Wednesday, July 14, 2010

Job seeker to job opening ratio over time

I've mentioned a few time that one of my interests is gross labor market flows - separations, hiring, job creation, job destruction - which I think are underemphasized relative to numbers like unemployment and employment.

Yglesias has a good post up on the ratio of unemployed to job openings - a good measure of how tight the labor market is, or in this case whether it's even functioning (a higher ratio means that more job seekers are chasing fewer jobs). He grabs a chart from Scott Winship that illustrates the point brilliantly, tracking the ratio back fifty years:


It's fundamentally the same ratio. The green line uses the Conference Board's help-wanted ad index (an old measure of job openings), the red line uses the Job Opening and Labor Turnover Survey from the BLS (I've used this data for a paper I wrote up and presented at the Southern Economic Association, but never tried to publish), and the blue line uses a newer version of the same Conference Board measure.

I'll just put this bluntly: this is not a normal business cycle - the labor market is fundamentally broken.

That spike is a labor market surplus. Can wage reductions clear the market or not? Econ 101 says it should, but if Econ 101 described the world under all circumstances there would be no reason for Econ 102 - and since we know there is an Econ 102, our suspicion should be "maybe not - maybe something somewhat different is going on this time around".

Maybe this isn't a temporary disequilibrium that will adjust itself - maybe there is an inherent suboptimality affecting the market this time.

21 comments:

  1. It is taking a while for the poison to get out of the system.

    ReplyDelete
  2. A related issue: http://bit.ly/9pvuIr

    ReplyDelete
  3. On the link -
    Veronique de Rugy has already demonstrated that she has very little credibility on stimulus questions. It's not just the issue that she doubts fiscal stimulus. In the recent past she's embarassed the Mercatus Center by bungling stimulus disbursement data that an undergrad should have had no problem working with. This was picked up by all the major blogs.

    But her thoughts on stimulus oughta be addressed too.

    First, she says Romer thinks the multiplier is 2.5. That's absurd. The Romer-prefered mutliplier is 1.5. I'm guessing de Rugy read the 1.5 multiplier, thought it meant 1.5 additional dollars of stimulus for every dollar of government spending, and called that a multiplier of 2.5. Whatever the confusion, that's absolutely not the multiplier that Romer has been using - and given de Rugy's well document confusion over the data in the past, something like this seems plausible.

    I've noted several times on here why Barro's estimates are legitimate, but should not be applied "out of sample" as stasticians say. de Rugy covers none of this. She doesn't even attempt to explain why Barro got his result and Romer got hers. Instead, Barro is reasonable and Romer is "fantasy" with exactly zero explanation for the assessment.

    I, on the other hand, can provide a very good reason why Barro and Romer are both right and why neither result is surprising, and why both have a best-that-we-can-hope-for estimate of the multiplier.

    But feel free to trust de Rugy's assertions that are oddly self-serving.

    ReplyDelete
  4. regarding this: Recovery.gov, which actually counts the number of jobs created (if in a very favorable light), only displays roughly 680,000 jobs, not 3 million. Why would the White House not up that number if in fact 3 million jobs had been created? Because they don’t have names and addresses to back up their gargantuan projections.

    Again, de Rugy demonstrates she simply doesn't understand the issue at hand. The 680,000 jobs from recovery.gov, are (1.) not updated, as I understand it - which is bad enough, but (2.) jobs directly created by stimulus disbursements. The 3 million job figure is from direct AND indirect jobs created. It's a counter-factual, in other words, which isn't even measuring the same thing that recover.gov is measuring. The two figures shouldn't even be compared to each other. The fact that she even tries to compare them shows how little she understands what's going on.

    In the next section, where she discusses investment, she again doesn't seem to understand the difference between a counterfactual - which they Fed can't report with survey data - and an actual investment figure, which they can.

    No person that bungles the very concept of counterfactual has any business testifying before Congress on fiscal stimulus. It's disheartening to see that de Rugy is already 45 minutes into her testimony.

    It's not just that she disagrees on the multiplier - de Rugy repeatedly demonstrates that she doesn't even know what she's talking about.

    If you want to furnish me with a counter-argument, give me Scott Sumner or even perhaps Robert Barro. de Rugy is with Wayne Anderson and Don Boudreaux on that list of people that for the most part aren't even worth reading.

    ReplyDelete
  5. "In the recent past she's embarassed the Mercatus Center by bungling stimulus disbursement data that an undergrad should have had no problem working with."

    Didn't she win that lunch, actually? Oh, yeah she did.

    This was reported at Marginal Revolution here: http://www.marginalrevolution.com/marginalrevolution/2010/04/nate-silver-wins-a-lunch-date-with-vero-de-rugy.html

    ReplyDelete
  6. 1. Her explanation sounds odd for several reasons - I'm trying to check out the revised report to actually look at the regressions, but it's slow loading for some reason.

    2. I'm not saying she was wrong to say that party matters. I imagine it does. It would be absurd to think that Democrats don't benefit from having control of Congress when a major spending bill passed. I never critiqued her on those grounds. I'm saying she clearly didn't understand the data itself and the way ARRA worked - and that still holds true regardless of whatever Nate Silver was wrong about.

    3. She still shows an obvious lack of understanding on other stimulus issues that have nothing to do with stimulus that I listed.

    She's really not worth your time, Xenophon. There are enough good libertarians/conservatives out there. You shouldn't waste your time with her.

    I'll report back if I end up finding anything notable in the new regressions.

    ReplyDelete
  7. No, this is what Romer stated about the multiplier: "We confess to considerable uncertainty about our choice of multipliers for this element of the package." pg. 12 Add to this the problem that Romer was basically basing her multiplier on past government spending in situations not remotely analogous to the current problems and you see why the multiplier used is sort of pointless.

    The 2.5 is likely a typo.

    "...are (1.) not updated, as I understand it..."

    Well, that is no mark in the favor of the government, is it? Anyway, it is not just that it isn't updated, private audits have shown it to be woefully inaccurate.

    "It's a counter-factual, in other words..."

    No it is a number propagated as a means to spin political B.S.

    I really don't care what your opinion of de Rugy is.

    ReplyDelete
  8. Right - as I've said countless times on here before mutliplier estimation is very uncertain.

    What are you thinking Romer looked at vs. what Barro looked at? Most people that compare them say that Barro's comparison looks less like what's happening today. Why do you argue the opposite? What are your reasons?

    No it is a number propagated as a means to spin political B.S.

    You really need to understand that this is not a counter-argument to anything that pisses you off, Xenophon. People genuinely disagree with you on these points. You have to grow up and deal with that fact. I don't buy what a lot of people say about the stimulus, but I don't dismiss it as "political B.S.".


    I can't find these revised regressions anywhere - they're not in the July paper. Do you know where they are?

    ReplyDelete
  9. Daniel,

    There are lots of people who are not really worth my time, but I still read them. Really, unless you've advocated some things I consider beyond the pale - you know, preventative war, torture, expanding the range of eminent domain, the unitary executive theory, etc. - I'll still read you.

    ReplyDelete
  10. Daniel,

    "I don't buy what a lot of people say about the stimulus, but I don't dismiss it as "political B.S."."

    If it quacks like political B.S., it probably is political B.S. It is a rule of thumb that has worked very well for me in the past.

    Ahh, aren't they at the Mercatus Center website? That's what the original article that Cowen linked to states as I recall.

    ReplyDelete
  11. Also, I am not making an argument in favor of Barro.

    ReplyDelete
  12. The original article is there and a four page article recently from de Rugy is there, but I can't find the new regressions.

    Maybe they're updated in the original article? I'll look again.

    ReplyDelete
  13. I am still of the opinion that we have a money shortage. I believe I understand the theoretical case for a fiscal stimulus, but it seems to me the least desirable and least effective means to address the problem. Since I do not believe the "zero nominal bound" is a serious impediment to expansionary policy, I believe the buck stops with the Federal Reserve.

    All that said, the Obama administration has been quite proactive about changing the rules of the game. Surely this can't be helping the labour market.

    ReplyDelete
  14. Lee Kelly,

    Remember, Obama is all about "bold, persistent experimentation."

    ReplyDelete
  15. Lee Kelly -
    I circle around that view - I'm certainly sympathetic to it. And I usually don't make a big deal of the zero lower bound on here. But I do think liquidity preference is an important burden on prosperity right now, and even if the zero lower bound isn't the end of the discussion, you still have a lot of thirst for liquidity that needs to be slaked before you're going to get upward pressure on the price level, which is presumably what's going to make the zero lower bound an irrelevant constraint.

    Presumably, the demand for liquidity is a psychosis of sorts. It's a result of uncertainty about demand, rather than careful calculation. It seems to me that the most direct route for addressing that is to augment demand rather than money supply.

    And while I don't think the zero lower bound has to be an impediment to monetary policy, it doesn't mean it won't weaken the general economy even further. If we are suffering from liquidity preference that is far too strong, driving the real interest rate below zero will, at best, keep pace with the value of cash which won't solve our liquidity preference problem and, at worst because of fear of debt monetization, drive the real interest rate on government bonds even lower than the real return on cash. That would make liquidity preference increase, not decrease - which would hurt our output problem.

    That's just off the top of my head - I may have to think through some of that again.

    I think augmenting demand seems like the safer route, if you think of this not just as a money demand problem, but also a liquidity preference problem.

    I'm not opposed to helicopter drops. At this point I'll take about anything relative to what we're doing. But it doesn't seem to me to be the smartest way of doing it.

    ReplyDelete
  16. For some reason, I first read that as "bold, persistent perspiration." Kinda makes sense, too.

    ReplyDelete
  17. Daniel,

    When you write "demand for liquidity," what exactly do you mean?

    ReplyDelete
  18. Demand for cash or highly liquid assets. The desire to hold money without putting it at interest.

    ReplyDelete
  19. For a moment I thought you might have meant something different. Thanks for the clarification.

    I find it strange that you consider augmenting demand to be the safer route, because it seems to me that changing supply is both safer and easier. Both of us can't be right.

    In any case, I tend to worry about ways in which our current monetary institutions may exacerbate fluxuations in money demand.

    ReplyDelete
  20. I do think under most conditions adjusting money supply is the safer route. I think when depressed aggregate demand and liquidity preference is the primary problem - and especially when we're in or near a liquidity trap - it becomes much less safe.

    Which isn't to say I think monetary policy is useless in a liquidity trap - I just think it's weaker and may even risk exascerbating an important part of the problem while it's providing a solution.

    A big part of my view on this is completely separate from macroeconomics as well. I simply think there are a lot of important public investments that we could make. If those investments are sitting their waiting, why not augment demand by pursuing them?

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.