Job Growth Numbers Send Stocks Lower - New York Times

“This is a far weaker report than we expected,” Jared Bernstein, an economist at the Economics Policy Institute, wrote in an e-mail message. “The uptick in the unemployment rate alone, which won’t be revised away, is flashing recession.”

I’ve been talking about the weaker trend in employment growth for some time.  This report didn’t help at all. The year/year gain in employment was 1.0% on the seasonally adjusted basis I find difficult to trust, and was down 0.6% on a non-seasonally adjusted basis.

All the stories about the jobs report being “in line with expectations” never mention the implications of those expectations. If people are expecting a crappy report and are right, it is still a crappy report even though it is in line with expectations.

This morning’s jobs report was one of those.

Sources: Bureau of Labor Statistics data, Stock Market Beat math

The year/year growth in jobs, whether seasonally adjusted or not, is the weakest it has been since June of 2004 (which is when I started tracking it) and the trend is clearly weakening.

According to Reuters, U.S. employers added 110,000 new jobs in September and hiring in the two previous months was revised up strongly, the government said on Friday in a report showing a more resilient labor market than previously thought.

More resilient? That’s not what the year/year growth chart is telling me.

From this perspective, it is looking less and less resilient.

When the Bureau of Labor Statistics (BLS) released their Employment Situation report for July last month, I called it Employment Situation Normal: All Fouled Up, saying:

According to the Bureau of Labor Statistics, nonfarm payroll employment continued to trend up (92,000) in July, and the unemployment rate (4.6 percent) was essentially unchanged. Employment grew in several service-providing industries. Average hourly earnings rose by 6 cents, or 0.3 percent.

Their version of “trending up” clearly means that it continues to be positive. At least, as far as they can tell. The 92,000 increase in July is below the 136,000 average this year, which is below the 186,000 average last year. Some would call that trending down. On the other hand, the number of jobs assumed into existence by the birth/death model was just 26,000 compared to a 2007 average of 110,000 - so based solely on the numbers reported to the BLS this report was actually stronger than others we have seen. Whether the overall report is stronger depends on how accurately the BLS statistical estimates compensates for data they are unable to initially collect.

Given that the BLS calls “trending down” “trending up” their characterization of the current report should come as no surprise.

Nonfarm payroll employment was essentially unchanged (-4,000) in August, and the unemployment rate remained at 4.6 percent, the Bureau of Labor Statistics of the U.S. Department of Labor reported today.

An outright decline now equals “essentially unchanged,” despite expectations prior to the report called for a 100,000 job gain. Oh, and the birth/death model added 120,000 jobs, meaning that the actual data collected by the BLS showed a decline of 124,000 jobs. The only thing “essentially unchanged” about this report is the PR gamesmanship.

The image below shows the year/year change in employment, non-seasonally adjusted but inclusive of the birth/death model.

Stock Market Beat subscribers who want to see the actual spreadsheet behind this can download it here.

...

This post contains additional content for Stock Market Beat premium subscribers. Login to view.

When it comes to reading financial statements and government statistics, it is a good practice to follow Ronald Reagan’s admonition to “trust, but verify” whenever possible. Many have pointed out the high percentage of US job growth generated by the “birth/death model” in recent years, and I have noted that despite headlines to the contrary even the government’s statistics on job growth look sluggish.

So today I am trying to verify. By looking at what companies that are related to the job market are saying, perhaps I can see whether there is more strength or weakness than is otherwise apparent. The companies I chose were Paychex (PAYX) and Automatic Data Processing (ADP), both of which manage payrolls for a large number of businesses, and Hewitt Associates (HEW), which is an outsourcing firm providing benefits management. I reviewed their recent conference calls to collect their thoughts on the job market.

Paychex, for one, is doing its own part to help the employment market.

As of May 31, 2007 our number of employees increased 7% from 10,900 employees one year ago, to 11,700 employees today.

(Excerpt from full PAYX conference call transcript). While that doesn’t say much for the overall economy, it is certainly a start. On the bigger picture, management said:

When we look at the specifics of the things that we have a visibility to and the main number that we look at is the new hire transactions per client and the change in that from a growth year-to-year basis. Night of 2007 ended up at 4.2% for the full year and 5.1% for the fourth quarter. So, those are two very healthy numbers. Last year as an example was 3.1% on the full year.

So, John has often talked about it has been the economy stuck in a good place and the stuck that we, the elements of the equation that we see specifically new hire transactions because we do the compliance reporting.

Bonuses paid is other one that we look at and at the year end the bonus is paid by our clients to their employees were also up. And we look at checks per client, that was also up on a year-to-year basis, although modestly. So, the things that we look at suggest to us that the economy at least the one that we’re addressing, seems to be okay and it doesn’t seem to be moving dramatically one way or the other.

(Excerpt from full PAYX conference call transcript)

ADP does not think things are slowing significantly.

Liz Grausam - Goldman Sachs

Okay. And your pay per control has decelerated a little bit from 3Q into 4Q. Are you seeing anything kind of broad based across your client base one direction or the other in terms of what their hiring trends have been across the year?

Gary Butler

No, we’re not, and you can’t it’s hard to look at it quarter-to-quarter. Now, if you look at the trends for the past year, we were as low as 1.7, and as high as 3%. You have got to really look at it on the full year basis, and we aren’t seeing anything that would lead us just believe that’s decelerating.

(Excerpt from full ADP conference call transcript)

In case things do start to slow, ADP also helps us figure out how to spot it.

Typically in economic slowdowns in the past, what becomes more difficult is selling new business because people stop expending capital or the time to convert the business, et cetera. So, the thing that would probably be the most visible would be a slight slowing of new sales bookings.

Additionally, in severe downturns in the past we’ve seen some clients cancel ancillary reports or supplemental kinds of things that we were doing for them to try to get the bill down a little bit. And certainly if you had a major slowdown you would see some abatement in the pay growth that we’ve enjoyed at the 2% plus level over the last three years.

(Excerpt from full ADP conference call transcript)

Hewitt also appears to see things as solid.

The HR BPO business reported strong top line growth in the third quarter. Overall, after adjusting for the decline in third party revenue and currency, revenues grew 14%, driven primarily by growth of existing clients, including an increase in the project work as well as by contracts that went live in the twelve-month period.

(Excerpt from full HEW conference call transcript)

All in all, the conference calls seem to verify what I had been seeing from the government statistics - a fairly trendless situation.

According to the Bureau of Labor Statistics, nonfarm payroll employment continued to trend up (92,000) in July, and the unemployment rate (4.6 percent) was essentially unchanged. Employment grew in several service-providing industries. Average hourly earnings rose by 6 cents, or 0.3 percent.

Their version of “trending up” clearly means that it continues to be positive. At least, as far as they can tell. The 92,000 increase in July is below the 136,000 average this year, which is below the 186,000 average last year. Some would call that trending down. On the other hand, the number of jobs assumed into existence by the birth/death model was just 26,000 compared to a 2007 average of 110,000 - so based solely on the numbers reported to the BLS this report was actually stronger than others we have seen. Whether the overall report is stronger depends on how accurately the BLS statistical estimates compensates for data they are unable to initially collect.

According to Reuters, June job growth tops forecasts:

Employers added a stronger-than-expected 132,000 new jobs in June and also boosted payrolls more strongly than previously thought in April and May, according to a Labor Department report that underlined a strengthening job market.

Just because economists have low expectations does not mean that higher numbers are a sign of strength. When I look at the year/year change in employment levels, not seasonally adjusted, the current growth rate of about 1.4% looks neither strong nor strengthening.

Classifying the report within my economic taxonomy, however, is tricky. 1.4% doesn’t look good so I am going with bad. But really looking at the last several months the trend is flat, while I offered space for only deteriorating or improving. Since the June growth was (slightly) worse than May’s, I am going with deteriorating. This decision is also due to the fact that more than 100% of the reported change in employment is due to assumptions (the Birth/Death Model) rather than reported data.

EconomicData