State of the State: Ohio
Ohio's unemployment rate rose half a percentage point during a recent period, a larger increase than that of many other states. A Cleveland Fed senior regional officer explains what that may or may not mean.
The Federal Reserve Bank of Cleveland serves the Fourth Federal Reserve District, which comprises Ohio, western Pennsylvania, the northern panhandle of West Virginia, and eastern Kentucky. Like the other Federal Reserve Banks, the Cleveland Fed collects anecdotal reports and analyzes data about the region it serves in order to inform national monetary policy. In the Bank's State of the State series, we share some of what our regional researchers find.
The unemployment rate in Ohio increased half a percentage point in the 6 months ending in April, from 4.7 percent to 5.2 percent, in turn raising the following question: Does an increase of this magnitude signal the start of deterioration in the state's labor market conditions?
While Ohio's unemployment rate remains low—it fell slightly to 5.1 percent in May 2016—its increase during this recent 6-month interval is among the largest for any state. Of interest is that several other Midwestern states saw unemployment rate increases as large as or larger than Ohio's during this time period, including Pennsylvania (up 0.5 percentage points) and Indiana and Illinois (both up 0.7 percentage points). Prior to the recent increase, in the third quarter of 2015 Ohio's unemployment rate had reached its lowest point in nearly 15 years.
Using US unemployment rate data, some analysts have noted that an increase in the unemployment rate above a certain threshold over a predefined period has, in fact, been a good guide to impending economic weakness.
Some analysts have noted that an increase in the unemployment rate above a certain threshold over a predefined period has, in fact, been a good guide to impending economic weakness.
Edward Leamer, professor of economics and statistics at UCLA, reported in a 2008 paper titled “What's a Recession, Anyway?” that an increase in the national unemployment rate that exceeds 0.8 percentage points in a 6-month period has presaged every recession in the post-World War II period, with only 1 false positive. Similarly, analysts at Goldman Sachs have noted that an increase in the 3-month average of the US unemployment rate that exceeds 0.3 percentage points has also been a reliable indication that the US economy is either currently in recession or headed for a recession in the subsequent 6 months. The single exception is in periods immediately following a recession, during which time the unemployment rate may continue to trend up before beginning to decline during an ensuing recovery.
In Ohio, when considering data back to the mid-1970s, a 6-month increase in the unemployment rate of the magnitude we've witnessed recently has indeed tended to occur around US business cycle turning points. The only exception was in the mid-1990s, when Ohio's unemployment rate rose for several months after falling to 4.3 percent, a rate that was more than a percentage point below the US average at the time. In the other cases, by the time the 6-month increase in Ohio's unemployment rate had equaled or exceeded 0.5 percentage points, a national recession had already begun or had recently ended.
Can we conclude, then, that a recession is imminent or perhaps already underway? Not necessarily.
It's important to be cautious in interpreting the recent increase in Ohio's unemployment rate because of the annual revision process that often changes these data in significant ways.
The national unemployment rate continues to decline, and other indicators continue to point to ongoing growth. In addition, it's important to be cautious in interpreting the recent increase in Ohio's unemployment rate because of the annual revision process that often changes these data in significant ways.
A good example of such change is the most recent annual revision to the 2015 data. Initially, the unemployment rate for June was estimated to be 5.2 percent, but it was subsequently revised down to 4.8 percent, a 0.4 percentage point change. Similarly, the initial estimate for October was reported to be 4.4 percent, but was subsequently revised up to 4.7 percent, a 0.3 percentage point change. Clearly, revisions—which can change several years' worth of data—could eliminate the recent increase in the state's unemployment rate.
There is a far more benign interpretation of the recent increase in Ohio's unemployment rate: It could be the result of more people entering the labor force because they believe their likelihood of securing jobs has improved. Indeed, after remaining relatively flat for several years at about 5.7 million workers, Ohio's labor force has increased—and sharply. Over the 6-month periods ending in April and May, the state's labor force grew by 2.3 percent and 2.4 percent, respectively, the largest increases in any 6-month period since the mid-1970s.
The recent increase in Ohio's unemployment rate … could be the result of more people entering the labor force because they believe their likelihood of securing jobs has improved.
However, these recent increases seem surprisingly large, and they therefore may not survive later revisions. If there are downward revisions to the recent labor force estimates without meaningful revisions to the estimates of employment, these numbers could cause a downward revision to the unemployment rate, thereby moderating or erasing recent increases.
While increases in the unemployment rate can be a good predictor of economic conditions to come, it's too early to conclude that a downturn is imminent. Indeed, many other indicators, including anecdotal reports from our District contacts, continue to point to ongoing moderate growth in the economy.
Sum and substance: Ohio's unemployment rate has increased more than that of many other states during a recent 6-month span, something that could signal a turn in the business cycle; however, most indicators continue to point to ongoing growth.
Forefront's Q&A with Charles Manski digs into how data can be revised materially after their initial release and what can be done to best address the resultant uncertainty.