The Automatic Data Processing (ADP - Analyst Report) employment survey was much weaker than expected in May. It shows that private sector employment rose by just 38,000, far below consensus expectations for a 170,000 increase.
That marks a massive deceleration from the 177,000 jobs added (as estimated by ADP) in April. Adding to the disappointment was the fact that the April numbers were revised down from a gain of 179,000 jobs.
This is a shockingly weak report, and if confirmed by the BLS on Friday it is something that policy makers (and investors and citizens) have to be seriously concerned about. In April, ADP was much more downbeat than the BLS, which reported a total of 268,000 new private sector jobs. Thus it is not a slam dunk that the Friday report will be disappointing, but that would be the way to bet.
In January it was far too optimistic, showing a gain of 187,000, and then the BLS only reported a gain in private sector jobs of 50,000 (later revised up to 68,000). In February, ADP was almost spot-on with a forecast of 217,000 private sector jobs added and reported a gain of 222,000 private sector jobs. However that was subsequently revised up by the BLS to a gain of 240,000.
In other words, the ADP numbers are generally in the ballpark, but there is still considerable room for the numbers on Friday to be significantly different. As the chart below shows, the ADP numbers track the BLS numbers pretty well over time, but the BLS numbers tend to be somewhat more volatile month to month.
Results by Business Size
Small businesses, defined as those with fewer than 50 employees, rose a total of 27,000 jobs in the month. Medium sized firms, those with between 50 and 499 employees, gained 30,000 jobs while large firms with 500 or more employees dropped 19,000 jobs.
Large businesses are a relatively small share of total employment in the country, accounting for just 17.446 million out of a total of 108.506 million private sector jobs in the country (16.1%). Small business is the largest source of employment at 49.128 (45.3%) million, followed by medium businesses at 41.932 million (38.6%).
The goods producing sector lost a total of 10,000 jobs. Overall, goods producing industries are not that big a source of jobs in this country, just 17.802 million (16.4%) in total. Employment in goods producing industries tends to be more volatile than in the service sector, and thus the goods producing industries have an outsized influence on the overall strength of the job market.
Goods producing jobs, particularly manufacturing jobs, have been in a secular decline, particularly as a share of total employment for more than 30 years now. Relative to the overall increase, it looks like that trend is continuing.
The goods producing sector is made up of Manufacturing, Construction and Mining. While construction jobs did increase during the housing bubble, those jobs were particularly hard hit in the Great Recession.
Construction industry employment was down by 8,000 in May. That eliminated the gains from April, and was only the second gain for the sector since June 2007. Construction jobs peaked well before overall employment in the country, in January 2007. Over that period, employment has shrunk by a total of 2.124 million jobs. That is more than one fourth of the total jobs lost in the entire economy since the recession started.
Historically, construction employment (especially residential construction) is one of the first areas to recover when the economy starts to rebound, but that is not happening this time around. With the extraordinary weakness in new home sales in recent months, there is very little reason to believe in that construction employment is going to pick up anytime soon.
High Vacancy Rates
High vacancy rates in most forms of commercial real estate also means that there is not going to be much of a pick up in commercial construction any time soon. Eventually higher employment is going to lead to higher rates of household formation.
That, combined with population growth, will increase the demand for housing and the massive inventory overhang we have now will be absorbed. That, however, is not a first half of 2011 story, but it could well start to occur late in 2011 or in 2012. The sharp slowdown in employment growth this month is going to push that back further.
Manufacturing had been a bright spot in this recovery, but it faltered in the fall. It had looked like it was getting back on track with a gain of 25,000 jobs in April on top of a gain of 37,000 jobs in March. Now it looks like it is stumbling again, with the loss of 9,000 jobs in May.
There were 11.652 million Manufacturing jobs, or just 10.7% of the overall private sector workforce. Some of the weakness in manufacturing is probably related to supply chain disruptions stemming from the Japanese disaster. That has been a particular problem in the Auto industry, but it has not been alone in feeling those effects. ADP does not break out mining jobs separately, but given the overall rise in goods producing jobs, we can surmise that the number of mining jobs was up 7,000 on the month.
Within the goods producing sector, most of the gains came from the medium-sized firms, which added 7,000 jobs. The small goods producing firms gained 2,000 jobs for the month. Those gains were more than offset by the loss of 19,000 jobs among the large goods producing firms.
The Service sector is far larger, accounting for 90.704 million jobs or 83.6% of the private sector total. It added 48,000, down from adding 138,000 jobs in April. Of the jobs gained in May, 25,000 were added by small service firms, while medium sized firms added 23,000 and large service firms left their payrolls unchanged in May. Far more people are employed by small service firms, (42.495 million) than by either medium sized firms (34.174 million) or by large sized firms (14.035 million).
The ADP report only covers private sector employment, not government jobs at any level. The two series do tend to move in the same direction, and tend to be closer once all of the revisions are in.
Government employment has been falling in recent months, particularly at the state and local level, and that trend is widely expected to continue. Thus if the ADP numbers prove accurate for May, it means that the headline number on Friday is probably going to be below 20,000. That is simply pathetic. It is sure not very inspiring coming out of a deep recession.
That level of job growth will not put much of a dent into the vast army of unemployed and underemployed in this country. We have been getting a lot of weak signals on the economy over the last few weeks.
Wider View: More Bad News
GDP in the first quarter only grew at 1.8%, and the quality of that growth was downgraded in the first revision of the numbers. Initial claims for unemployment insurance have been once again consistently running over the 400,000 level.
The trade deficit numbers were disappointing, as were the data on personal income and spending. Housing is still a disaster zone. It seems pretty clear that the economy is not growing nearly fast enough to put a serious dent in the vast army of the unemployed, and the even larger reserve force of the underemployed.
Downward Revisions Before Friday?
The consensus is looking for a gain of 185,000 jobs on Friday, with more than all of the gains coming from the private sector. The ADP numbers should make a significant change in those expectations. The consensus is looking for a loss of 35,000 government jobs, mostly at the State and Local levels.
The apples-to-apples private sector expectations are for a gain of 220,000 jobs on Friday. State and Local governments have been under severe fiscal strain and are likely to be laying off people. The State and Local layoffs were 22,000 in April. It would not shock me if the State and Local losses are greater than expected again this month, and/or if the April lay-off numbers were revised substantially higher. That means that there is a real possibility (not the most likely case, but possible) that we will see a negative print on total jobs for May.
With a stalemate going on between the GOP house and the Democratic Senate and White House, don’t look for any help to the States from the Federal level. After all, such aid made up about one quarter of the ARRA or Stimulus Plan that the GOP criticized in the election.
Since States are legally not allowed to run operating deficits, they either have to raise taxes or cut spending. Raising taxes is less politically popular right now than cutting spending, and the States continue to cut taxes, particularly on businesses. For the most part, cutting spending at the State and Local level will mean laying people off, or cutting take-home pay of public servants. The State and Local cutbacks are a major source of “de-stimulus that offsets the stimulus from the ARRA on the Federal side.
From the point of view of the overall economy and aggregate demand, it really doesn’t matter if the spending is coming from the Federal Government or the State government. (It does matter on a couple of other levels, but not in terms of total demand in the economy.) Thus, the total amount of stimulus in the economy is much less than is commonly believed.
Even so, there is going to be a lot less of it going forward than we have had over the last two years. Further complicating the picture -- if private sector employment is starting to falter, and this report clearly points in that direction, then overall incomes will stagnate, and those States with income taxes will see revenues weaken again. Assuming people start to spend less when they don’t have jobs, then sales tax revenues will also fall.
The third major source of State and Local revenues, property taxes, are still likely to be strained as housing prices are likely to continue to fall for most of 2011, and that will result in lower assessed values, and hence lower property tax revenues.
This was an extremely discouraging report. It came in far below expectations and points to slower job growth than we saw in either March or April. Those job gains were only slightly above the level needed to keep up with the growth in the labor force to begin with. The numbers implied by this report are well below that level.
The unemployment rate has fallen sharply relative to last fall, but a big part of it (not all, but a big part) was due to a falling civilian participation rate. And if the economy is really starting to turn around, the participation rate is unlikely to continue to fall, and is more likely to rebound. That would put upward pressure on the unemployment rate even as the economy starts to do better in job creation.
With the numbers implied from this report, a continued drop in the participation rate would be the only way to keep the unemployment rate down. That might look OK on paper, but would not reflect any improvement in the real world. The civilian participation rate, and the employment to population ratio (aka the employment rate) are two of the most important numbers to look at when the report comes out on Friday.
Looking Ahead to Friday's Report
When the big report comes out on Friday, another important thing to look at will be the revisions to the March and April numbers. In recent months the revisions have been running large and positive. If that changes this month and we get negative revisions, that will be another sign that something is seriously amiss in this recovery.
Total employment in April was still 6.97 million below the January 2008 peak (and private sector jobs were 6.75 million lower). At a rate of 38,000 new jobs a month, it would take 178 more months from here before we passed the prior private sector employment peak -- in other words, late in 2025. Add in a growing population and workforce, and bringing down unemployment to what we thought of as "normal" before the Great Recession appears to be a glacial process at best.
No, the pace we were on in the previous few months was glacial; this is more like the speed of continental drift. The graph below shows the path of employment, both total and private sector over the last thirty years. Note that relative to the last two downturns, the increase in private sector job growth has been relatively strong, but not as strong as following the 1982-83 recession, but that the decline was also much larger.
If there is a silver lining to this very dark cloud, it is that perhaps some people in Washington will wake up to the problem. The unemployment rate at 9.0% and having been there for such an extended period should be considered a national crisis. It is by far the most pressing economic concern right now.
On the monetary policy front, however, the discussion is all about how to unwind the stimulus from QE2 and of fears of inflation. This, despite inflation -- especially core inflation -- that is running at levels that is far below the average level of the last few decades. People are fretting about the dollar being weak, despite the fact that it is actually very helpful to have a weak currency when you are running massive trade deficits and have very high unemployment.
This economy needs stimulus, not austerity and tight money. Fiscal stimulus would be best, but monetary stimulus in the form of QE3 would also be helpful. Unfortunately, the best case realistic scenario right now is that Washington does not take aggressive action that hurts the economy. The probability of them reversing course and actually doing something constructive is extremely small.