There was an interesting comment from Andrew Coleman on a recent post of ours in which he laid out the reasons he sees for government intervention in saving behaviour. As I read it he put forward four reasons:
- People are time inconsistent
- People are poor at choosing good investment options
- People underinsure against catastrophe
- People take advantage of government intervention
The first three pertain to cognitive biases that may justify intervention to correct. There is widespread debate about whether such biases warrant intervention and, if so, what sort. However, what I want to draw attention to here is the fourth point: when governments intervene they often give people perverse incentives to take advantage of the system that then require further, costly intervention to remedy. Even if the intervention seems justified it may be that the mechanism implementing it is so costly to administer or enforce that the intervention ends up being costly to society.
On this blog we seem to spend a lot of time talking about whether it’s appropriate for governments to intervene in markets but what we’re talking about is only the first step in justifying regulation. We’re only suggesting that the markets are inefficient but the next step — that any policy analysis would have to consider — is to ask whether there is a practical means of intervention that is less costly than the benefits it generates. It may well be that the costs of the policy, whether in administration or the incentives it creates, are greater than the cost of the market inefficiency.
In particular, think about the four reasons for intervention and suppose there was a benefit to society from intervening to correct the first three and that intervention was administratively costless. Nonetheless there would be a cost to society from the perverse incentives created by the intervention. Now that might require further regulation to correct and that regulation would itself be costly. When the government justified the second round of regulation it might point to the inefficiencies generated by the perverse incentives and claim that it was fixing a problem. But that would not be correct comparison: the correct counterfactual is the situation where there was no intervention at all. The original intervention would only be justifiable if it provided a net benefit once all of the costs of the successive rounds of patching up problems were taken into account.