Jobs Report Misses Estimates
The headline jobs growth missed expectations, but the report had plenty of positive takeaways. On a broad sense, my analysis might sound like it contradicts itself because sometimes what is good is bad and vice versa. This is because the goal is to have a Goldilocks economy. When wage growth gets too high and job creation is too good, there is usually inflation which causes the Fed to raise rates which ends the cycle. In this case, the strong February report simply led to a weaker headline report for March. Averaging both reports means the labor market is in the same place it has been in for months.
The strong February report was revised to be even stronger as it went from 313,000 to 326,000. The March report showed 103,000 jobs which missed the consensus for 175,000. It was below the lowest estimate which was for 102,000 jobs created. I was expecting a better number, but I also recognized there was some impact from the winter storms. I also expected some revision from February which was strong. It’s also worth noting, this report can easily be revised up to 150,000, making it closer to the recent numbers. As you can see from the red bar below, it sticks out as a very low number. It was the weakest report since March 2017.
The winter weather prevented 159,000 Americans from going to work. As you can see from the chart below, the number of workers affected by bad weather wasn’t much different from previous winter seasons, but obviously if the nor’easters didn’t occur, there would’ve been more job growth. Last month was a big month for full time job creation as 739,000 were added. In March, 311,000 were lost and 310,000 part time jobs were added.
Participation Rate Falls
Private payrolls were 102,000 which is less than half the ADP report. Month to month reports have major vacillations within them, but generally, the ADP and BLS reports have similar trends in the long term. If there was a recession and the ADP report didn’t accurately show the job losses, it would be fixed, but that’s an unlikely scenario. As you can see from the chart below, 63% labor force participation seems to be the ceiling in the past few years. It acted as a ceiling again because the participation rate fell from 63% to 62.9%. That was better than the analyst expectation for 62.8%. The all important employment to population ratio for workers 25-54 years old fell from 79.3% to 79.2%. That’s not surprising because it jumped 0.3% in February which is a large amount for a one month move. This metric often falls in the midst of uptrends. Whether or not you consider this good news depends on your vantage point. On the one hand, it means the labor market likely isn’t near being filled. On the other hand, it means the job market wasn’t amazing in March. This doesn’t mean there’s any shift in the cyclical trend.
Wage Growth Strong
The low number of jobs created gives the noisy signal that the labor market might be full. To be clear, noise is when a data point vacillates sharply without it having much meaning. This could be because of one time effects or problems with how the data is collected. A better way to look at the trend is to measure the year over year growth as you can see in the chart below. What I find interesting is that the decline in growth is always very sudden. The yellow highlight shows mid-cycle weakness and the red arrow points to the pre-recessionary period. Clearly, there’s no evidence of a recession yet, but the rapid decline should come in the next couple years. On the other hand, the participation rate tells us the labor market isn’t full. A sharp decline in the participation rate would mean a recession is coming, but we’re not seeing that. The low level now compared to previous peaks leads me to believe there’s more slack in the labor market.
The much followed average hourly earnings growth was 2.7% which was up one tenth of a percent from last month. It met estimates. The month over month change was 0.3% which met estimates and was up 2 tenths from the previous month. Getting down to the details, the report was 5 basis points below the one from January. The peak of this cycle was the 2.83% growth in September. This report signals the labor market is close to being tight. I only check the hourly earnings report because it is widely followed. The far more important result is weekly earnings because it ignores the effect from the length of the work week. The work week was 34.5 hours which met estimates and was the same as last month. Even though the hourly earnings report wasn’t changed by this metric, past data was impacted by the length so it’s always worth looking at the weekly report.
Weekly earnings growth was very strong. The 3.32% growth was the highest since the 3.33% growth in February 2011. It was the third highest report in this business cycle. The highest was 3.43% in October 2010. The chart below shows another way to look at the data. As you can see, the aggregate weekly payrolls index was up 5.2% from last year which is consistent with nominal GDP growth of 5%.
Conclusion
The headline number was weak and the participation rate fell. The wage growth improved. There’s not much to take away from this information because it mostly goes against what was seen in February. February went against what was seen in January, so the data is vacillating quickly. Looking at the labor market, the economy looks like it is a couple years away from the end of the business cycle. The great job growth in February is completely different from some of the other economic data which suggests a slowing economy. The March report is more consistent with the ‘growth slowing’ narrative.
Recent Comments