My bet with Bryan Caplan

Surely the obvious (not necessarily correct) answer is that inflexible labour markets are inflexible on the way down as well as on the way up. That would mean that EU15 unemployment rises slower per loss of GDP in recession, but will decrease slower per gain of GDP in the recovery.

German GDP declined 5% in 2009 – twice as much as the US – but unemployment doesn’t seem to have gone up at all. This is usually explained as a specific result of the Kuzarbeit programme, but a broader inflexibility of the labour market could well play a role too.

JoB, could you please clarify that statement?

I don’t have a full explanation for the apparent failure of economic liberalism in the labor market, but increasingly entrenched inequalities in education (which exist everywhere, but are worse in the US) and the associated lack of social mobility must play a role, I think.

I’m not sure the educational inequalities are playing a major role. Zamfir pointed out the time factor, and I can also tell you there are thousands of lawyers and MBAs who are looking for work right now, too.

Joshua Holmes 01.16.10 at 7:51 pm

” I’m not sure the educational inequalities are playing a major role. Zamfir pointed out the time factor, and I can also tell you there are thousands of lawyers and MBAs who are looking for work right now, too.”

True, but I’m sure that there are millions (10 million?) of fresh high school graduates looking for work.

James Conran @2 -inflexible labour markets- employee protections -are inflexible- prevent kneejerk sackings for the sake of short-term cost-reduction on the way down as well as -on the way up- moaned about a lot by employers’ organisations when the economic climate permits.

(OK, simplistic, perhaps a bit overstated, but still…)

Tim Wilkinson @10: your formulation seems entirely isomorphic with Conran’s @2, for those who don’t import value judgments into their social science. Do you think that you are saying something different?

When I say “isomorphic” I mean that, under either formulation, Prof. Quiggin will lose his bet.

‘your formulation seems entirely isomorphic… for those who don’t import value judgments into their social science…’

Really? For all three of them?

Prof. Quiggin @13: Agreed that, as a theoretical matter, reducing the variance of a measurement does not have a predictable effect on its mean. (No one can dispute the statement at that level of generality.) So it is possible for Conran’s and Wilkinson’s statement to be true, and you still to win, and I was wrong to imply that no further analysis was required. In fact, considerable further empirical analysis would be required. I don’t read the empirical evidence the same way you do, and there are plenty of econ Ph.D.s who also don’t agree with your conclusion (starting with Bryan Caplan, I believe), but time will tell.

What can be predicted with a high degree of certainty is that the outcome of the bet will not have much effect on either Caplan’s or Quiggin’s policy prescriptions. I predict that, if Caplan loses, he will argue that the Obama administration has done such economic damage that the United States has been afflicted with European levels of unemployment, and if Quiggin loses, he will probably revert to the argument that if you included prisoners in the count, he would have won. Because we have no true understanding of macroeconomics, but human nature is relatively clear.

y81 @12: <Tim Wilkinson @10: your formulation seems entirely isomorphic with Conran’s @2

[To repeat a general reminder: this weird and terribly destructive ‘business cycle’ thing is not a force of nature, but a very very nasty and persistent deficiency in ‘free market’ systems – which is not ameliorated by drawing a best fit line through the wildly oscillating curve and calling it ‘equilibrium’. Or what Barry said @15.]

imposed isomorphism

The ‘formulation’ @10 was (I suppose) intended largely, as JQ says, to provide a different description of the issue from the point of view of the ordinary person. Without being too grandiose, I wanted (again: I suppose) to provide a counterweight to the bland phraseology of those who export economics from their value judgements.

But I don’t see that the two versions are isomorphic, except in the sense that the main events they purport to describe are necessarily isomorphic, in virtue of being one and the same set of events.

I mean, of course those who weren’t laid off on the way down don’t need to be re-employed on the way up. (And, yes, those who weren’t employed in the first place because companies are so fretful about the cost of sacking them one day won’t be laid off on the way down.)

But that obvious kind of airy generalisation is one thing, what actually happens another. To attempt to put the point in terms closer to the economystic idiom, for ad hominem purposes:

assymmetric inflexibility

There is a big risk that managers in ‘flexible’ labour markets sack people too easily because next quarter’s profits are the main issue. If they reduce production for a bit due to a short-term cut in income (while they wait for the stampede of expectations to change course and the trend to start moving toward – and past – the notional equilibrium point), they still won’t want to take a hit to that all-important equity premium (still less repudiate bonuses or interrupt the remuneration committees’ circle-jerk by cutting salaries). They can just dump employees costlessly and pick them back up when it suits them.

My formulation was indeed in non-evaluative terms, compatible with the economystic story wrt the way down – employee protections damp the effects of recession on unemployment – and, plausibly, btw, on consumption and on sentiment-at-large – by making the labour market less ‘flexible’ where sacking people is concerned, largely by making sackings directly more expensive (redundancy packages etc).

But wrt the way up, I focussed on employers ‘moaning’ – i.e. making opportunistic and exaggerated complaints – about inflexible markets, rather than actually being terribly constrained by them when it comes to hiring. (Inflexibility is a standard thing to moan about if you are an employer – more power is better, and being able to sack peope for all kinds of non-cost-related reasons is very useful, not east to keep them fearful and compliant.)

The reason my armchair-science persona doesn’t think employers are probably as constrained by ‘inflexibility’ on the way up are two. One is obvious – that ‘inflexibility’ doesn’t impose (and certainly isnlt geared to imposing and need not impose) significant legal penalties or financial burdens (the former largely – and in theory entirely – reducing to the the latter anyway, for corporate persons) directly on hiring.

The second is that the short-termism that leads to sackings – and contributes to the snowballing panic that leads ‘corrections’ in the market to overshoot into recession – may also tend to mean that future costs of the next round of redundancies are discounted. As we know, in the boom times, firms (the management in firms) are as irrationaly exuberant as anyone, so it’s not entirely clear that the fuure costs of redundancy are goingto e factored in.

fucking awful bloody disasters

And this needn’t mean I’m committed to some ‘ratchet’ effect whereby employment always expands in boom times more than it contracts in the ‘downturns’ (as these fucking awful bloody disasters are termed). That, taken as an indefinitely continued trend, would be obviously unsustainable. Why not? First, just because if ‘full’ employment were really reached, a boundary condition would be reached and any such process would stop. But also because of familiar negative feedback mechanisms that mean, e.g. hiring does become more costly as potential employees become more scarce, etc – mechanisms that have nothing to do with the flexibility or otherwise of the markets.

The underlying argument is really about whether ‘inflexibility’ means that the mean level of unemployment is going to be lower (as JQ of course points out). And that depends on all sorts of stuff to do with firms going out of business (mpowell) and how fast the dizzy heights of supply-push can get the economy in good times. Just two things on that before I shut up –

1. regardless of (‘orthogonal to’) that issue, it’s much more important to try and mitigate the horrible effects of recession than to ensure that the production of widgets (still less consumer flibbertigibbets) can expand at breakneck speed at the first sign of a return to normality.

2. Where are all the complaints about a lack of flexibility when it comes to the prospect of temporarily unprofitable (insufficiently profitable) businesses being quickly liquidated (‘sacked’ from the economy)? Three things – in general, nc economics pays insufficent attention to what happens in those great big shaded areas between the actual curve and the nice smooth equilibrium line. There’s something not even that metaphorically fascistic about fixating on the overall performance of GDMP (gross domestic monetised product) – or even unemployment statistics – while largely ignoring the welfare of those buffeted by market forces.

Gross Domestic Monetised Product

Darwin was appalled by the horrible waste and misery involved in natural selection. He also didn’t pretend that the outcome of evolution – increased ‘fitness’ – could be assumed to be an improvement sub spec aet.. Or rather, he appreceiated that talk of ‘improvement’ makes little sense in the absence of a criterion to which it can be anchored.

NC economics is the same, if understood correctly: the invisible hand makes sure that revealed preferences are satisfied as well as they can be under a ‘laissez faire the fait accompli’ system. What isn’t done through trade isn’t ‘revealedly’ preferred, and whatever is done must be however counterintuitive. Darwin got the idea of natural selection by an imperfect analogy with the intentional selection carried out by breeders of animals. And oeconomics – at least in etymological terms – arose from the model of the planned activity of housekeeping. ‘Don’t interfere in the markets because they will self-correct’ is an injunction that needs a lot more justification than I’ve ever heard anyone provide.

[Feel compelled to apologise for length. If only empty Word documents got me tapping as enthusiastically as comment boxes do. Still, noone is forced to read it, and! not only are there no Halasz-style paragraphless slabs – (jch – that is very offputting, to me anyway) – but there are even enticing little headings added.]