[1]While recent signs indicate that wage growth may be on the rise, many factors can make trends in wage growth difficult to predict, according to a recent report [2] from the Federal Reserve Bank of Cleveland [3].
Cleveland Fed VP and economist Edward S. Knotek II noted in the report the importance of wage growth to the economy, since household income drives consumer spending, which in turn drives the bulk of economic activity. But while the unemployment rate has fallen steadily in the last five years (from 9.8 percent in January 2010 to 5.5. percent in March 2015), wage growth has remained steady during that period, at an annual rate of about 2 percent.
However, the 5.5 percent unemployment rate, which is consistent with many analysts' and policymakers' assessment of relatively normal economic conditions, and there have been greater increases among wage growth in the Bureau of Labor Statistics' Employment Cost Index (2.8 percent year-over-year growth in private industry compensation from January 2010 to March 2015). These factors, combined with the announcement from many businesses, including notable retail chains, that say they intend to raise wages indicate that greater wage growth is on the horizon.
Knotek states in his report that despite recent signs pointing to an imminent increase in worker compensation, factors such as slack in the labor market, bargaining power, worker productivity, inflation, and many other variables and factors, make wage growth impossible to accurately forecast.
Three models were used in the Cleveland Fed report to make the point that wage growth can be impossible to accurately predict for the Employment Cost Index (ECI, which is a measure of wages, salaries, and benefits): first, a Bayesian vector autoregression (BVAR), which includes ECI growth, unemployment rate, productivity, inflation, and other macroeconomic data; second, information from businesses to forecast growth in the ECI; and third, a "random walk" model, which is a simple model that assumes future year-over-year ECI growth will equal its most recently observed value.
Using the BVAR, the possibility exists that a declining unemployment rate could put upward pressure on wage growth, but wage growth could also be influenced by other factors such as productivity and inflation, according to Knotek. Using data through the end of 2014, Knotek's wage growth forecast using the BVAR showed an increase of about 3 percent in ECI by the end of 2017.
In the second model, Knotek used the survey results from the National Federation of Independent Business monthly survey which reports the net percentages of small business responding in the survey who said they plan to increase worker compensation in the next three months or who have increased it over the previous three months. The forecast using this model accurately predicted the annual growth rate of 2.8 percent for Q1 2015, but showed that ECI growth would taper off toward the end of 2017 down to about a 2.5 percent annual growth rate.
Knoteck found the random walk model to be the most reliable of the three models to forecast wage growth.
"By construction, the random walk forecast calls for ECI growth to be steady at a little under 2 and a half percent for the next three years," Knotek said. "Of course, if I were to redo the forecasts using the most recent ECI reading of 2.8 percent, the random walk model would now call for that rate of ECI growth to persist going forward."