes for bundles of goods produced and consumed in an economy.  The bundle of goods measured by a GDP deflator represents the country-specific components of GDP, whereas the CPI measures the price of a country-specific representative consumption bundle.  In this data set, both indices are set to a common base year (2010). The CPI series included in this data set is based on the CPI estimates reported in the World Bank’s World Development Indicators (WDI) and the International Monetary Fund’s (IMF) International Financial Statistics (IFS), rebased to 2010 as necessary and with missing values filled in using back estimation, interpolation, and moving average extrapolation as necessary. 

Real exchange rates. Real exchange rates describe changes in the value of a country’s currency holding prices constant.  They are calculated using data on nominal exchange rates and CPIs for the United States and each foreign country.  They are defined as:

 

Equation 3 


where  is the nominal dollar exchange rate in terms of country ’s local currency (LCi), and  and  are the CPI for country and the United States, respectively.  More specifically:



Equation 4 

which can be written as:



Equation 5 



The real exchange rate for country  can thus be seen as the ratio of the product of the local currency value and the U.S. CPI to the product of the value of one dollar and the local currency CPI.  A real appreciation (depreciation) of a currency represents an increase (decrease) in its relative value against another currency, in this case the U.S. dollar. Thus, an appreciation implies that less of that currency is needed to purchase another currency. Relative rates of inflation are critical for determining real exchange rates:  A higher rate of inflation in the local currency than in the U.S. dollar results in a real appreciation of that currency relative to the dollar—unless the nominal exchange rate depreciates to offset the difference.

How the Projections Are Made

The purpose of the projections included in ERS’s International Macroeconomic Data Set is to provide the long-term global macroeconomic scenario behind the 10-year projections for U.S. and global agricultural markets updated and released annually by USDA.  USDA macroeconomic projections are not made independently but are a composite scenario that draws on projections and forecasts made by other organizations and vendors.  The projections are informed primarily by analysis from Oxford Economic Forecasts (Oxford), IHS Economics and Country Risk (IHS), the IMF World Economic Outlook (IMF-WEO), and other country-specific sources.  U.S. projections also make use of U.S. Congressional Budget Office Economic Projections (CBO) and the White House Office of Management and Budget Mid-Term Review (OMB).  Long-term projections are made under the assumption that global policies are largely held constant.  Therefore, projections generally imply a smooth
future path. 

The ERS projections are restricted to be within the range of the other forecasts and projections.  On a qualitative basis, few differences are observed among the alternative projections for the major countries that account for most of world trade.  While significant differences can be observed for smaller and less developed countries, these differences do not have a major impact on world growth or world trade. For use in generating the USDA agricultural long-term projections, the macroeconomic outlook provided by the sources used is smoothed to remove the impact of abrupt changes due to business cycles or other shocks that may have been incorporated by the source.

Data Sources

The International Macroeconomic Data Set includes both historical series by country and region for seven indicators from 1969 to 2013 and projections for the same variables for 2014 to 2030.  Each variable is reported in both levels and growth rates.  In most cases, historical data are obtained in levels from the various sources, while projections are made in terms of growth rates that are also used to calculate projected levels.  Historical data on GDP come primarily from Census global population data. 

Real GDP. Historical GDP levels are obtained from the World Bank’s WDI for all countries except Taiwan.  Taiwan’s GDP is obtained from reports issued by the Taiwan government.  GDP data are reported in billion constant 2010 U.S. dollars.  For years where GDP is missing from the WDI, values are either calculated based on previously forecast growth rates or interpolated using data from surrounding years.  Projected GDP growth rates for each country are based on the range of projections provided in the published projections.  Regional aggregate GDP levels are the sum of the GDP values for each country in the region. 

GDP deflators. Historical GDP deflator values are derived from deflators reported in the WDI.  All deflators have been adjusted to a common base year of 2010.  For years in which the WDI data are missing, the deflator is interpolated from surrounding data or calculated using previously forecast percent changes.  U.S. GDP deflator values are projected by ERS analysts, and projected deflator growth rates for other countries are taken from IHS Global Insight projections.  Regional aggregate deflator levels are a GDP-weighted average value across countries in the region.

GDP shares. GDP shares in both the historical and the projected series are a ratio of each country’s GDP to world GDP.  Decade average shares are simple averages. 

Population. The historical and projected population series are both obtained from the U.S. Bureau of the Census International Database in levels, with growth rates calculated from these levels.  Regional aggregate population is the sum of the population over countries in the region.   

GDP per capita. GDP per capita is calculated by dividing GDP by population at either the country or regional aggregate level. GDP per capita is reported in constant 2010 U.S. dollars. 

Consumer price index (CPI). Historical CPI values are obtained from the IMF-IFS for all countries except Taiwan. Taiwan values are obtained from the Taiwan government.  The CPI base year is 2010 for all countries.  For countries included in the ERS Agricultural Exchange Rate Data Set, historical annual CPI values are the average of monthly CPIs; for all other countries, annual CPI values are taken directly from the IMF-IFS.  For cases in which data are missing from the IFS database, values are interpolated from surrounding data.  For the projections, U.S. CPI growth rates are projected by ERS.  For all other countries, projected CPI growth rates are averages drawn from IHS Global Insight and Oxford Economic Forecasting.   Regional aggregates are a GDP-weighted average over countries in
the region. 

Real country and regions exchange rates. Real exchange rates are calculated using nominal exchange rates and the U.S. and foreign country CPIs.  See the documentation for the ERS Agricultural Exchange Rate Data Set for a detailed description of this calculation.  All exchange rates are presented in terms of local currency per U.S. dollar.  This means that a rise in a country’s exchange rate implies an appreciation of the dollar or a depreciation of the local currency(ies).  Similarly, a decline implies a depreciation of the dollar or appreciation of the local currency(ies). 

For the 79 countries included in the Agricultural Exchange Rate Data Set, historical nominal and real exchange rates are an average over monthly exchange rates.  For all other countries, real exchange rates are calculated from annual nominal exchange rate and CPI data.  Historical series nominal exchange rates are obtained primarily from the IMF-IFS.  Where available data are considered inaccurate or misleading, various methods are used to obtain a data set representative of global market conditions.  These cases are described in the documentation for the ERS Agricultural Exchange Rate Data Set.  For the projections, projected percent changes in exchange rates are obtained from the IMF-WEO, IHS Global Insight, and Oxford Economic Forecasting.  Regional aggregate exchange rates are a GDP-weighted average of real exchange rates.

Country Coverage

The data described here comprise a complete, annual historical series of GDP, prices, and exchange rates for 189 countries, 37 regional aggregations and 12 income-based aggregations from 1969 to 2013 and forecasts from 2014 to 2030.  Countries are selected based primarily on data availability.  Only countries with data on GDP, prices, and exchange rates sufficient to derive a relatively reliable and complete data series are included.  All countries for which there were sufficient data to fill in missing values are included.  Countries for which one or more of the variables were completely missing from the international data set are omitted.  Thus, North Korea and many island states with very small populations are omitted.  We used back estimation techniques to create data for political units that did not exist during some part of the historical period covered by the data set (1969-2014).

This table  provides a complete list of the countries and regions included in the International Macroeconomic Data Set. 

References

Board of Governors of the Federal Reserve System (2014).  Foreign Exchange Rates.

Callen, T. (2007). “PPP Versus the Market: Which Weight Matters?” Finance and Development, 44/1, March.

IHS Global Insight (2014).  Global Insight’s Comparative World Overview Spreadsheets.

International Monetary Fund (2014). 

National Statistical Bureau, Republic of China, Taiwan (2014). National Accounts and Related Statistics, Directorate of Budget, Accounting and Statistics, Taipei City.

Organisation for Economic Co-operation (OECD) (2006). “A Contribution to the “Purchasing Power Parity vs. Market Exchange Rates for Use in Long-Term Scenarios” Discussion,” OECD Papers, Vol. 6/2.

Oxford Economic Forecasting (2014).  Monthly Model and Forecast Data Set.

Shane, Mathew, Terry Roe, and Agapi Somwaru (2008).  “Exchange Rates, Foreign Income and U.S. Agricultural Exports,” Agricultural and Resource Economics Review 27/2, pp. 160-175, October.

Schuh, G.E. (1974).  “The Exchange Rate and U.S. Agriculture,” American Journal of Agricultural Economics 56/1,
pp. 1-13.

The University of British Columbia (2014).  Pacific Exchange Rate Service.

Westcott, Paul, and Ronald Trostle (2014).  USDA Agricultural Projections to 2023, OCE-141, U.S. Department of Agriculture, February.

World Bank (2014).  World DataBank, World Development Indicators.

 

 

Last updated: Tuesday, February 10, 2015

For more information contact: Kari Heerman