The Calculus of Labor Imbalances

By Rainer StrackJens BaierMatthew Marchingo, and Shailesh Sharda

To identify global labor gaps and surpluses, BCG compared the supply and demand of labor in 25 developing and emerging economies through the year 2030. In our analysis, we looked at four key components and took a quantitative view of workforce shortages and surpluses.

Labor Supply. The labor supply includes the economically active population, or all people who are providing labor to produce goods and services during a given time period. We were interested in how many people each country will have in its labor force, taking into account birth rates, immigration trends, and labor force participation for specific demographic groups.

To calculate this number, we relied on the forecasts of population growth and labor force participation from the UN and one of its agencies, the International Labor Organization (ILO). 1 1 The latest available labor-participation-rate forecasts from the ILO run only through 2020. We kept this number constant through 2030. Notes: 1 The latest available labor-participation-rate forecasts from the ILO run only through 2020. We kept this number constant through 2030. We sought a source that was not only highly credible but also sufficiently global in order to obtain the utmost consistency and comparability across the selected countries. Every country we analyzed is a member of the UN, and all UN data agencies, including the ILO, coordinate directly with each country’s economic agency to obtain data and projections. Since people born in 2014 will enter the labor market in 2029 at the earliest, at age 15, the predictions are fairly reliable—barring, of course, changes in immigration or workforce participation rates.

Labor Demand. We used a simple identity to calculate labor demand. (See the exhibit below.) To know what labor demand will be in a given year, we need to know what GDP and labor productivity will be in that year. Of course, nobody knows these numbers, and forecasts are highly speculative. However, we can calculate two scenarios. If a country aims for the same GDP and labor-productivity growth over the next 10 or 20 years as it has had in the past 10 or 20 years, we can directly calculate the number of people it will need to generate this growth.

To illustrate, let’s take the example of the U.S. Over the past 20 years, average annual real GDP growth (CAGR) was 2.6 percent, and annual labor productivity grew on average by 1.6 percent. 2 2 In this report, all GDP figures (including growth rates) are real (inflation-corrected) GDP, not nominal. Notes: 2 In this report, all GDP figures (including growth rates) are real (inflation-corrected) GDP, not nominal. Suppose the U.S. aspires to have the same growth rate in the future as it has had in the past. We would divide GDP by labor productivity to calculate how many people it will need in its labor force.

Using historical GDP and labor productivity growth rates from the Economist Intelligence Unit, we calculated annual GDP and labor productivity growth rates over the past 10- and 20-year periods (2003 through 2012 and 1993 through 2012) for every country we studied. Using these historical growth rates, we extrapolated the labor force each country would need in order to keep GDP and productivity growth for 2020 and 2030 at their historical levels. We did not use forecast numbers because forecasts take various factors into account, including the forecast labor supply. Because we explicitly looked at the labor supply, we needed GDP figures that were not yet constrained by any changes in labor supply numbers. Using historic averages allowed us to treat both variables independently.

Surpluses and Gaps. By subtracting demand from supply, we calculated labor force surpluses and shortfalls by year (2020 and 2030) and by country, for each of the two GDP- and productivity-growth scenarios.

Measures and Interventions. We analyzed the impact of changing certain key variables that directly influence the labor supply, such as immigration rates and workforce participation rates for specific demographic groups. This might include delaying the retirement age or increasing the share of women in the labor force. We also analyzed the effects of changing variables affecting labor demand. What would happen, for instance, if productivity were raised as a result of educational initiatives? Then we extrapolated the effects on the supply-and-demand curves.

A Quantitative View. Our analysis takes a purely quantitative perspective of aggregate workforces. It does not consider employability and job qualifications. When supply-and-demand numbers are broken down by education levels and job families, as we will do in subsequent reports, balances will likely change—revealing more significant shortages and surpluses than the averages here suggest.