Regional Labor Market Recessions and Recoveries
All the recent talk of “green shoots” has led to speculation about what a recovery of the U.S. labor market will look like. Will employment begin to bounce back in a V-shaped recovery curve, or will a U-shaped or even an L-shaped curve ensue? While this alphabet soup of recovery patterns (there’s a W, as well) describe previous national labor market recoveries, the patterns also emerge at the regional level.
At the national level, labor market recoveries after the four most recent U.S. recessions (1981, 1990, 2001, and the current one) were not all alike. Once the trough in nonfarm payroll employment had been reached in the 1981 recession, employment rebounded sharply, and unemployment fell markedly. The next two recessions, on the other hand, experienced relatively “jobless” recoveries, with payroll employment growth picking up slowly (if at all) after the trough was reached and unemployment rising for several quarters. Seventeen months into the current recession, nonfarm payroll employment has declined 4.0 percent, and the unemployment rate has risen 4.0 percentage points. The trough in this recession will be both deeper and occur later than in the previous three. The current unemployment rate of 8.9 percent is still below what it was in the 1981 recession, but most analysts expect that it will continue to rise over the next several quarters.
Labor market recovery patterns have varied across the states of the Fourth District in these recessions as well. Past business cycles indicate that Ohio typically has been relatively slow to recover lost employment. This was true in the recovery cycle after the 1981 recession, when Ohio’s recovery lagged the nation, as well as after the 2001 recession, when Ohio experienced a truly “jobless” recovery, as nonfarm payroll employment remained essentially flat from 2002 through 2007. Since the start of the current recession, the percentage decline in Ohio’s payroll employment has exceeded the nation’s decline by 1.3 percentage points, but the decline is still less than in the 1981 recession. Ohio’s unemployment rate rose to 10.2 percent in May, above the national unemployment rate but well below the peak unemployment rate of 13.9 percent, seen in the 1981 recession.
Pennsylvania’s payroll employment, in contrast, initially held up relatively well in the current recession, although it started to deteriorate in 2009. Still, the state’s labor market performance is better than the nation’s as a whole, with Pennsylvania’s unemployment rate a full percentage point below the national rate and its nonfarm payroll employment declining by 1.2 percentage points less than the nation’s drop.
Kentucky’s employment losses in the current cycle are somewhat greater than the national decline. Kentucky has lost 4.5 percent of its nonfarm payroll employment, and the state’s unemployment rate has risen to 9.8 percent. In terms of nonfarm payroll employment losses, this is shaping up to be Kentucky’s worst downturn of the past four cycles. Kentucky’s unemployment rate is up 4.3 percent, with 3.4 percentage points coming in the past five months.
According to the current data, West Virginia’s downturn occurred somewhat later and has been milder than Ohio’s, Kentucky’s, and Pennsylvania’s. However, since the beginning of 2009, West Virginia has been playing catch-up, as its payroll employment declined sharply and its unemployment rate rose. Indeed, West Virginia’s unemployment rate is up 3.2 percentage points in this cycle, with almost the entire increase taking place since December 2008 (3.0 percentage points). Still, this is markedly different from the 1981 recession, when West Virginia experienced the highest unemployment rate of any state (18.2 percent), and nonfarm payroll employment remained depressed for years after the recession.
Hidden in these state patterns is the possibility that different industries in Fourth District states may be experiencing different patterns of job losses across the recessions. We illustrate this for Ohio and Pennsylvania by comparing the employment growth rates of various industries in these states to national growth rates in the same industries. We do this comparison for both the 1981 recession and the current recession. In the charts below, the solid red line represents what would be equal growth rates at the state and national level. Thus, for data points below the line, the growth rate of the industry in that state is below the national level, and vice versa for points above the line.
During the 1981 recession, almost all of Ohio’s and Pennsylvania’s industries grew at rates that were below their corresponding national rates. National growth rates in the construction and manufacturing sectors were quite low, but in Ohio and Pennsylvania they were even lower. Employment growth in services remained positive, as well as in mining. Still, Ohio and Pennsylvania experienced below-national growth across almost all sectors.
In the current recession, growth rates in most Ohio and Pennsylvania industries have remained relatively close to the nation. Again, construction and manufacturing have experienced the sharpest national declines in employment. Pennsylvania’s construction industry has outperformed the nation, while Ohio’s manufacturing industry has underperformed the nation. This underperformance of manufacturing in Ohio reflects, in part, the fact that Ohio has an above-average concentration of automotive industries, and these industries have recently experienced sharp declines in employment. Sectors that have expanded employment nationally include education and health services and government. These sectors grew at similar rates in Ohio and Pennsylvania, as well.
Overall, states with counties in the Fourth District have experienced somewhat different labor market cycles in the current recession. Ohio has had the weakest labor market, while Pennsylvania and West Virginia have had relatively strong labor markets. Employment recoveries from recessions have also differed markedly. In recent recessions, job losses have been relatively minor but the accompanying recovery was also anemic. A current fear is that while we are experiencing a sharp labor market contraction similar in magnitude to the 1981 recession, we will have a labor market recovery similar to those which occurred after the 1990 or 2001 recessions—the L-shaped scenario.