[UPDATED 11/5/15. See below.]
“The Economy Is Up, but Pay Is Not.” So sayeth Steven Greenhouse in the New York Times this past weekend in a piece titled, “The Mystery of the Vanishing Pay Raise.” Well, consider this mystery solved.
To Greenhouse, and to Jared Bernstein as well, the economy is, as ever, defective. While unemployment has dropped to 5.1% from near 10% in 2009, “wages haven’t accelerated upward, as many had expected.” The Greenhouse piece has a chart showing average hourly wages rising from $24.66 in 2009 to $25.09 today. Bernstein presents his own, even scarier, chart showing a cyclical but steady deterioration in…something (more on that in a second).
But their basic conclusion is just wrong.
Here is my version of Jared’s chart, using data from the Current Employment Survey on the hourly wages of production and nonsupervisory workers (seasonally adjusted):
Look at that decline over 33 years! But Jared’s chart, and mine, shows growth rates for nominal pay. While these rates are declining, the careful observer will note that the trend line is above zero over the whole period, meaning that nominal pay grew every year. What matters, however, is whether nominal pay growth beats inflation growth, which means real pay increases. If there has also been a steady decline in the growth of inflation, then real pay growth has also increased.
And it has. The blue line in the chart below shows how real (inflation-adjusted) average hourly pay has grown over time, using the inflation adjustment that Bernstein tends to prefer and that Greenhouse uses in his chart (the “CPI-U-RS”). First, note that the line shows that average hourly pay is about 11% higher than at the beginning of 1982, and it has been rising—with some pauses—since 1996. Not such a scary chart anymore.
The dark black vertical line indicates the start of 2009, and it is clear that real pay has stagnated between then and now, as Greenhouse suggests. However, note what happened during 2008.
[THE FOLLOWING DISCUSSION HAS BEEN UPDATED BASED ON FEEDBACK FROM THE HERITAGE FOUNDATION'S SALIM FURTH. I THANK HIM FOR THE COMMENTS.]
Average inflation-adjusted pay jumped that year. The explanation, as Salim Furth points out, is that prices dropped significantly during this period, and more so than nominal pay did. With businesses facing lower prices for their products but wage bills that had fallen less, we would have expected to see real wages actually fall after 2008, but they instead flattened out at this elevated level. This pattern reflects the well-known phenomenon of "sticky wages"--it is difficult to reduce existing workers' pay, so employers reduce their costs, in part, by laying off lower-productivity workers. Nominal pay falls, in part because new hires are offered less, but not as much as prices do, since existing workers who keep their jobs don't get a pay cut.
Because of sticky wages, rather than seeing flat or declining real wages in 2008, followed (eventually) by real wage gains as the economy recovered, we instead see most of the real wage increase that would have happened occurring in 2008. If we compare January 2008 to September 2015, we find that, seasonally adjusted, average pay is 5.5% higher. Average pay has not stagnated, it is just that Greenhouse has selected a baseline where wages were, in some sense, artificially higher than they would have been if employers could have reduced nominal wages more in 2008. The increase in real pay is hidden by its shifting from a gradual rise after 2008 to a sharp rise during 2008.
[END OF REVISION]
But the blue line uses an adjustment that overstates the increase in inflation. The orange line uses a better measure, preferred by the Congressional Budget Office and the Federal Reserve Board and consistent with a different measure that the Bureau of Labor Statistics says is best. The orange line indicates average pay has been rising since 1993 and is 23% above its January 1982 level. It has risen a bit even since 2009. The monthly average for 2015 (through September) was 3.0% higher than the 2009 monthly average, and September’s average pay was 3.5% higher than in January 2009 (seasonally adjusted). September’s average was 7.3% higher than in January 2008. That constitutes an annual growth rate of 0.9%, which is only a bit below the annual rate from January 1993 to September 2015 (1.1%) and higher than the annual rate from January 1965 to September 2015 (0.6%).
The charts I have shown are restricted to production and nonsupervisory workers, while Greenhouse’s looks at “all employees” (actually all private employees). Doing so allows me to show trends from before 2006, the first year that averages for all private employees are available. But the trend for all private employees is very similar since 2006. Using the best inflation adjustment available (and adjusting for seasonality), pay was 6.7% higher in September than it had been in January of 2008, while Greenhouse’s chart shows only a 1.7% increase since January of 2009.
Other data yield the same conclusion. The Employment Cost Index indicates that adjusted for inflation using Greenhouse’s approach, average wages and salaries rose 1.6% between the third quarter of 2009 and the third quarter of 2015. But switch to the third quarter of 2008 as the baseline and the increase was 4.3%. Use the better inflation adjustment and that figure rises to 4.6%. Add nonwage benefits in, which cannot be done with Greenhouse’s data source, and average compensation rose by 5.2%. The average compensation for all civilian employees (including government employees) also rose by 5.2% between the third quarter of 2008 and the third quarter of 2015.
Before closing, I want to be clear that these figures are all looking at average pay, not median pay. As inequality has grown, growth in the average has outpaced growth in the median. Median pay has grown in recent decades among women but been stagnant among men. That’s a more complicated story, but the point here is that the mystery identified by Greenhouse and heralded by Bernstein hinges on what has happened to average pay. But there is no “mystery of the vanishing pay raise” in this sense to be explained.
