U.S. job market at a crossroads?

May 5, 2016

A pigeon flies over Times Square in New York City November 15, 2006. Plagued by dirt and noise, a recruitment center shared by several branches of the U.S. armed forces has installed a sound system in New York's neon-bedecked "Crossroads of the World" intended to scare off the winged marauders by playing the sounds of predatory birds.   REUTERS/Lucas Jackson    ( UNITED STATES) - RTR1JDDH

The U.S. job market seems to be at a crossroads: just as wage inflation starts to show flickers of life, hiring growth appears to have lost a bit of momentum.

It is hard to tell, though, whether this is just a normal slowdown late in the business cycle or the earliest signs of a worse downturn – one that could eventually prevent the Federal Reserve from lifting rates much further above zero.

The world’s largest economy is creating a solid number of jobs each month, around 200,000 or so, that will inevitably lead to lower unemployment in the months ahead and, possibly, higher wage inflation. Several economists argue that it would take a halving, or perhaps even more, in the pace of jobs growth to stop the jobless rate from falling and make it start rising again.

Since last October, the gap between the consensus forecast in Reuters Polls for non-farm payrolls and the actual outcome has narrowed considerably. For an indicator that is notorious for coming in wide of the mark – partly because of the wide margin of error for the payroll survey itself – that suggests most are confident of no huge surprises in coming months.

But it’s also clear the pace of hiring is slowing a bit. This is normal so many years into a job market expansion and the trend in first-time claims for unemployment insurance, which has touched a 43-year low on a weekly basis recently, suggests there is no real reason for concern yet.

The problem is, in a normal business cycle the Federal Reserve would be well into a campaign of raising interest long before the jobless rate got this low, in anticipation that higher wage inflation would soon follow.

Jobless claims and funds rate

Given how the unemployment rate has halved in the last six years without any accompanying surge in wage growth, it has become difficult for the Fed and forecasters to judge at what level of employment pay — and with it, inflation — will start to pick up again.

Some simple analysis of Reuters poll data for the U.S. jobs report due Friday offers a few interesting observations on developing trends. As ever, whether or not the median forecast is correct isn’t what’s most useful; what matters more is what the range of opinion is saying, and how close to the edges, or outside the range altogether, the actual data end up.

  • Optimism about the unemployment rate being able to fall much lower and quickly is fading. This is to be expected to some extent as we get closer to what most would accept as full employment. The lowest jobless rate forecast polled so far in regular monthly polls is 4.8 percent, but it has been nearly a year since the actual rate touched the lowest forecast.
  • Opinion is becoming more split from this quarter onward on the trajectory of the job market over the next year. The Reuters consensus for the unemployment rate points to 4.7 percent this time next year compared with 5.0 percent in March. But the spread between the lowest and the highest forecasts begins to widen from 0.5 percentage point in the current quarter (4.6%-5.1%) to 1.6pp by Q2 next year (4.1%-5.7%). The diverging extremes of optimism and belief the job market won’t improve any further, at least on this measure, really start to kick in a bit later this year, although the majority don’t expect the jobless rate to start rising again by then.
  • The April jobless rate is forecast steady at 5.0 percent, marking the first rise in the Reuters Poll median forecast since February 2013, when unemployment was much higher, at 7.7 percent. We know that some of the rise in the unemployment rate last month had to do with evidence of a surge of people coming back to the job market. So this might not mean anything specific, but is worth watching.

US jobless rate

The main thing missing for monetary policy is clear, not just tentative, evidence that wage inflation is picking up. The Federal Reserve took a pass on raising rates in April partly because of this, and partly because official growth data showed an economy barely expanding at all.

With a sluggish world economy as a backdrop, higher wages will be one of the key drivers for U.S. inflation that Fed Chair Janet Yellen needs to see in order to justify more interest rate hikes after a first tentative step up in December from a record low.

On pay growth, recent forecasting data clearly show how entrenched low wages have become in everyone’s minds, but also give a sign that resolve may be breaking.

  • Median forecasts for average earnings growth, always volatile on a monthly basis, have tended to flat-line around 0.2 percent since the financial crisis. That kind of sustained monthly rises is clearly not enough to push annual wage inflation significantly above where it is now, 2.3 percent.
  • Only before September 2008 was there last an extended period of 0.3 percent medians for monthly pay rises. While most economists forecast the volatile monthly figure, the year-on-year figure is more useful to look at for the broader trend, and it has been creeping higher, and even more on an annualized basis.
  • The consensus this month for average earnings is 0.3 percent, with 34 in the sample going for that number, up from 21 last month. Only twice since Lehman Brothers collapsed in September 2008 has the median been 0.3 percent; once in January this year and the other in January 2015. The chart looks like this, and gives perhaps a slight hint that long trend is under threat.

US average earnings history

Also notable is increasing extreme optimism on monthly pay growth. Again, there is usually a technical explanation for out-of-the-ordinary changes, often having to do with pay dates in the calendar month that can cause big declines followed by spikes as in December 2014-January 2015. But we’ve had two forecasts for 0.6 percent monthly pay growth in the last year, despite the fact that has only actually happened twice in the past decade.

— Additional analysis and reporting by Krishna Eluri and Sumanta Dey

No comments so far

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/