Weirdly Impressive ADP Report

May ADP report was terrible when using normal calculations, but spectacular compared to expectations. This may have been the biggest beat ever, because other than last month, no estimate would ever be this negative. Specifically, the April reading was revised from 20.236 million job losses to 19.557 million losses. The chart below doesn’t include the revision, but it still gives a great picture of how bad the job losses have been. Before the recession started, employment was about 15 million higher than the prior cycle peak. It has dropped all the way below the previous trough.

(Click on image to enlarge)

In May there were 2.76 million job losses which vastly beat estimates for 8.663 million in job losses. That beat the most optimistic estimate which was for 3 million in losses. This report doesn’t make sense. If it is accurate, the BLS estimates will be completely destroyed. 

The current estimate for Friday’s report is for 6.5 million private-sector job losses and 7.725 million total job losses. That will bring the unemployment rate to 19.8%. Some have been saying that the unemployment rate will get above 20%. It may still despite this ADP report. Initial jobless claims are consistent with worse results than ADP indicates. That being said, because there is a possibility of it being accurate, it’s worth taking a look at the numbers.

There were only 435,000 job losses from private small businesses. That sounds like a mistake. Small businesses have been severely impacted by this recession. Surely, more jobs were lost in May. There were probably some job losses due to the riots, but that’s not included in this sample period. Within small firms, very small firms lost 253,000 jobs and other small firms lost 182,000 jobs. 

Goods-producing small firms lost 91,000 jobs and service providing small firms lost 344,000 jobs. The possibility that the small business data is wrong is supported by the fact that midsized and large firms lost more jobs. You would think small firms would be doing the worst since they have been shut down, don’t have access to capital, and all haven’t been able to get Fed loans.

Midsized firms lost 722,000 jobs and large firms lost 1.604 million jobs. To be clear, the discrepancy between small and large firms doesn’t make up for the entirety of the beat. Even if small firms lost 2 million jobs, the headline number would have beaten estimates. We'll see what the BLS report shows. Within ADP, 794,000 of the job losses were in the goods-producing sector and 1.967 million were in the service-providing sector. Manufacturing lost 719,000.

Trade, transportation, and utilities lost 826,000 jobs which made it the worst industry in May. Professional and business lost 250,000 jobs. Information lost 115,000 jobs. Remember, last month wage growth was strong because there were many more job losses in leisure and hospitality than information (the former pays less than the latter). This time there were only 105,000 losses in leisure and hospitality. 

If this holds true in the BLS report, wage growth will fall sharply. Obviously, we aren’t too concerned with wage growth because so few people have jobs. This is just a tidbit to look for. Leisure and hospitality have the most room to bounce back because restaurants and hotels were closed. There is only one way to go from literally zero sales; that’s up!

Non-Manufacturing ISM PMI

ISM-nonmanufacturing PMI beat estimates just like its counterpart the manufacturing PMI. It rose from 41.8 to 45.4 which was above the consensus of 44. As you can see from the chart below, the 6 month average of yearly growth was -10.5%. Recession average is -12.2%. This report was different from manufacturing because supplier deliveries skyrocketed. The index rose 11.3 points to 78.3 which is terrible news. 

Backed up deliveries doesn’t mean the economy is doing well. It artificially boosts the PMI which is why it’s consistent with such high GDP growth. This report is consistent with -1.1% GDP growth which is slightly worse than the manufacturing PMI, but dramatically higher than what will actually be reported.

(Click on image to enlarge)

Within this report 3 industries saw an increase and 13 saw a decrease. Three with an increase were the following: agriculture, forestry, fishing & hunting; finance & insurance; and retail trade. The good news is the business activity index rose 15 points to 41 and the new orders index rose 9 points to 41.9. 

The bad news is the employment index was only up by 1.8 to 31.8 which implies the labor market contracted swiftly. This is close to the April reading unlike the ADP report which was much better than April. Inventories were up 1.1 to 48 which signals a mild contraction. The prices index was up 0.5 to 55.6 which supports consumer sentiment that inflation is coming.

Within the comments section, COVID-19 was mentioned in 4 out of 10 quotes. Optimism didn’t seem to be as high as it was in manufacturing. That might be because manufacturing is used to wild swings. An accommodation and food service firm stated, “Current operations have been reduced to accommodate a slowed economy, but the business is taking time to review operation and implement more efficient processes. Purchasing has slowed for client-facing goods/services but has increased for internal operating supplies/services.” 

On OpenTable, restaurant bookings growth in America was -83% on June 2nd. In Germany, growth fell from 32% to -39%. Data points in Germany have been all over the map.