Saturday, 2 June 2018

Political risk aversion

Few bureaucrats get fired for being too cautious about health and safety. Maybe this latest Housing New Zealand debacle will change that.

Recall our base theory here, going back to the kind of stuff Alex Tabarrok talks about. Consider a government agency trying to mitigate some kind of risk. It can screw up by being too cautious, or by not being cautious enough. We should all want it to minimise the expected cost of error, so to weigh the costs of getting things wrong in either direction. But the political incentives go the other way. If you err on the side of being too cautious, putting more weight on health and safety risks, for example, you'll typically impose costs across the whole sector that aren't all that visible - everybody pays $10k more than they should on a house repair because of really stupid scaffolding rules, for example. But nobody gets fired. If the bureau errs the other way, and somebody dies for lack of caution, then John Campbell's all over it and the Minister starts busting heads down the Ministry because the Minister's incurred political costs.

It's made worse by health and safety liabilities on directors, where they can take on pretty substantial cost if they're not sufficiently risk-averse.

And then we wind up in the meth-mess, where Housing New Zealand imposed stupidly risk-averse standards for determining meth contamination and so the whole sector wound up bearing a lot of cost - and a pile of people got needlessly kicked out of their houses.

Blaise Drinkwater's summary seems best here.



I think we need to figure out ways of making the public sector, and regulation more generally, bring the real costs and political costs of the two kinds of error into line.

This episode might help in that - there are, in this case, political consequences for having been too risk averse around health and safety. The Ministries behave as though they expect massive penalties from the Minister for not being sufficiently risk averse, but no particular consequence for being too risk averse. Those expectations have to change or we'll keep getting excessively risk-averse Ministerial interpretations.

No comments:

Post a Comment