State of Employment: Are Fourth District Labor Markets Tight?
The people on both sides of any paycheck have a vested stake in what's changing in the labor market, but for different reasons.
Job seekers' odds of finding employment are better in a tighter labor market, and in such a market employees are better able to bargain for wage increases. But a tightening labor market also means that employers may be challenged to fill vacancies and may see their labor costs rise. As it does for the nation, here in the Fourth District, which comprises Ohio, western Pennsylvania, the northern panhandle of West Virginia, and eastern Kentucky, labor market tightness varies.
Federal Reserve policymakers have their own reason to be interested in labor market tightness.
The Federal Reserve's “dual mandate” is price stability and maximum employment. The Federal Open Market Committee (FOMC), which sets the stance of monetary policy for the United States, recently reaffirmed that in its view price stability is achieved when the longer-run inflation rate is 2 percent.
Maximum employment is a harder concept to nail down. In the language of the FOMC, the maximum employment level is “largely determined by nonmonetary factors” that “may change over time and may not be directly measurable.” It's inappropriate, then, to set a specific unemployment rate as an objective.
But when are labor markets at their maximum sustainable level, or, more colloquially, “tight”?
The unemployment rate is the primary indicator of the tightness of labor markets, and, indeed, the national unemployment rate has dropped considerably, from 10 percent in 2009 to just 5 percent at the end of 2015. So here's the question: What rate of unemployment is the natural outcome of normal movement in and out of jobs, movement that continues even when the economy is operating at its potential?
This “natural rate of unemployment” is a difficult concept to define, and it's even harder to estimate. The first quarter 2016 “Summary of Economic Projections” released by the FOMC shows that a median projection of longer-run normal unemployment would be 4.8 percent in a healthy economy. This number suggests that the national labor market is at least close to the FOMC's current estimate.
However, the decisions households make pose challenges in determining what a normal unemployment rate in a healthy economy would be. For instance, there may be people who are not currently looking for work but who might take jobs as the labor market tightens, thus affecting the rate.
Following the Great Recession, unemployment numbers have risen in part because they include a rise in the number of people who are not looking for or who are otherwise unavailable for work. If growth in the economy can bring those individuals back into the labor market, then there would be less tightness associated with today's unemployment rate (that is, more available workers).
There are many categories of potential workers, and while “potential workers” may sound an amorphous and difficult to calculate assembly, the Bureau of Labor Statistics (BLS) does publish alternative unemployment rates based on including some of the people who fall into this category.
In one case, adding “marginally attached workers”—a group that includes, among others, people who were discouraged in their prospects of finding a job and have thus stopped looking—produces a more inclusive measurement of unemployment. In general, these are people who are most likely to enter the labor force once prospects open up. Taking the calculations a step further, we could include anyone who expressed a desire for full-time work, even if they have not looked for work in the last year or are not currently available to work. People in this group often have significant impediments in their ability to work.
Two conclusions can be reached from accounting for people who are not working but not included in the official unemployment rate.
First, there are potential workers who could re-enter the labor force, and the number of those individuals rose during the Great Recession. Second, the number of these potential workers has declined largely in parallel to the official unemployment rate since early 2010, with the number nearly returning to 2007 levels by the end of 2015.
These conclusions show why we should be cautious about a single statistic such as the unemployment rate, but also why using more inclusive measures of labor market tightness leaves today's market in the rough vicinity of the hard to define concept of maximum employment, or a median value of 4.8 percent.
An alternative approach is to think about the effects of a tight labor market: wage growth.
Nationally, we have seen relatively mild compensation increases. The most encompassing compensation measure, the Employment Cost Index (ECI), shows relatively little compensation growth during the last year—just 2 percent—including all benefits costs such as health insurance and paid time off. In past recoveries, ECI has been considerably higher, suggesting that today's labor market is not as tight as the unemployment rate alone would indicate.
These national challenges are apparent, too, in the Fourth District.
Regions have very distinct workforces and industries, and these distinctions mean that they vary in the level of unemployment rates they can sustain. Rural northwestern Ohio counties often have an unemployment rate under 4 percent, a rate that, at least in part, reflects older workers in more stable employment relationships that come with experience.
On the other side of the spectrum within the District, many eastern Kentucky counties have persistently higher unemployment rates that in part reflect the low education outcomes of residents. Individuals with lower education levels have higher unemployment rates nationally.
And of course, in any county there can arise a significant mismatch between sought-after skills and occupations and the experience and schooling of the workforce. Issues such as these make it very difficult to assess what level of unemployment could be achieved in a region through a stronger national economy.
The District shows promise, though.
One way to see labor market progress within the region is to compare today's unemployment rates to the pre-recession levels of 2006, when the nation's unemployment rate was last near where many economists might define maximum employment. Doing so can give us a broad sense of whether local unemployment rates are getting objectively low.
During this pre-recession period, unemployment was quite low in the District. But by 2010, most counties in the District were at least 4 percentage points over their 2006 levels, with many counties experiencing rates more than 6 percentage points higher. These numbers are largely consistent with the national rise of 5 percentage points in the unemployment rate during the same period, from a 4.6 percent to a 9.6 percent annual average.
The decline in unemployment rates in the District since 2010 has been steep, as well. In 2015, many District counties were between 0 and 2 percentage points lower than the levels they sustained in 2006. And 2 counties, Wyandot County, Ohio, and McCreary County, Kentucky, had unemployment rates more than 2 percentage points lower than their 2006 levels.
It remains impossible to tell how much lower District counties' unemployment rates might fall, but it's clear that many counties are seeing tighter labor market conditions than they've experienced in the last decade.
Reflecting this downward trend, many District contacts have been reporting tighter labor market conditions.
Of course, one would expect those conditions to result in a gradual boost of compensation for affected workers. While local compensation figures comparable to those in the ECI are not available, the Cleveland Fed has asked District contacts about their expected wage costs for new employees. As of last December, only about 18 percent of our contacts who were hiring “were raising wages and/or salaries for most job categories.” Almost 50 percent of our contacts who were hiring reported making adjustments to only selected job categories, while roughly 30 percent reported no change in wages or salaries.
There is certainly evidence of a tighter labor market in the District, but generally District contacts have not witnessed a broad-based increase in wage rates, a situation consistent with the national data on compensation growth.
Sum and substance: National and regional unemployment numbers have fallen markedly since the Great Recession, and while compensation levels have risen in some areas, that rise hasn't been seen across the board.
Conference Board Help Wanted Online (HWOL)
Another way to examine the tightness of labor markets is to compare the number of unemployed individuals to the number of job vacancies. In general, the easier employers find it to fill positions, the greater the number of unemployed.
The Conference Board's HWOL program collects data on online advertised job demand from over 16,000 online sources. Before the recession, major metropolitan areas within the Fourth District had about 2 unemployed individuals per single job opening, a number similar to the national average. Those figures all rose substantially during the Great Recession but are now nearly 1 to 1.
This type of data should be a great way to examine labor market tightness going forward. Over the recent past, though, the nature of job postings has changed dramatically from newspaper advertisements as the primary approach to online advertisements for open positions, skewing the comparative data. This trend tends to increase the number of “counted” job postings, but it doesn't necessarily affect the number of people looking for jobs. This is true even if the overall number of postings, including newspapers and other traditional methods, has not changed.
Pennsylvania's exposure to the oil and gas industry is affecting its jobs numbers. For a look at employment growth in the state, check out “State of the State: Pennsylvania,” the first of a 4-part series on state economies within the Cleveland Fed's region.
Find more on the Conference Board HWOL here.