Another View of the Data
While I applaud the cautious optimism of Spencer and Tom, I’m more inclined to quote Joseph Brusuelas:
[T]he January payrolls added a dollop of Zen like logic to a recovery that is shaping up like no other. An additional 111,000 workers entered the labor force, yet the unemployment rate fell to 9.7% while private sector employment continued to contract. Hours worked, demand for temporary workers and the hiring in the service sector all improved. However, without the benchmark revisions, the unemployment rate would have increased to 10.6% which better captures the condition of an economy that has seen 8.4 million workers displaced during the recession.
The bump in manufacturing was more than balanced by the drop in the Service Sector, as more and more flower shops cut staff in the face of slack demand and unavailable credit.
If the bank bailout was to bailout the banks—defibrillating them to kick-start the economy’s heart, as it were—then it appears to be time to admit that that program was too small. Or to stop the other programs that are making it more advantageous for banks to hold funds than lend them. Any way you look at it, the optimistic view that declining unemployment has started doesn’t appear to be the way to bet.
I guess I keep seeing a disconnect in terms of bank lending. Banks got into trouble lending money directly or indirectly to people who were not able or willing to pay the money back and were pilloried for doing so. I think it is reasonable for banks to take a closer look at risk in deciding whether to loan available funds. To be sure there are plenty of good risks out there, but the good risks do not see a huge need for credit because they are better off deleveraging in the face of weak demand rather than leveraging their investment. The flower shops do not need to expand their credit lines or employment to meet increased demands and if the flower shop is actually losing money it is not a good risk to lend money.
I am aware that, absent seasonal adjustment, the jobless rate would be at 10.6%. I’m not sure that leaving out benchmark revisions would put the jobless rate at 10.6%, but I am sure that leaving out benchmark revisions is a really bad idea. I don’t know why one would try to argue that the data are somehow more valid before they are corrected for error than after.
It seems to me a matter of story-telling rather than analysis to claim that one particular statistic does a better job of characterizing the economy than do a handfull of others. Any one of us can insist that some data series is the one that tells the “real” story, and we could all choose different ones and all tell different stories. So for completeness and objectivity, I think we need to stick with spencer’s look.
And, by the way, the service sector showed a rise in employment in January, not a drop. The rise in factory employment can be argued to have been more than offset by by declines in construction employment, state and local government, finance, leisure and hospitality jobs – any of those categories, but not the broad service employment category.