Tuesday, August 19, 2014

Wage Inflation and Price Inflation

Real wage growth has been disappointingly flat since the Great Recession. Reuters reports that "Most economists do not expect the U.S. central bank to raise benchmark rates until around the middle of next year, given sluggish wage growth." If and when wage growth picks up, I anticipate many debates about the relationship between wage inflation and price inflation by Fed officials and Fed-watchers. The Wall Street Journal blog recently wrote about how Fed-watchers should pay attention to wage growth as an indicator of inflation pressures:
Wage growth is a sign of labor market health but also spills into the Fed’s other mandate: price stability. Labor costs are the driver of costs for most businesses overall. Bigger pay gains push businesses to mark up their own selling prices, leading to higher inflation overall.
But the transmission of wage growth to growth in prices is not straightforward or perfectly understood. It is not safe to assume that firms translate some fixed proportion of labor cost increases into price increases. A 1997 New York Fed study called "Do Rising Labor Costs Trigger Higher Inflation?" found that the answer to its title question is, "It depends." There are several major groups of industries for which labor cost increases and price increases are not directly linked. In industries with high import-penetration ratios, global competition limits firms' abilities to translate higher unit labor costs into higher prices. In some large industries such as utilities, public transportation, and medical care, the government plays a large role in setting prices, so there is not a direct transmission of increased labor costs into increased prices. Housing prices are little affected by rising labor costs because the short-run supply of housing is essentially fixed, so prices depend more on land values and material costs than on labor costs. And finally, some firms are able to respond to labor cost increases by taking steps to increase productivity to maintain profitability without raising prices. The study finds that only in the service sector are cost increases easily passed on to customers.

The relationship between wage inflation and price inflation may be even more tenuous in the context of the post-Great Recession labor market. A new Cleveland Fed study that just came out today, by Edward Knotek II and Saeed Zaman, readdresses the question of whether rising labor costs trigger higher inflation. They look at the cross-correlations of inflation and various wage measures. Their cross-correlation graphs below are meant to show how inflation is correlated with future or past wage growth, as an indication of whether price inflation predicts wage inflation or vice versa. Using data from 1960-present, there is moderate positive correlation at several leads and lags, suggesting that price inflation and wage inflation move together, without one clearly preceding the other. When they use data only from 1984-present, the correlations are still positive, but smaller, indicating a weaker relationship between wage inflation and price inflation in more recent decades.


While Knotek and Zaman only break the time sample into pre- and post-1984, there is reason to believe that the relationship between wage and price inflation may have changed since the Great Recession. For instance, the figure below, from Cleveland Fed researchers, shows that labor income as a share of total income is extremely low by historical standards. This means that firms facing rising labor costs may have a bit more flexibility than usual to absorb the cost increases rather than passing them on to customers.

Several Federal Reserve surveys attempt to gauge firms' perceptions and expectations of their cost increases and price changes. The New York Fed's Empire State Manufacturing Survey is a monthly survey sent to about 200 manufacturing executives in New York State. The executives are asked about current business conditions and their expectations of conditions in six months. Several of the questions ask about prices paid (for inputs) and prices received (for sales).

In the just-released August survey, 30% of firms say they are paying higher prices, but only 16% say they are receiving higher prices. And 47% expect to pay higher prices six months from now, but only 31% expect to receive higher prices. The discrepancy between changes in prices paid and changes in prices received suggests that these firms will not be fully passing on increased input costs to customers. A supplemental section of the Empire State Manufacturing Survey this month asked firms about their response to the Affordable Care Act, and 36% said they will increase or have increased the prices they charge to customers as a result of the Act.

The Atlanta Fed also surveys businesses about cost and price expectations.  This survey asks respondents how various factors will influence the prices they charge over the next 12 months. Just over half of firms expect labor costs to have a "moderate upward influence" on the prices they charge, and another third of firms expect labor costs to have little or no influence on the prices they charge. Meanwhile, 64% of firms expect non-labor costs to have moderate upward influence on the prices they charge and 24% expect non-labor costs to have little or no influence. A special question on the latest round of the Atlanta Fed survey asked about year-ahead compensation expectations. Firms expect an average 2.8% compensation growth (see figure below).

An important takeaway from Knotek and Zaman's study is that "given wages’ limited forecasting power, they are but one piece in a larger puzzle about where the economy and inflation are going."