Friday, June 4, 2010

1.3 million jobs and counting

Today's headlines are focused on the disappointing number of private sector jobs created (41K vs. expectations of 180K) in May. But that's only one side—the establishment survey side—of the story. The other side is the 178K new private sector jobs that were uncovered by the household survey. Growth in private sector jobs, according to the household survey, has totaled 1.33 million so far this year, but only 495K according to the establishment survey.

As the chart above suggests, the household survey can and does tend to lead the establishment survey, particularly in the first years of a recovery, so it is not unreasonable to conclude from the conflicting data coming from these two surveys that things are looking better than most people who watch the nightly news realize.

The unemployment rate (which is calculated based on the household survey) is still hovering at relatively high levels, however. That's because there have been lots of new entrants to the labor force that are now looking for work. It's going to be awhile—most likely some time after the November elections—before there is any meaningful decline in the unemployment rate.

In other news, average weekly hours rose a bit more than expected, and have now retraced all the decline of last year. At the current rate of improvement, however, we won't see a full recovery in hours worked until the end of this year.

My take on all this is that the recovery is progressing just about as expected. We're not in a full-blown, V-shaped recovery, but we are seeing the ingredients of a moderate recovery, in which the economy grows somewhere around 3-4% per year. We would be in much better shape if it weren't for the massive amount of "stimulus" spending, since that has only sucked over a trillion dollars out of the economy (to finance the increased deficit) that could have been used for more productive purposes.


Unknown said...


Do you have historical data of past recessions supporting your thesis that the household indicator is a leading indicator of the establishment?

1 data point is prone to signifcant sampling error :S

brodero said...

Those in the private sector had a
nice income increase...771.89 from
767.25...on an aggregate basis the
private sector looks like it generated income of 7.7% annualized....

Anonymous said...

Can you overlay a chart of the S&P, to show how the markets are doing relative to employment?

Markets are leading indicators of economic health, while jobs are lagging. Now that its clear the mkt has moved off the lows as has employment, it would be interesting to see how they continued historically.

Thanks for the continued great postings, Scott.

John said...


Looking at the BLS chart in the post it appears the household survey turned up several months before the establishment.

John said...

In 2002, that is.

brodero said...

Help me Scott...

Aggregate Hours index went up
from 98.9 to 99.3 and yet total
private workers increased only from
107.561 to 107.602 million.Are they
just working the current workers a lot more???

John said...


Scott may disagree but that's the way it looks to me too.

A lot of folks are fixating on Hungary this AM, a country that is so small it probably will only affect its small neighbors marginally and their larger ones hardly at all. Why not take a look at Canada? They are our largest trading partner and weigh in at about 10% of our size...not inconsequential. They are growing their economy at twice the rate of the USA, about 6.1%. Growth is so good that the CCB is raising their benchmark short term rate and the Loonie is strong. Unemployment is falling and their banks are strong. Its a good bet SOME of this will impact areas of the USA in a meaningful way... unlike small, remote areas of Europe.

Just a thought when the market is focused on the speck of one disappointing lagging unemployment number.

brodero said...

Hungary's GDP is 1.1% of the Eurozone GDP....

brodero said...

Excellent commentary on a blog
by Charles Lieberman...

"Instead of lipstick, how about some real analysis? As the economy recovers, firms need man-hours. Late in a cycle, after the workweek is back to normal, firms can't get much in the way of additional hours, so they must resort to adding more workers. At this stage of the cycle, they can add either to headcount or to hours or some combination of the two. So, let's look at the combination of the two.

In table B4, the index of man-hours is shown to have increased 0.3% for the month. That's a 3.6% annual rate. If we add a rough estimate for productivity of 2%, (2.5% to 3% might be a better estimate), we would ballpark GDP as rising at a 5.5% rate in real terms. That is hardly shabby.

Another critical element in the report is whether this implies a good gain for household income. This would combine headcount, the workweek and wage rates. That estimate is also provided in table B4 and it shows an increase of 0.7% for the month. If we subtract 0.1% for inflation, we would get an estimate of 0.6% for monthly gain in real income, or 7.2% at an annual rate. This implies a very strong gain in the ability of households to finance spending.

The above suggests that the details of the report are far better than suggested by a superficial reading of the top line number and an analysis based solely on headcount. "

John said...


Well done.

Today's market decline is IMO hot money exiting ahead of the weekend.

This is OT but we've discussed it before. Scott Rothbort has a good piece on High frequency trading at TSCM.

I'm all for economic free speech (naket shorting, etc) but these algorithmic HFCs are no friend of markets. There are limits to free speech (yelling 'fire' in crowded areas, instigating riots, etc.).

I still say make 'em trade the old fashion way. Shut down the machines.

Scott Grannis said...

One reason the household survey tends to lead coming out of recessions (and it does, I've checked but don't have the time to document it) is that it uses a telephone survey, whereas the establishment survey is based on sending out questionnaires to known businesses. Coming out of a recession there are usually a lot of startups and people working on their own that the establishment doesn't find, but that are picked up by a random phone survey.

Scott Grannis said...

Furthermore, even if you doubt that the household survey leads coming out of recessions, there is no reason a priori to disregard its findings. Simply taking the average of the two surveys would tell you that things are picking up, although only moderately.