Workers’ share of national income
All around the world, labour is losing out to capital
The “labour share” of national income has been falling across much of the world since the 1980s (see chart). The Organisation for Economic Co-operation and Development (OECD), a club of mostly rich countries, reckons that labour captured just 62% of all income in the 2000s, down from over 66% in the early 1990s. That sort of decline is not supposed to happen. For decades economists treated the shares of income flowing to labour and capital as fixed (apart from short-run wiggles due to business cycles). When Nicholas Kaldor set out six “stylised facts” about economic growth in 1957, the roughly constant share of income flowing to labour made the list. Many in the profession now wonder whether it still belongs there.
Workers in America tend to blame cheap labour in poorer places for this trend. They are broadly right to do so, according to new research by Michael Elsby of the University of Edinburgh, Bart Hobijn of the Federal Reserve Bank of San Francisco and Aysegul Sahin of the Federal Reserve Bank of New York. They calculated how much different industries in America are exposed to competition from imports, and compared the results with the decline in the labour share in each industry. A greater reliance on imports, they found, is associated with a bigger decline in labour’s take. Of the 3.9 percentage-point fall in the labour share in America over the past 25 years, 3.3 percentage points can be pinned on the likes of Foxconn.
Yet trade cannot account for all labour’s woes in America or elsewhere. Workers in many developing countries, from China to Mexico, have also struggled to seize the benefits of growth over the past two decades. The likeliest culprit is technology, which, the OECD estimates, accounts for roughly 80% of the drop in the labour share among its members. Foxconn, for example, is looking for something different in its new employees: circuitry. The firm says it will add 1m robots to its factories next year.
Cheaper and more powerful equipment, in robotics and computing, has allowed firms to automate an ever larger array of tasks. New research by Loukas Karabarbounis and Brent Neiman of the University of Chicago illustrates the point. They reckon that the cost of investment goods, relative to consumption goods, has dropped 25% over the past 35 years. That made it attractive for firms to swap labour for software whenever possible, which has contributed to a decline in the labour share of five percentage points. In places and industries where the cost of investment goods fell by more, the drop in the labour share was correspondingly larger.
Other work reinforces their conclusion. Despite their emphasis on trade, Messrs Elsby and Hobijn and Ms Sahin note that American labour productivity grew faster than worker compensation in the 1980s and 1990s, before the period of the most rapid growth in imports. Studies looking at the increasing inequality among workers tell a similar story. In recent decades jobs requiring middling skills have declined sharply as a share of total employment, while employment in high- and low-skill occupations has increased. Work by David Autor of MIT, David Dorn of the Centre for Monetary and Financial Studies and Gordon Hanson of the University of California, San Diego, shows that computerisation and automation laid waste mid-level jobs in the 1990s. Trade, by contrast, only became an important cause of the growing disparity in wages in the 2000s.
Trade and technology’s toll on wages has in some cases been abetted by changes in employment laws. In the late 1970s European workers enjoyed high labour shares thanks to stiff labour-market regulation. The labour share topped 75% in Spain and 80% in France. When labour- and product-market liberalisation swept Europe in the early 1980s—motivated in part by stubbornly high unemployment—labour shares tumbled. Privatisation has further weakened labour’s hold.
Such trends may tempt governments to adopt new protections for workers as a means to support the labour share. Yet regulation might instead lead to more unemployment, or to an even faster shift to automation. Trade’s impact could become more benign in future as emerging-market wages rise, but that too could simply hasten automation, as at Foxconn.
Accelerating technological change and rising productivity create the potential for rapid improvements in living standards. Yet if the resulting income gains prove elusive to wage and salary workers, that promise may not be realised.