A Backstory on the U.S. Jobs Report

If there was one report to count on to shake up markets, it was the monthly US jobs report.  Part of the issue was that of all the economic data that the US reports, economists had the greatest difficulty in forecasting the monthly change in non-farm payrolls.  It is simply the residual of a great churn—many job gains and losses–and there are not many reliable inputs.  Another source of volatility was what the labor market news meant for policy.

If there was one report to count on to shake up markets, it was the monthly US jobs report.  Part of the issue was that of all the economic data that the US reports, economists had the greatest difficulty in forecasting the monthly change in non-farm payrolls.  It is simply the residual of a great churn—many job gains and losses–and there are not many reliable inputs.  Another source of volatility was what the labor market news meant for policy.

That was then and this is now.  Now the Federal Reserve has distanced its forward guidance from the unemployment rate, preferring a broader range of metrics, some of which are not even contained in the monthly employment report.  The data itself has been amazingly steady, despite the volatility in the underlying economy as measured by GDP. 

Consider the recent averages 224k over the past three months, 235k over six months, and 220k over past 12 months.  Beyond that, over the past two years the average monthly increase has been 210k.  Over the past three years, the average is a little more than 200k. 

The market learns the non-farm payroll growth well before it learns how fast the economy grew during the quarter.  Even though growth is a function of work hours and the productivity, on a quarterly basis, knowing how many net new created jobs does not help one forecast GDP.  In Q3 2103, non-farm payrolls rose 531k.  The economy grew by 4.5%.  In Q2 2014, non-farm payrolls rose by almost 800k and the economy grew by 4.6%.  The US economy added nearly 570k jobs in Q1 14, and the economy contracted by 2.1%.  

Similarly counter-intuitively, there does not seem to be a relationship between wages and unemployment.  The St. Louis Federal Reserve recently re-examined this old chestnut.   It argues that in a workforce characterized by a wide distribution of wages, high wage and low wageworkers experience different labor market conditions.  

Workers with wages at or above the median have a “higher job-finding rate and are more likely to be able to move directly from job to job.”  Therefore, the authors conclude, their bargaining power and wages are less sensitive to changes in the labor market. Lower paid workers are less likely to move directly from job to job and have a lower job-finding rate.  They have less opportunities, and therefore less bargaining power.  

The economists found that across the distribution of wages, the correlation with unemployment was low, and since the Great Financial Crisis has eased a bit more.  They conclude that, “…it seems misguided to look for welfare or signaling movements of mean wages.”

That said, in October, the three-month average at 224k was the lowest since March and is 38k below the three-month average peak in June of almost 267k.  The US economy appears to be tracking a slower rate than Q2 or Q3 2014. 

Barring a significant surprise, we expect the main investment theme, which is one of divergence between the US and the other high-income countries, will remain intact.  The Fed’s leadership continues to suggest that expectations of a rate hike around the middle of next year is remains reasonable.

US Employment Report: Not What it Used to Be is republished with permission from Marc to Market