The Magic Number for Continued US Jobs Growth

The good news, for the moment, is that credit spreads are far below 6.5%.  In July, high-yield bonds averaged 3.9%.  According to our model, a 3.9%-reading, even given the other factors, including the relatively flat US yield curve, is still consistent with a net new jobs creation of 154K for the month of August (+/-122K one standard deviation). That implies about an 89% chance of a positive reading on NFP when the August numbers are released on September 6.

Looking further ahead, the increased volatility in August not only flattened the yield curve further – a concern for jobs growth into 2020– but by Friday August 16, the Bloomberg Barclays US Corporate High Yield Average Option Adjusted Spread widened to around 4.4%.  If it says around this level for the rest of August, this would be consistent with September jobs growth slowing to around 122K (+/-124K one standard deviation).  This estimate could move a bit higher or lower, depending on whether credit spreads narrow or widen, respectively.

One might object to this analysis, pointing out that when oil prices collapsed between late 2014-early 2016, high yield bond spreads widened to as much as 8.37% over US Treasuries – well beyond our current 6.5% threshold for negative net growth in the jobs market.  Yet, there as no downturn in the jobs market in 2016.  All of that is true, but two factors offset the negative impact of credit spreads.  One of those factors we have already mentioned: oil.  While collapsing oil prices were indeed bad for the oil sector, which lost a significant number of jobs in 2015 and 2016, it was akin to a gigantic tax cut for everyone else, boosting consumer spending outside of gas stations. As such, the 2015-16 widening of spreads was mainly limited to the energy sector and did not choke off credit to the rest of the economy.

If credit spreads experience a broader widening later this year or in 2020, job creation may not hold up so well, which brings us to the second reason why the jobs market stayed positive in 2016, despite wider credit spreads. In the years leading up to the collapse in oil prices, the US yield curve had been extraordinarily steep.  In our model we take a one-year moving average of the yield curve and lag it by one year.  In the year to February 2015, the yield curve averaged 350 basis points (bps) of positive slope (Figure 2), given the other factors, that helped to create 110K jobs per month at that time.  In the year to August 2018, by contrast, the yield curve had only 145 bps of slope, which helps to generate about 45K jobs per month, given the other factors (and according to our regression model).

Today’s flat yield curve will likely have little immediate impact on the economy or job creation. That said, job creation is likely to slow further as we go into 2020.  If our model’s beta coefficients are to be believed, the flattening of the yield curve will likely slow the trend rate of job creation by another 25K jobs by this time next year.  Moreover, if the yield curve stays flat, we could see an even bigger drag on job creation in late 2020 and early 2021. The flatter yield curve implies a jobs market that is likely to become less and less resilient in the face of shocks to the credit market. See appendix for details.

Finally, the yield curve itself may be a driver of credit spreads.  Past periods of flat yield curves eventually translated into much wider spreads.  We may already be in the early stages of a credit-spread widening.  So long as the Federal Reserve (Fed) cuts rates and the trade war doesn’t escalate too much,  there is a chance that credit spreads may not widen too much more. That said, the yield curve’s current degree of steepness is constant with levels that triggered much wider spreads at the end of the 1990s (Figure 3) and in 2007 and 2008 (Figure 4).  The Fed’s 2016-18 tightening cycle not only flattened the yield curve, it might have imperiled the credit spread. Over the next several months we’ll see if the Fed can defuse the situation by cutting rates and credit spreads widen dramatically.