The Case Against “Jobs Day”
November 7, 2014
I find myself increasingly puzzled by the monthly ritual of wall-to-wall reporting on “jobs day.” The first Friday of every month, the Bureau of Labor Statistics releases a report on the employment situation in the country. While I appreciate the importance of the size of the labor market in general – in fact, much of my work on our Taxes and Growth project is about the labor supply – I think monthly updates can be a needless distraction from the big picture. This is especially because the two most commonly reported statistics, the payroll additions and the unemployment rate, are both misleading out of context.
The fact is that the labor market is still deeply depressed, and recovering only very slowly. Too slowly. This remains true regardless of the sort of tiny wiggles we see in month-to-month data. The payroll additions are very small relative to total population, and, of course, they only measure the change in number of jobs, rather than the total quantity of jobs. Therefore, they are a useful statistic mostly over very long time horizons.
How long? Let’s put it this way: after the financial crisis, the employment-population ratio fell from a high of about 63% to an all-time low of 58.2% in November of 2010. Over the last four years, the policy blogosphere has dutifully reported about ”jobs day” every month, as if it were a momentous event that could tell us big things about where the economy was going next.
We have had nearly fifty job days since then. And in all that time, the employment-population ratio has edged back up to – wait for it – 59.2%. That little movement is all we have to show for years’ worth of parsing jobs day.
The other statistic people follow, the unemployment rate, has moved far more substantially, but only because it is becoming increasingly irrelevant and misleading. Ostensible improvements in the unemployment rate can come from the unemployed gaining new jobs, signaling a healthy labor market, or from the unemployed dropping out of the labor force, signaling an unhealthy labor market. By itself, the unemployment rate doesn’t tell us which is which.
In truth, both are part of the story. The economy has improved somewhat for some people, while others still struggle. On net, the situation is not improving nearly fast enough.
Our fellow fiscal policy research organization, the Center on Budget and Policy Priorities, releases a statement on the jobs report every month. For example, this is their statement from last month. This is their statement from today, which has very similar content.
Every month, their content is accurate, and every month, it is roughly the same. This is not a criticism of their analysis. Their statement remains the same because the conditions remain the same. The CBPP remain concerned about the overall labor force participation rate, and they remain concerned about the long-term unemployed.
So do we.
While the CBPP does a good job of putting the numbers in a larger context, not everyone does. We need to think about a bigger picture than what "jobs day" represents. It's almost pathetic to see policy journalists parse the report for whether it shows 150,000 or 200,000 jobs this time. We need something more like ten million.