Wages...When? Ever?

The direction of wage growth in the U.S. and across the world remains one of the key questions for markets and policymakers. Conceptually, the notion that demand and supply should drive the price of labor, just as they drive the price for goods, makes sense—if the supply of labor declines, then the price of labor should rise. And yet the paltry rate of wage growth in the U.S. seems significantly out of line with the level of unemployment. What gives?

The answer is complex because there are lots of moving parts. Some argue that we are not appropriately measuring labor supply. Although the U.S. unemployment rate is low, by historical standards, there are other measures of labor supply that are more elevated. Chair Yellen has noted a number of these metrics in recent years—e.g. part-time employment, marginally attached workers, participation rates. While some of these metrics have also shown material improvement, other metrics still have scope to improve further. Labor supply may also be mis-measured to the extent there is a large skills mismatch today. On the one hand, a large skills mismatch might imply that labor supply is even more scarce, and wages should be even higher. However, the consequent negative implications for productivity may fully—and perhaps more than fully—offset the labor scarcity premium. And, in fact, the weakness of productivity growth in this cycle is notable. Unit labor costs, which adjust wages for productivity growth, are actually growing near the historical highs of prior economic peaks.

There are other structural explanations for why wage growth is lagging measures of labor capacity. Technology remains an important factor. The shift towards robotics and the development of applications utilizing artificial intelligence (AI) are disruptive in new ways; although this has not prevented companies from hiring human beings, it may nevertheless have constrained wage growth in the labor cohort that robots/AI are effectively replacing.

The other structural explanation is globalization. This is certainly not a new theme, but the Great Financial Crisis (GFC) may have compounded the effects of globalization. By the time of the crisis, globalization was already an issue for labor supply. Developed markets reacted to the GFC in markedly different ways—the disparity between fiscal and monetary policy responses between the U.S. and the European Union partly explain the dispersion in unemployment rates. The European and U.S. post-GFC recoveries were unsynchronized, with Europe lagging by almost four years. While the Federal Reserve (Fed) and European Central Bank (ECB) both pursued open-ended quantitative easing, the Fed’s dual-mandated focus on the labor market may have broadened the chasm between the two recoveries. By the beginning of 2013, the U.S. unemployment rate had fallen from 10% to 8% even as eurozone unemployment was still climbing its way north of 12%.

Let’s look at some data to understand the broader effects of globalization on the labor market. One of the most significant events of globalization was the return of China to the world economy, notably marked by the entry of China into the World Trade Organization (WTO) in 2000. The introduction of Chinese workers to the universal labor supply was clearly disruptive. One can observe this in the relationship between U.S. wages and the unemployment rate both before and after 2000. From 1985–2000, before China entered the WTO, the correlation between hourly earnings growth and the unemployment rate was 0.82, indicating a strong relationship. From 2000–2016, after China entered the WTO, that same correlation declined to 0.69— a meaningful but weaker relationship.

Interestingly, one can observe that the more closed an economy—where trade is a smaller percentage of GDP—the stronger the relationship between wages and unemployment. Although the relationship between wages and unemployment has weakened for all countries, it has weakened much more for trade-dependent economies. Since 2000, the average correlation between wages and unemployment of more open countries, such as Australia, Canada, South Korea, Poland, and Sweden is 0.32. The same correlation for more closed economies, such as the eurozone, Japan, and the U.S., has averaged 0.73—significantly higher. This data argues to consider not only the domestic unemployment rate in the U.S. when assessing the prospects for U.S. wages, but also a more global unemployment rate.

On this basis, things now look more promising for U.S. wage growth. The Organization for Economic Cooperation and Development (OECD) unemployment rate has fallen to its lowest level since mid-2008. The decline in the unemployment rate in Japan, the U.K., Canada, and even in Europe suggests a global improvement in labor market conditions.

The OECD unemployment rate is a broader measure of unemployment but it still excludes China. However, China’s labor data also looks constructive.

The current cycle is also different because the successive shocks of the GFC, the European sovereign debt crisis, and the commodity bust created very large regional differences in growth. This shows up in the large dispersion of unemployment rates during much of this expansion. Unlike the recovery that unfolded following the 2001 recession, the current recovery has been far less synchronized. In that sense, the decline in the U.S. unemployment rate for much of the past eight years overstated the tightening of labor supply. But here, too, things have improved—the more synchronized global growth impulse over the past year, and the decline in unemployment dispersion suggests we might see more wage growth ahead.

The question on when U.S. wages will accelerate in a meaningful way is a complex question. There are number of drivers of wage growth, some cyclical and some structural. However, one important factor that has changed in recent years is the greater synchronization of the global cycle and the consequent broad-based tightening of global labor markets. The prospects for wage growth in the U.S., and globally, are arguably the strongest in nearly a decade. The Fed and other major central banks will increasingly have to consider monetary policy in light of this shift.