How healthy is the U.S. labor market? Economists can’t seem to agree. Some argue it has regained much of its pre-crisis strength, citing the significant drop in the unemployment rate, the relatively low number of “short-term” unemployed and the recent acceleration in wage growth. Others say it remains in the doldrums, suggesting that the unemployment rate has fallen mostly because workers have left the labor force, and that the large number of “discouraged,” underemployed and “long-term” unemployed workers – as well as the still-anemic pace of wage growth – indicates substantial labor market slack.
While the unemployment rate has historically been one of the Federal Reserve’s key measures of spare capacity, and thus inflation risk, those eagerly awaiting each month’s employment report for signals on the Fed’s likely response may be barking up the wrong tree. Why? The answer lies in the debate itself: Precisely because there are so many conflicting signals coming out of the labor market, the Fed’s reaction function may have changed. Though it will continue to monitor a combination of employment and price signals – Chairman Yellen’s employment “dashboard” – the Fed seems to have shifted its focus to inflation. In other words, the Federal Open Market Committee (FOMC) may have decided to follow the advice given to militia men at the Battle of Bunker Hill: “Don’t fire until you see the whites of their eyes.” Rather than hike rates preemptively based on forward-looking indicators of inflation, the Fed plans to hold onto its policy bullets until it sees wage growth and inflation back at more “normal” levels.
The participation rate conundrum
The participation rate measures the percent of the population that is in the labor force – that is, people who either have a job or are unemployed but actively looking for a job. Historically, the unemployment rate and the participation rate have had an inverse relationship; as one goes up, the other tends to go down. The intuition is that a tight labor market both lowers the unemployment rate (as more of the unemployed find jobs) and entices otherwise discouraged people – and perhaps a stay-at-home parent or two – to look for work. The conundrum is that the participation rate, which dropped quite sharply during the financial crisis, has continued to fall in subsequent years, even as the unemployment rate has steadily declined (Figure 1). If the falling unemployment rate is a sign of a strong labor market, then why is the participation rate not rising?
There are essentially two competing responses. Employment “bears” argue that the unemployment rate is falling in part because workers are becoming discouraged by the bleak job outlook. When a worker stops looking for a job he is no longer counted as unemployed because he is deemed to have fallen out of the labor force, resulting in a drop in both the unemployment rate and the participation rate. Hence, the bears believe the declining unemployment rate is really a sign that the labor market is so bad that workers are simply losing hope. Bulls, on the other hand, argue that lower participation is structural and related to the aging of the population. Since participation rates tend to be lower among those over 65, as this group grows as a percentage of the population the participation rate declines naturally. Indeed, the data show a substantial increase in the percent of the population aged 65+ in recent years, from 15.5% in 2005 to nearly 18% today (Figure 2).
So who's right? The truth lies somewhere in the middle. Changing demographics have certainly added a structural element to the decline in the participation rate over the past decade, but they don’t explain the whole story. Since 1997, demographic shifts have been responsible for nearly two percentage points of the decline in the participation rate (Figure 3). Interestingly, however, approximately half of this impact had already occurred by 2008. The upshot is that of the three-percentage-point decline in the participation rate since 2008, only one-third can be explained by demographics. The remainder, or two percentage points, reflects declining participation that is due to other factors. The bulls would be quick to point out that increasing numbers of workers have been filing for disability in recent years, and this may indeed be part of the explanation. Bears, by contrast, would highlight the still-elevated numbers of Americans who say they are working part-time for economic reasons or that they would like a job even though they are not looking for one (and are thus not counted among the unemployed). The bottom line: It seems clear that the participation rate has been and will continue to be affected by demographics, but there appear to be additional, more cyclical, factors contributing to its decline in recent years.
The debate matters, of course, because of the participation rate’s impact on wages. If the participation rate is low for structural reasons, then there may not be lots of “excess” workers waiting until the labor market strengthens to return to the labor force; any increase in demand for labor could thus result in higher wages. If it's low simply because the job market is poor (i.e., for more cyclical reasons), then it follows that an improving job market will bring some of these previously discouraged workers off the sidelines to compete for jobs; in this case wages might remain stable despite increasing demand for workers.
Short- versus long-term unemployment
When it comes to wage pressures, there is a similar disagreement between those who argue that short-term unemployment matters more than long-term and those who believe the distinction doesn’t matter. The debate has arisen partly because the rates of short- and long-term unemployment, which have historically moved in lock step, have diverged in recent years: The long-term rate has remained stubbornly high, while the short-term rate has returned to pre-crisis levels (Figure 4). Some argue that the divergence is structural and reflects the fact that the long-term unemployed – who have been out of a job for more than six months – are less likely to have the skills required to compete for any given job that becomes available. Fed Chair Janet Yellen, on the other hand, has said that she believes the unusually high level of long-term unemployment is essentially cyclical in nature, and a recent Fed paper argued that there is no appreciable difference between the effects of long- and short-term unemployment on inflation. Why does it matter? If short-term unemployment is what drives wages, then we can expect wage growth to return to pre-crisis levels in short order. If long-term unemployment matters just as much, then wages may not see much upward pressure just yet.
The real question, however, is whether any of these debates about the internal dynamics of the labor market really matter anymore when it comes to predicting Fed policy. While the central bank still attempts to estimate the natural rate of unemployment (also called the NAIRU, or the level of unemployment at which inflation may begin to accelerate), the conflicting signals from the labor market have clearly made the Fed less willing to trust its models. The result: Though it may still monitor a combination of employment and inflation signals, the Fed has decided that inflation is what really matters.
A shift in the reaction function
In Fed Chair Yellen’s February testimony before Congress, she suggested that the decline in the unemployment rate did not tell the whole story and thus called into question the value of the Fed’s 6.5% unemployment rate “threshold” as a policy indicator. At the March Fed meeting, the unemployment threshold was ditched completely. Most recently, the Fed has even suggested that it may pay just as much attention to actual inflation as it does expected inflation.
The Fed’s changing reaction function has two main implications for investors. The first is that rates are likely to stay on hold until the personal consumption expenditures (PCE) deflator (currently just over 1%) moves decisively toward the Fed’s 2% target. In a developed economy like the U.S., price inflation usually comes with wage inflation, so as the minutes of the March FOMC meeting suggested, the committee is likely to pay particularly close attention to wages. Investors should do the same.
The second is that the Fed is less likely to be proactive. This is partly because it has less faith in its ability to forecast inflation, but – somewhat ironically – also because it has gained a great deal of credibility by winning the war on inflation. Not only can it wait for clear signs of inflation before hiking rates, but its destination – the long-term equilibrium fed funds rate – may be closer to 2% than 4%. So the Fed will wait to see the whites of inflation’s eyes before shooting, but the market’s faith in it means that even when it does start shooting, it may need to let very few bullets fly. FOMC members may not have reached the vaunted status of Bunker Hill’s sure-shot militiamen, but being veterans of a successful multi-decade war on inflation expectations has its privileges.