Penn World Table

Penn World Table
Name	Penn World Table
Producer	University of Pennsylvania
First	1968
Latest	10.0
Subject	National accounts, purchasing power parity
Languages	English

Penn World Table is a widely used cross-country dataset that provides comparable measures of national income, output, and productivity for many countries and territories. It is used by researchers at institutions such as Harvard University, Massachusetts Institute of Technology, World Bank, International Monetary Fund, and Organisation for Economic Co-operation and Development for empirical work on growth, development, and business cycles. Policymakers at United Nations, European Commission, Asian Development Bank, and central banks consult it alongside datasets from OECD and National Bureau of Economic Research for international comparisons and historical analysis.

Overview

The dataset supplies time-series of real GDP, real GDP per capita, and related aggregates adjusted for purchasing power parity across countries and years, facilitating comparisons among entities such as United States, China, India, United Kingdom, and Brazil. It harmonizes national accounts concepts used by organizations like United Nations Statistical Division and Eurostat and aligns with price-level information from projects such as the International Comparison Program hosted by the World Bank. The dataset is maintained by scholars at centers including University of Groningen, University of California, Berkeley, and the University of Pennsylvania economics department, and it is cited in journals such as American Economic Review, Quarterly Journal of Economics, and Journal of Political Economy.

Methodology

The core methodology builds on the concept of purchasing power parity developed in studies associated with Gustav Cassel and operationalized in projects led by researchers like Irving Kravis and Alan Heston. It combines national accounts data from agencies such as U.S. Bureau of Economic Analysis, Office for National Statistics (UK), and Statistics Canada with price relatives from international price surveys. For aggregation and valuation it uses chain-weighting techniques related to work by Fisher Ideal Index and index number theory advanced by Ewart Kemp, applying methods comparable to those in the International Comparison Program. The methodology incorporates adjustments for capital services inspired by literature from Dale Jorgenson and Zvi Griliches to produce measures of total factor productivity comparable across countries like Japan, Germany, and France.

Data Content and Revisions

The dataset contains variables including real GDP (at constant PPPs), nominal GDP (in national currency), population, capital stock estimates, and measures of hours worked and output per worker for economies ranging from Nigeria to South Korea and Mexico. Major versions have introduced revisions to PPPs, base years, and extrapolation methods; for example, updates have reflected new rounds of the International Comparison Program and revisions from national agencies such as Instituto Nacional de Estadística y Geografía and National Bureau of Statistics of China. Contributors have released major versions numbered sequentially, each addressing methodological critiques and incorporating new data from surveys by organizations like United Nations Conference on Trade and Development and Asia-Pacific Economic Cooperation statistical units.

Uses and Applications

Researchers use the dataset to study long-run growth across regions such as Sub-Saharan Africa, Latin America, East Asia, and Eastern Europe, to analyze convergence debates associated with scholars like Robert Solow and Paul Romer, and to estimate productivity gaps highlighted by Angus Maddison. Applied work in fields connected to institutions like International Monetary Fund and World Bank uses the data for poverty assessments, welfare comparisons, and calibration of macroeconomic models for countries including Argentina, Turkey, and Indonesia. The dataset also supports empirical investigations in labor studies referencing work by James Heckman and capital measurement studies linked to Robert Hall.

Criticisms and Limitations

Critics highlight measurement issues when comparing countries with divergent consumption baskets, citing debates involving researchers from Stanford University, Princeton University, and Yale University. Limitations include sensitivity to base-year choices discussed in papers by Angus Deaton and John Muellbauer, potential bias from extrapolating PPPs between benchmark years noted by scholars at London School of Economics, and challenges measuring informal sector output in economies like Pakistan and Bangladesh. Concerns also arise about the treatment of quality change in manufactured goods and services referenced by researchers at Carnegie Mellon University and Columbia University, and about the interpretation of capital stock and productivity series in policy fora at International Labour Organization and G20 meetings.

History and Development

The project traces intellectual roots to early 20th-century price comparison work by economists such as Gustav Cassel and institutional efforts like the League of Nations comparisons. The modern dataset evolved from initiatives in the 1960s and 1970s involving researchers at University of Pennsylvania, with prominent contributors including Alan Heston, Robert Summers, and colleagues who coordinated with international statistical efforts like the International Comparison Program. Subsequent development involved collaborations among scholars from institutions including University of Groningen, University of California, Los Angeles, and World Bank analysts, responding to methodological advances in index number theory and new international price data from rounds of the International Comparison Program.

Category:Macroeconomic data sets