PWT Snapshots
The long run of every country, from the Penn World Table 11.0
The data cover 185 countries from 1950 to 2023. Hover over a chart to read its values, compare up to five countries, change the benchmark and the years, and open the Python code under any chart to rebuild it in Google Colab.
The page is inspired by Charles I. Jones’s Country Snapshots, the one-page summaries of every country that accompany his textbook Macroeconomics. Panels 1–6 follow his snapshots; panels 7–12 are new.
These snapshots follow the long run: income, growth and its sources. Their companion, the WEO Snapshots, looks at the short run and economic policy of every country (business cycles, inflation, unemployment, government budgets and debt) with the IMF’s World Economic Outlook.
The snapshot
Loading the data…
Enter or a click shows the country; + Compare (or Shift+Enter) adds it to the charts.
Browse all 185 countries
Explore
Each question opens the page on the countries, years and chart it is about.
- Growth miracles. South Korea and Ghana: which country was richer in 1960, and which is richer today? Then look at the investment rates (panel 3) of the two countries. Try it →
- China’s take-off. Start the years in 1978, when China’s reforms began. What is China’s average growth rate since then (panel 2), and how far has it closed the gap with the United States? Try it →
- Two neighbors in Africa. Botswana and Zambia were both poor in 1960. Which one caught up, and when did their paths split? Try it →
- Catching up with the world. With the World average as the benchmark, which of India, China and Brazil has gained the most since 1970? Try it →
- Where does a gap come from? Japan’s GDP per capita is about two-thirds of the US level. Compare its TFP (panel 7), capital-output ratio (panel 8) and hours worked (panel 9) with the United States. Try it →
- Is the labor share constant? The Cobb-Douglas production function says it should be. Compare the United States, Japan and Mexico. Try it →
- The Penn effect. Compare the price level of Switzerland, the United States and India. Why are poorer countries cheaper? Try it →
- High inflation. Argentina’s exchange rate on a ratio scale: in which decades did the peso lose most of its value? Try it →
- A very open economy. How open is Ireland, and does it run a trade surplus or a deficit? Then compare its openness with the United States (choose X+M/GDP above the chart). Try it →
Notes
Reading the charts.
- Each title gives the latest value of each series in the years shown, in parentheses, as in Jones’s snapshots. When several countries are compared, the latest values are in the legend.
- With one country, the first series of a chart is the solid blue line and the second is the dashed line. In panels 5 and 9, the second series has its own axis on the right. When several countries are compared, each chart shows one series (choose it above the chart), with one line per country.
- Benchmark. GDP per capita, GDP per worker and TFP are shown relative to the benchmark in the same year (benchmark = 100). Human capital is shown as a percent of the benchmark’s latest value, so its line still shows the country’s own progress. Jones uses the United States as the benchmark (and its 2019 value for human capital). Choose Levels to see the values themselves. A country that is its own benchmark is shown in levels.
- World average. The world average is population-weighted, over a fixed sample of countries with complete data for 1970–2023, so that the average does not jump when countries enter the data. The samples have 158 (GDP per capita), 113 (GDP per worker), 129 (human capital) and 81 (TFP) countries.
- Growth rate. Panel 2 shows the percent change of GDP per capita from the year before. The average in its title (the dashed line) is the average annual growth rate over the years shown, 100 × ln(ylast/yfirst) / (number of years), as in Jones’s snapshots.
- The ratio (log) scale turns a constant growth rate into a straight line. GDP starts on the ratio scale, as in Jones’s snapshots. Switch it on or off for GDP, the exchange rate and population.
- Missing years are gaps in the lines. Small countries and the years before 1970 have the least data.
The variables. Every number is computed from the Penn World Table 11.0. These are the columns of the data file that the Python code reads (and that Download CSV gives):
| Column | Definition (PWT 11.0 variables) | Unit |
|---|---|---|
pop |
population, pop | millions |
gdp_pc |
GDP per capita, rgdpo / pop | 2021 US dollars (PPP) |
gdp_pw |
GDP per worker, rgdpo / emp | 2021 US dollars (PPP) |
growth |
growth rate of GDP per capita, percent change of gdp_pc from the year before | % per year |
inv |
investment share of GDP, 100 x csh_i | % of GDP (current PPP) |
gov |
government consumption share of GDP, 100 x csh_g | % of GDP (current PPP) |
hc |
human capital index, hc | index |
nx |
net exports including the residual, 100 x (csh_x + csh_m + csh_r) | % of GDP (current PPP) |
open |
openness, exports plus imports, 100 x (csh_x - csh_m) | % of GDP (current PPP) |
xr |
exchange rate, xr | national currency per US dollar |
tfp |
TFP level at current PPPs, ctfp | USA = 1 |
tfp_index |
TFP at constant national prices, rtfpna | 2021 = 1 |
ky |
capital-output ratio, cn / cgdpo | ratio (current PPP) |
emp_pop |
employment-population ratio, 100 x emp / pop | % of population |
hours |
average annual hours worked per worker, avh | hours |
labsh |
labor share, 100 x labsh | % of GDP |
price |
price level of GDP, pl_gdpo | US GDP price level in 2021 = 1 |
Rebuild a chart in Python. Open Show the Python code under any chart. The code reflects what the page shows (the countries, the benchmark and the years), uses only pandas and Plotly Express, and reads the data file from the book’s GitHub repository, so it runs in any empty Colab notebook. Change the choices in Step 2 and run the cell again. The data file is country_snapshots.csv, described in its README.
Differences from Jones’s snapshots. These charts use PWT 11.0 (data through 2023); Jones’s latest version (January 2023) uses PWT 10.01 (data through 2019). PWT revises its past data with each release, so the values for the same year differ slightly between the two.
Sources
- Data: Feenstra, Robert C., Robert Inklaar and Marcel P. Timmer (2015), “The Next Generation of the Penn World Table,” American Economic Review, 105(10), 3150–3182. Penn World Table version 11.0, www.ggdc.net/pwt (DOI 10.34894/FABVLR), licensed CC BY 4.0. The series rgdpo is used for GDP, as in Jones’s snapshots.
- Inspiration: Charles I. Jones, Country Snapshots, Stanford University. The charts, code and text of this page are our own.