The blame for three decades of stagnant wages in most advanced countries is often laid at the doorstep of globalization, particularly competition from low-wage developing exporters. Globalization is clearly contributing to increased integration of labor markets and closing the wage gap between workers in advanced and developing economies, especially through the spread of technology. It also plays a part in increasing domestic income inequality. But erecting protectionist policies to stanch the forces of globalization is not the best response. Policymakers must instead focus on what can be done to help workers adjust to a changing world.
The Imperfect Integration of Labor Markets
We are very far from a global labor market, as evidenced by a wide disparity in wages. One study finds that the median wage for jobs in advanced countries is two and a half times the wage level for jobs with similar skill levels in the most advanced developing countries, and five times the level in low-income countries. In 2008, a Chinese manufacturing worker earned about one-twentieth the wage level of a U.S. manufacturing worker; a Mexican, one-sixth.
That gap, however, is narrowing in part due to globalization. From 1999 to 2009 (the year of the worst global recession since the 1930s), average real wages rose by about 0.5 percent per year in advanced countries, compared to about 1.5 percent in Africa and Latin America, and almost 8 percent in developing Asia.
Globalization is far from being the whole story behind the narrowing gaps. If wage convergence were principally the result of an integrating global labor market, one would see wages in Africa, the poorest region, rise much faster than the others. But differences in domestic factors, such as the business climate, governance, and education, also play a vital role in determining wage growth.
Global Forces Behind Wage Convergence
Migration, trade, foreign investment, and the spread of technology—all channels of globalization—work to induce wage convergence in interconnected and mutually reinforcing ways.
Increased migration probably plays only a small role in wage convergence. The stock of emigrants from developing countries is just 2 percent of their population, so emigration has little role in raising wages by limiting the growth in labor supply in developing countries. Most studies have found that immigration also has had only modest long-term effects on wages in advanced countries. There are many potential reasons for this: Immigrants typically only account for 10–15 percent of the labor force. Migrants and native workers are imperfect substitutes and may even complement each other, as migrants increase aggregate demand for the services of native workers. And migrants reduce the price of services consumed by native workers.
Trade can promote wage convergence even when workers do not move. Developing countries with abundant labor export goods intensive in labor, so trade induces their wages to rise relative to rich countries, which have less labor and plenty of capital. The more-than-quadrupling of developing countries’ manufactures exports relative to their GDP from 1985 to 2008 almost certainly contributed to wage convergence, especially in middle-income countries that have typically been the most successful exporters. Numerous studies of the direct impact of trade on wages in advanced countries conclude that the depressing effect is small.
Foreign direct investment (FDI) in capital-scarce developing countries can raise the productivity of workers, and thus their wages, by transferring management skills, capital, and technology, and in the process sometimes outsourcing jobs from advanced countries. FDI inflows to developing countries rose from 0.6 percent of their GDP in 1980 to 3.5 percent in 2008. However, over the past decade, total net capital flows (including official and private portfolio flows) equal to 2.6 percent of developing countries’ GDP actually have gone from developing to advanced countries, largely in the form of Treasury bill purchases. This type of investment did not generate jobs directly in advanced countries, but may have reduced the need for domestic borrowing that would have crowded out domestic investors. On balance, it is therefore not clear whether capital flows to and from developing countries have played a large role in promoting wage convergence, though capital flows taking the form of FDI almost certainly have.
Various studies have shown that skill-biased technological change is a major driver in reducing the demand for unskilled workers. At the same time, the transfer of all types of technology through FDI, international trade (imports of machines and learning from competitors and sophisticated customers), and migration (via contacts with diasporas and returning migrants) from advanced to developing countries provides an enormously important opportunity for raising productivity and thus wages in the latter. Moreover, increased competition from low-wage countries tends to spur labor-saving technology in advanced countries.
However, despite the deepening of these channels, remaining barriers to trade and investment, inefficiencies in transport and communications, and structural limitations to the absorption of technology in developing countries, such as insufficient levels of education and business climates that discourage ventures that could adopt technologies, continue to impede the spread of technology. Moreover, in many poor countries, even relatively old technologies are only available in selected locales and to elites, or used only by a few firms. Thus much of the technology adoption that contributes to wage convergence is actually about bringing backward regions, inefficient firms, and disadvantaged groups up to the level of the more advanced in the same country.
Globalization of the Labor Market and Income Distribution
These forces of globalization have been associated with both rising living standards and a deterioration in income distribution in advanced countries: Low-skilled wages have remained flat or even declined, while high-skilled wages have increased sharply. Labor income fell as a share of GDP by 3.5 percentage points from 1993 to 2009. And Gini coefficients, which provide an aggregate measure of income inequality, rose from the mid-1980s to the mid-2000s in all G-7 countries except France.
Inequality also has increased in many developing countries. According to the International Labor Organization, of the 28 developing countries for which data are available, 21 experienced increased income inequality from the early 1990s to the mid-2000s. As in advanced countries, openness to trade and foreign investment have increased the relative return to skilled labor and capital, while reducing the relative return to unskilled labor.
Indeed, some analyses find that trade and financial liberalization episodes, or openness in general, have contributed to worsening income inequality, at least in the middle-income countries. The link between openness and inequality depends in part on the policies adopted, as well as the structure of the economy and the initial income distribution.
How Should Policymakers Respond?
Developing countries’ weak social safety nets and relatively unequal income distributions make them very sensitive to the impact on the losers from globalization. However, their rapid growth and good fiscal positions mean that over time they should be able to build the safety nets available to workers in advanced countries. In designing safety nets, policymakers in developing countries should try to protect workers if unemployed or injured, but would be ill-advised to erect the barriers to firing common in many advanced countries that reduce the overall demand for labor in formal (that is, decent) employment.
Despite their wealth, the labor challenge facing advanced countries, still mired in high unemployment in the wake of the crisis, is more daunting. Identifying the correct policy response is made more arduous by the difficulty of attributing the worsening of income distribution with any precision to trade, technology, demography (increased female participation in the labor force, for example), or other factors such as increasing returns to education.
What is certain is that the answer cannot be to stop the spread of technology or trade. With developing countries likely to be home to the vast majority of the global middle class—people with significant disposable income—in the coming decades, the opportunities available to advanced countries from international trade and technological innovation are likely to greatly increase. At the same time, competition in technology-intensive sectors as developing countries learn will intensify, forcing the pace of innovation to increase even more.
The U.S. tax code could be made more progressive by increasing rates for the highest-income taxpayers, eliminating deductions that favor higher-income groups, reducing social security taxes that weigh most heavily on wage earners, and eliminating unproductive tax loopholes that favor corporations.
But taxes are only part of the problem. Poverty has also been driven by an erosion of the quality of public goods. Most importantly, public education in many advanced countries, beginning with the United States, no longer offers the same opportunities for advancement that it did in the first half of the twentieth century. Increased investment in education and training is all the more important with rapid technological progress and shifts in the demand for workers with different skill levels.
Globalized markets may be changing too quickly to provide the security enjoyed by manufacturing workers in the 1950s. Setting up trade barriers in an attempt to slow the pace of change is not an appropriate response, as they would in effect throw out the baby (efficiency) with the bathwater (inequality). But a much better job can be done to soften the blow for those who are adversely affected and help workers adjust to the demands of a rapidly changing global economy.
Uri Dadush is the director of Carnegie’s International Economics Program. William Shaw is a visiting scholar in Carnegie’s International Economics Program. This note is based on a longer article by the authors “Is the Labor Market Global?” which appeared in Current History in January 2012.