Robert Waldmann responds to Arnold Kling’s response
Arnold Kling was generous enough with his time to respond here to a post of mine
Sorry I was AWOL. Thank you Arnold Kling for responding to me. I’m flattered.
“Not quite. The more heterogeneous labor force does not necessarily raise the natural rate of unemployment. However, it makes it harder to recover from an adverse shock.”
On heterogeneity and the Beveridge curve, I don’t see how heterogeneity can make it harder to recover from a shock or reduce the beneficial impact of a stimulus and increase the costly impact without causing the Beveridge curve to shift out. Shocks occur constantly, gross worker employer separations and hires occur at a huge rate compared to net changes in employment. I can’t think of a model in which the task of matching workers with jobs has become much more difficult, yet the Beveridge curve is about where it was.
I’d say the explanation is that people differ in dimensions other than educational qualifications, so total heterogeneity hasn’t increased that much, and that matching technology has improved (you know the internet and all that).
more on his response after the jump.
Now, in order to get them back to work you need to add demand that specifically uses their skills. This is a much harder trick to pull off. My Hayekian point is that this sounds like a task better suited to the market, not for Washington–hence my remark about “central planning.”
I think that a stimulus that allows the private sector to figure out resource allocation (my preferred approach being a cut in the employer contribution for Social security) is more likely to be effective than a stimulus where the government tries to figure out the resource allocation (health care IT, “green jobs,” etc.).
You seem to be assuming that workers can’t shift from sector to sector if there is surplus demand for labor in one sector (do to the stimulus) and not in another. This is a rather pessimistic view of the magic of the market.
You think that an increase in demand for labor should therefore match the education, geographic, occupation, education, age gender etc of the excess supply (unemployment). So you conclude that the right policy is an equal across the board shift in demand.
It is not enough to believe that the market is better than the government in every way to conclude this. You have to assume that any effort at targeting will be worse than no targeting, so, for example, there is no hint as to relative sudden employment in construction and in agriculture.
Whether there is an accross the board stimulus (cut in payroll taxes) or a sector specific one, the market will figure out how to get people into jobs in roughly the same way. Each intervention implies miss match and need for the market to match people. Which creates a bigger problem is one of “payroll tax cuts vs the current stimulus” not “the market vs the current stimulus.”
I think I can reproduce the logic that leads to the payroll tax proposal
1. Things that happen in the market are optimal, therefore the government should not tilt the playing field.
2. Something very bad has happened in the market as is shown by aggregates cliff diving.
3. Therefore we should assume that that which has clearly gone terribly wrong in the market has gone equally wrong in every sector so the government shouldn’t tilt the playing field.
I see a coherent position that everything is for the best in the best of all possible market economies. I can’t see the logic of a position that the economy is messed up but there is no way for policy makers to estimate which sectors are more messed up than others.