Evan Soltas
Mar 19, 2012

Why Real Wages May Rise

Could the median income stagnation be coming to an end?

It's well established, I think, that the last few decades have not been the best for l'homme moyen américain -- that is, the median real hourly wage is virtually unchanged since the 1960s, and mean real compensation has long stagnated, except for a brief window of growth in the late 90s (see graph above). Similarly, the share of labor in firm income has dropped 10% since the 80s.

(Note: Mark Perry has attracted attention for claiming that "consumers have never had it so good," citing the enormous drop in prices in terms of hours-of-labor-equivalent for items in the Sears catalog. The problem with his point is that the catalog captures a narrow range of goods -- consumer durables, electronics, and technology. In all of these categories there has been several decades of deflation driven by falling production costs, which is not representative in the least of the prices in hours-of-labor terms of a broader basket of consumer goods.)

The economic Left likes to blame the decline in private unionization rates, which they suggest undermined the bargaining power of the median worker; most explanations include inequalities of opportunity in education, skill-biased technical change, and the rise of an international market for labor.

I'd like to talk a little more about the last item there. I think, in short, we are going to see the reversal of the trends which have characterized globalization -- most importantly, rising median wages in the United States.

To explain why, we need to look more deeply at the drivers of real median wages in recent years. Improvements in technology and communications made low- and mid-skill labor more closely substitutable across the world, which set off what Arnold Kling has correctly called "The Great Factor-Price Equalization."

To quickly summarize, the idea behind factor-price equalization is that, in the presence of competition, the prices for factors of production (inputs of materials and, relevant to our analysis now, labor) tend to converge across the area of competition.

In the case of wages, my view is that rising productivity in the US more or less canceled out the pressure to reduce real wages. This is effectively saying that American labor became a "better deal" because the productivity has risen much faster than compensation for several decades now.

I also imagine that the traditional considerations in economics about wages -- namely downward nominal wage rigidity -- have also restrained the impact of factor-price equalization in the US.

The main downward pressure on the American compensation-productivity ratio, in my view, has been low international wages. Let's take Chinese wages as a proxy for this broader variable, which I think makes sense given where the particular forms of labor we're consider have been substituted to.

As factor-price equalization would predict, Chinese mean real wages have soared -- rising between 10 and 15 percent year after year since the late 90s. The growth towards the end of this period is increasingly important to the price of labor, since the percent growth occurs upon a larger base.

For much of this period, the increases in mean real wages did not make Chinese labor and goods uncompetitive, as Chinese productivity was also rising fast -- 3.9 percent annually between 1979 and 1994 -- which restrained the growth in their compensation-productivity ratio. The other thing that's changed is that the People's Bank of China is directing a gradual appreciation of the yuan against the dollar.

Now the rising real wages are becoming a competitiveness problem for Chinese production. The Economist, for example, forecasts that by 2015, manufacturing costs will be even between China and the US.

What this suggests is that compensation-productivity ratios are converging between the US and China. As this happens, we should expect increases in productivity in either country to more directly correspond to one-for-one increases in the real wage. This implies that Chinese annual real wage growth should cool to 3 to 4 percent, or nominally 8 to 9 percent, anticipating a 5 percent annual inflation; US real wages should rise 2 to 3 percent annually, or nominally 4 to 5 percent, anticipating 2 percent annual inflation.