(probably the last post I’ll be doing on these issues for a while)
Tyler Cowen has a further response. His argument – I think I am presenting it correctly – is that regulation may not improve the lot of workers facing specific depredations, because firms will find other ways to screw them, leading to “indeterminate” outcomes, which might or might not be to workers’ long term benefit. His preferred alternative is an improved welfare state, which will enhance workers’ bargaining position vis-a-vis firms.
I think it’s right that regulations aimed at specific abuses will often have somewhat indeterminate consequences. It may also be empirically the case that for many abuses, the best we can do is to improve exit options for workers. But I also think that this is a decidedly poor basis for arguing that we should never regulate workplace conditions. There are regulations of workplace conditions (forbidding racial discrimination; forbidding sexual harassment; requiring basic workplace safety) that seem, as best as I can tell, to have worked out pretty well, to the extent that they have properly been implemented. Partial indeterminacy doesn’t say that we get rid of regulation altogether; it suggests that (a) one wants to think carefully about the tradeoffs and possible abuses, as best as one can given limited information, before introducing regulations, and (b) to be very willing to incorporate feedback ex post and to alter or abandon regulations that aren’t working.
I worry that I am beginning to sound like a cracked record, but to repeat my point again, very simple economic models don’t tell us much about regulating for worker protection, because they have nothing useful to say about asymmetric bargaining situations with radically incomplete contracts. There are more complicated and interesting economic models which might tell us more – but they are better as ways of reducing the possibility space into something intellectually tractable than at providing particular guidance about what to do across a variety of subtly different forms of power relationship. This is not, contra Tyler’s claim, ” cit[ing] a lack of perfect competition and then assum[ing] the proverbial pony.” It’s a simpler criticism – that those who start from the position of perfect competition and similar are assuming the pony, because they are assuming away problems of power asymmetry, incomplete contracts etc. There are a variety of reasonable and potentially useful ways of thinking about power relations in the workplace – but to be reasonable, they have to confront the actual problem rather than assuming that it doesn’t exist.
And there are quite a number of people who write in ways that assume it away. One striking example is Tyler’s co-blogger, Alex Tabarrok’s post of a couple of days ago, which claims that leftists like me are blinded by our assumptions, because we ignore how workers at risk of sexual harassment will seek and get higher wages in compensation for the risk.
Perfect competition and … well not just any old pony, but a fiery-eyed Death Dealing Destrier-Pony of Chaos and Destruction. To be specific, an article in the American Economics Review purporting to show that people working in industries where there was a high risk of sexual harassment demanded, and received, a wage premium in return for accepting this risk. Market competition working, just as market competition ought to work, allowing people to bargain and get what they want.
The only problem with all of this (and Alex’s post surely deserves to be read; he harrumphs and snorts very nearly as magnificently as his Death-Pony), is that the article looks to have been made from a common by-product of both ponies and horses. Again I’m repeating myself (I made these points in an update to the original post); its findings are based on the assumption that official complaints about sexual harassment are an unproblematic proxy for actual incidence of sexual harassment. This is at best a highly dubious assumption – the obvious counter-argument is that people’s willingness to report sexual harassment is likely to vary with their income level, independent of its actual incidence. Or to put it more plainly, women in industries with a lot of exploitation are more likely to worry about losing their jobs for complaining about sexual harassment and hence are less likely to complain when it happens.
One would think that this rather wonderful finding
The log wage difference between a job with zero sexual harassment risk and a job with the mean sexual harassment risk is 0.0155, or about 25 cents per hour for women, and 0.0252, or about 50 cents per hour for men. The large compensation for sexual harassment risk for men is surprising. One possible explanation is that since men infrequently file sexual harassment claims, those claims that are filed are particularly egregious, and exposure to such risk warrants a larger compensating differential than received by women.
would have caused the author, and indeed anyone even faintly connected to reality who read the article, to think again about whether the statistical findings could possibly be what they claimed to be. I can’t for the life of me think of a convincing micro-level story about how men systematically demand higher compensation for jobs that might have increased incidence of sexual harassment. Indeed, I don’t think that there is one. There is a possible explanation based on differences in willingness to report, but given the other flaws in the article (unless I am misreading it, the level of aggregation at which ‘industry’ is measured is so high as to be effectively meaningless for this kind of inquiry – ‘agriculture’ v. ‘information’ v. ‘Professional and business services’ etc) I’m now inclined to think that the finding is entirely spurious.
The point is this. There is a set of intellectual blinders (commonly although by no meansuniversally associated with both economics and libertarianism, and frequently most problematic among those who are both economists and libertarians) which precludes by fiat the possibility that regulation can do any good. These blinders are the result of a specific understanding of the implications of economic theory, which heavily emphasizes the explanatory power of free markets and contract, and correspondingly de-emphasizes power relations. This isn’t the only possible way to understand economic theory (one could – and people have done this – use the tools e.g. of game theory to analyze power relations). Still, the blinders are very common in US and UK policy discussions, and Alex’s post provides an outstanding example of how badly they can lead people astray. If he weren’t so consumed by this particular ideology, perhaps he would have seen the fundamental flaws in a paper that he instead tried to use as a bludgeon to belabor leftists. Talking about these blinders is a thankless and tedious task that has to be repeated over and over (there are many, many Augean stables worth of Death-Ponies, each its own industrious little producer). But it still needs to be done, I think.