Trahison Des Clercs

Morality splurge on the issue of the death penalty is always rattling away at fever pitch. We get the full political range, from “its state murder” glumness to “eye for an eye” preaching. The economics approach, admittedly with only partial success, tries to avoid such handbag haggling. There is genuine attempt to demonstrate whether the introduction (or continuation) of capital punishment is an efficient response to the problem of crime. Efficiency here refers to the attempt to find some notion of an optimal crime rate, where the diddy dismal scientist can take into account the benefits and costs associated with changes in the criminal justice system. But can we support capital punishment with the tools spawned?

Let’s hop and skip through the weak arguments in order to jump to the important argument. The pro-capital punishment commentator will often, due to a failure to consider the nature of economics, focus on two minor issues: recidivism and accountancy costs. Recidivism refers to the tendency to lapse into previous behavioural patterns. The death penalty can stop the more aggressive forms. No executed prisoner will have the opportunity to kill again after all. We then have a justification to execute the most heinous killers to prevent them from killing again. Of course the recidivism rate for murderers tends to the minor side and we can minimise such problems via more severe prison sentences. After wasting our time with recidivism, the pro-capital punishment commentator will find himself a grey suit and some comfortable slip-on shoes in order to refer to the accountancy of the issue. Simply put, he’ll crow that we can avoid the financial burdens of housing inmates for long prison sentences. Unfortunately the accountancy angle doesn’t hit home in the useful utterance. The constant appealing, in the hope that the innocent aren’t ‘accidentally’ executed, ensures accountancy ambles are second-rate.

The decision to use the death penalty must rest on the magnitude of any deterrence effects. These effects are based on a marvellously simple way of understanding criminal behaviour. The criminal is a simple rational agent who will, by taking into the account of costs and benefits, will attempt to maximise their satisfaction. We all become potential murderers, as we encounter the question “Does the expected utility from committing murder exceed the expected utility from alternative action?”. The argument for capital punishment then looks bullet proof: We must raise the price of murder for potential criminals in order to get less of it. The death penalty raises the price of homicide as long as execution is deemed to be more utility harming than life imprisonment.

The empirical testing of the theory looks like a hallelujah moment. Ehrlich(1975, 1977) uses econometric methods to estimate “murder supply equation”. His time series study suggests 1 execution deterred 7-8 murders; the cross section study suggests the effect is stronger at 20-24 murders. However, we must always be careful of such evidence. It is not possible to prove the deterrence hypothesis, given findings are reliant on the exact econometric techniques adopted. Thus, it is possible to use Ehrlich’s data and sample period but, by adopting different empirical specifications and functional forms, reach divergent conclusions. It becomes even more worrying when we note that there is a likelihood of empirical bias (either in favour or against) as the researcher uses his/her morality to deliberately bias results by choosing a specific econometric methodology that increases the probability of achieving the results they desire.

One might argue that it’s the anti-capital punishment knee-tremblers that are more prone to this bias problem. We can refer to the notion of a positive income elasticity of demand for humanitarianism. Thus, as our society sees increase in income we may move away from economically rational choices due to perceived fears of brutalisation. But what if we are to take that mighty leap and just accept the literature in favour of deterrence effects? Things aren’t actually made more straightforward. Rather disturbingly, the empirical analysis suggests that deterrence effects increase the quicker you execute. Thus, to maximise deterrence we'd have to maximise the number of “mistakes”. It also suggests only certain forms of execution work (at worst ensuring those bias doubts are fog horned back into our mind). We also should consider potential negative encouraged by the death penalty. For example, the marginal cost from committing multiple murders is eliminated if the death penalty is given to 'joe average' murderer. It therefore delivers perverse incentives to increase your kill rate.

Given the uncertainty over deterrence effects, it would seem prudent to maintain a non-capital punishment policy. Basic sense dictates!