Below are a series of pictures depicting the US labor market. We highlight a number of points.
1. The weak payroll numbers in May are significant in that
-
- they were well below even the lower bound estimates,
- prior month numbers were revised down significantly,
- temporary non-farm payrolls were also below estimates,
- there were no special mitigating factors
- Average hourly earnings and quits rates remain in an uptrend indicating a tight labour market.
- Falling participation rates can be explained by factors other than economic growth such as increased school and post graduate enrolment and the better health of new cohorts in the over 55 segment. We do not see falling participation as evidence of a weak economy.
- While the non-manufacturing PMI has weakened recently it remains above 50 (52.9) and the manufacturing PMI has turned over 50 in the last 3 months. Recession risk is low.
- We conclude that the labour market is at an inflexion point and is failing to adjust quickly enough to the evolving economy and that the May number is not a sign of a weak economy.
- The Fed is likely to look beyond the weak May data in their assessment of the economy. We maintain our outlook for a July rate hike. Our initial thesis for not expecting June was based not on the economy but rather the UK EU referendum due Jun 23, just 1 week after the Jun 15 FOMC.
With the exception of the 1990s, whenever labor productivity fell, unemployment fell. This is consistent with a model where labor’s share of output increases to compensate from lower labor productivity when technology could not pick up the slack. The 1990s was the era of the PC and internet which led to higher productivity even as unemployment fell.
The fall in labour participation is not a post 2008 phenomenon, it is a post 2000 phenomenon. The largest falls have been in the 16-19 year segment, presumable due to higher school enrolment. Post 2008, we have also seen declines in the 20-24 year segment with smaller declines in the 25-54 year segment. The falloff in the 20-24 year segment could be due to increased enrolment in post graduate education which was particularly popular in the post dotcom bust years. The 55+ segment has seen participation increase, presumably due to healthier populations. The trend of falling participation rates can therefore be explained mainly by demographic and non-economic factors ruling out the hypothesis of a weak economy.
The latest non-farm payroll numbers were quite poor, at 38K they fell below the lowest professional estimate of 90K and far below the average 160K. Taken as a percentage of the total labor force the number does not look better.
Two areas of strength, albeit not too much of it. One is the quits rate which is steadily climbing, although it has yet to reclaim the levels pre 2008. Quits rates are consistent with a tight labour market. Second is average hourly earnings which continues to recover. It also has yet to reclaim pre 2008.
Charts data source: Bloomberg and BLS