The relation between nominal and real wage growth

By: Alex Domash¹ and Lawrence H. Summers²

Over the past year, as employers faced a severe labor shortage and struggled to fill record levels of job openings, workers in the US gained significant leverage and won historically unprecedented pay raises. According to the best available wage data from the Federal Reserve Bank of Atlanta, median year-over-year wage growth in the US reached a series high of 6.5 percent in February 2022.

While it is tempting to think that higher average wages categorically make workers’ richer, we believe there is reason for uncertainty. It is useful to consider the following football analogy made by Arthur Okun to differentiate between individual and aggregate outcomes: When an individual football fan stands up in the crowd, he can see the football game better. But if everyone in the crowd stands up, then nobody sees any better, and everyone is made less comfortable and worse off. Similar to this logic, we believe there is reason to question whether aggregate wage increases are always better for workers.

We first highlight that in general, the correlation between economy-wide wage increases and worker purchasing power is close to zero or negative. Table 1 shows the historical relationship between nominal wage growth and real wage growth going back to 1965. We show the correlation coefficient using alternating wage measures and inflation measures, and include both the contemporaneous correlation and the correlation between lagged nominal wages and real wages. Across all our series, the correlation between nominal wage growth and real wage growth is very low, and usually becomes negative with a lag.

Looking more carefully at the data, one can see that up until a certain threshold, there is a positive association between nominal wage growth and real wage growth, but beyond this point, the relationship turns negative. Figures 1 and 2 show scatterplots of real wage growth vs nominal wage growth, using two different wage measures. Both figures show a clear parabolic relationship between nominal wage growth and real wage growth: as wages rise, workers’ purchasing power increases up until wage growth reaches about 5 percent, and falls thereafter.

While this correlation appears robust across different specifications, we do not believe it is appropriate to interpret it in a causal way. Both nominal wage growth and real wage growth reflect a variety of economic forces. Our suspicion about the best way to understand the documented relationship is that in environments of stable inflation, increases in wages are primarily driven by increases in productivity growth, which justify higher real wages. But past a certain point, it is likely that most increases in wages are driven either by adverse supply shocks or by increases in nominal aggregate demand, both of which are naturally associated with decreases in real wages.

It’s useful to look at nominal wages over the last 25 years to understand the challenge of taming accelerating wage growth. Since the late 1990s, nominal wage growth has reduced rapidly on two occasions — from 2001 to 2002 and 2008 to 2009 — and both times coincided with significant recessions and increases in unemployment (see Figure 3). Between 2001 and 2002, wage growth reduced by 2.4 percentage points, which required unemployment to remain above 5 percent for 4 years and 4 months. Between 2008 and 2009, wage growth reduced by 2.8 percentage points, which required unemployment to remain above 5 percent for 7 years and 10 months.

Historically, wage growth and price inflation have tracked each other very closely, implying that bringing down price inflation will likely require sharp reductions in wage growth, and large increases in economic slack. Table 2 below shows the relation between nominal wage growth (measured using the Employment Cost Index) and price inflation over the last two decades, using both the PCE and CPI. Since 2001, annual nominal wage growth has been 0.7 percentage points higher, on average, then annual price inflation (using the PCE), with a standard deviation of 0.9. Wage growth has never been more than 2.5 percentage points higher than price inflation.

These results imply that it is highly improbable that average price inflation will fall below 3 percent by 2023 — as predicted by the Federal Reserve — without a significant drop in wage growth from its current level of 6.5 percent. We have shown that such declines in wage growth have historically been associated with recessions and large increases in unemployment. Overall, the historical evidence presented in this note thus suggests that we have reason to be wary that higher nominal wages are always good for workers: on an economy wide basis, wage growth that runs too far ahead of productivity can contribute to underlying inflation and reverse the very gains in worker purchasing power that we are trying to achieve.

1. Alex Domash is a Research Fellow in the Mossavar-Rahmani Center for Business & Government at the Harvard Kennedy School. He is a recent graduate of the Masters in Public Administration in International Development (MPA/ID) program at the Harvard Kennedy School. Twitter: @asdomash
2. Lawrence H. Summers is the Charles W. Eliot University Professor and President Emeritus at Harvard University. He served as the 71st Secretary of the Treasury for President Clinton and the Director of the National Economic Council for President Obama. Twitter: @LHSummers