The big news of the week was always going to be the monthly BLS Employment Surveys. It was destined to move markets one way or the other, and since the +916K number from the Payroll Report (+1,072K if the +156K revision to February is included) significantly bested the 660K-675K consensus that was in the market, the equity markets are likely to move higher in the short-term, as is the longer end of the Treasury yield curve.
Labor Market Data
As expected, there were big numbers in the lower paying services sectors:
- Leisure/Hospitality: +280K
- Education/Health: +101K
- Retail: +23K
- State & Local Gov’t: +129K
And, as expected from the hot ISM and Regional Fed Manufacturing Surveys, factories added +53K (not bad for an economic segment accounting for 11% of GDP). Construction also rose +110K.
The Household Survey (the survey used to calculate the U3 and U6 unemployment rates) corroborated the strong Payroll Survey, showing +609K net new jobs in March. That took the official Unemployment Rate (U3) down to 6.0% in March from 6.2% in February. The U6 rate fell to 10.7% from 11.1%. (Note: U6 is still way too high to make us anxious about a return of “systemic” inflation.) In a turn from recent surveys, full-time jobs in the Household Survey rose +935K while part-time (“for economic reasons,” i.e., not able to find full-time) fell -262K. In other good news, both average hours worked (+1.5% M/M) and the workweek (+0.3 hours) expanded, pushing average weekly earnings up by +0.7%, and almost making up for the -0.9% drop in February’s weekly earnings data.
No report is ever perfect, not even this one. There are big issues to worry about going forward. And the biggest one revolves around “permanently” lost jobs – at least five million of them. The Survey data of the growing length of time to find a job confirm this issue. In the March Surveys, those unemployed for 27 weeks or more rose by +70K to +4.2 million. Given the supposed strength of the recovery, this number should be falling!
The median length of time of those unemployed now stands at 19.7 weeks, a rise of +1.4 weeks from the February survey (18.3 weeks) and +4.4 weeks from January (15.3 weeks). It appears, from this data that very few of those longer-term unemployed are finding jobs. (Is it the overgenerous unemployment benefits?) Many of the people with lengthier unemployment times will find that their old job has simply “disappeared,” as firms have found automated ways to accomplish the needed tasks.
On net, 62% of pandemic job losses have been recouped as of the survey period (week of March 7). The economy is still more than eight million jobs short of its February 2020 pre-pandemic level. As indicated above, we think that at least five million jobs are gone for good. The “new normal” economy is going to require a different set of skills, so the five million overhang will take a long-time to get worked off. With so many unemployed, it is hard to see robust “organic” economic growth. One of the Fed’s basic mandates is “to foster“ full-employment.” Given the above, it is hard to envision a policy tightening anytime on the horizon.
The economy was about 70% reopened at the time of the BLS Employment Surveys (week of March 7) and has reopened further since then (even CA and NY!). Nevertheless, the late March weekly employment data isn’t as robust as the BLS Surveys. State Initial Unemployment Claims (ICs) were 757K the week of March 13, up +25K from the BLS Survey week. State ICs represent claims by employees whose employers contribute to the state systems. Those “new” claims fell -96K the week of March 20, but rose +63K in the latest reported week (March 27). So, on net, despite the “barn burning” data from the BLS Surveys, the state ICs were only marginally lower (-7K) the week of March 27 than they were for the BLS survey week.
The PUA (Pandemic Unemployment Assistance) data is more encouraging. ICs there fell -241K from +479K the week of March 6 to +284K and have continued trending down since, standing at +237K the week of March 27. We expect this data to lead the state data by a week or two, as small business owners reopen and “recall” their employees as business begins to “normalize.” The first chart shows the substantial downward slope in the PUA IC data; the second chart shows the state ICs. Note the shallower downslope on the right-hand side in the second chart due to the delay in recalling employees.
While the ICs are starting to improve, in reality, the biggest issue facing the economy is Continuing Claims (CCs). These are people on the unemployment rolls for more than a week. As indicated above, long-term unemployment looks to be a huge future issue, and the CCs bear this out. The chart shows that they haven’t shown any downslope since mid-December. There was a January blip, but that was a technical timing issue due to the ending of the original PUA program in late December.
Economy and Inflation
No doubt the BLS Employment Survey will reignite the “inflation” narrative on Wall Street. We might yet see even higher equity prices. And the narrative is sure to push up bond yields, especially in the intermediate and long end of the yield curve. But, we see this as temporary (i.e., “transient”) because:
- Long-term permanent unemployment is an impediment to robust organic economic growth.
- The pandemic accelerated the move to automation; as explained in earlier blogs, an inflationary spiral is caused by something putting constant upward pressure on prices. In the 1970s, the cause of inflation was rising wage pressures (exceeding 8% annually). Today, pundits see commodity price spikes as inflationary. These price spikes are being caused by supply chain issues. Those pushing the narrative must implicitly believe that commodity price spikes will continue. However, the economic history of the post-WWII era tells us that these commodity price spikes are temporary (because supplies can rapidly increase). Even in the case of oil (time and capital needed), fracking eventually made the U.S. independent of foreign supplies.
- The move to automation holds down wage growth and increases worker productivity. Over the past year, wage growth has been about 1%; productivity, 4%. It is nearly impossible to have systemic inflation when productivity is growing faster than wages. Prices simply don’t need to rise to protect margins. The large number of unemployed will keep a lid on wages once the overgenerous federal unemployment benefits run out in early September.
- We do expect somewhat robust GDP growth for the quarter just ended (Q1/21). How can that not be? But, those early “knock your socks off” GDP forecasts have been significantly toned down. For example, the Atlanta Fed was at +10.0% Q1 GDP growth in early March, but ended the month with a +4.7% print. And the NY Fed went from +8.7% to +6.1% over the same period. Our view is that we are likely to see good numbers in March and April due to the recent “helicopter” money drop, but we believe that this will be treated like a “bonus” by the recipients, and not as “permanent” income. So, there will be a “blip” in spending, not a permanent rise.
The employment report was better than market expectations; there will likely be market reactions (higher equity prices; higher intermediate and long-term interest rates).
The high level of “permanent” job losses will slow organic GDP growth once the “stimulus” runs its course.
The Fed has a “full employment” mandate. It won’t let rates rise enough to choke off the modest level of organic GDP growth the economy can generate. The narrative of the “return of systemic inflation” is just that – a narrative.
And, we haven’t even talked about the impacts of higher taxes!
(Joshua Barone contributed to this blog)