June 6, 2012, by Jonathan
Efficiency: Real or Imagined?
A recent report by the think tank Centreforum has recommended reforms to the system of childcare in the UK with the aim of making it more affordable. BBC News – UK childcare needs to be more affordable – CentreForum.
One way of making it more affordable would be to decrease the quality of the service offered. But of course this is not what is proposed. The report makes suggestions about how to decrease the cost without decreasing the quality.
Whether or not its proposals are likely to achieve that aim, the report is quite rightly observing the distinction between merely cutting spending and achieving genuine efficiency savings. Sometimes this distinction is not observed in public discourse about ‘efficiency savings’: the phrase is used to describe any cut in spending. But you do not achieve efficiency gains merely by spending less, if that reduction comes at the cost of reducing the level of service or other aims of the spending. Cutting the libraries budget to zero, for example, would not be an efficiency saving.
Real efficiency savings occur when we can cut costs without reducing the desired output. But even this does not quite satisfy the definition of a Pareto Improvement–a concept used by economists and philosophers, named after the Italian economist and sociologist Vilfredo Pareto (1848-1923). A Pareto Improvement is a change that benefits some with no loss to anyone. In other words, it is the removal of pure waste. Most of the ways of achieving what we would normally call an efficiency saving–in which the broad level of service is maintained–do not satisfy this more demanding test. That’s because they involve redundancies, or changes in working conditions, that harm at least some people’s interests.
You might suspect that the concept of a Pareto Improvement has little application in the real world. But in fact such improvements are quite common in practice. The main reason for this is that people have different tastes or preferences. Suppose you love apples but hate grapefruits, and that I have the opposite inclinations. Then suppose that fate deals you a hand rich in grapefruits but poor in apples, and me the opposite. We can both get more of what we want, at no cost to others, by swapping or trading what we have.
Could it ever make sense to reject a Pareto Improvement when one is available? Some philosophers think we can have powerful reasons to do so even though it amounts to accepting pure waste. For example, some (not all) egalitarians are prepared to defend ‘levelling down’ on grounds of fairness. Critics of these forms of egalitarianism allege that this is an absurd position to take, just because it involves pure waste. However, it is worth noting–and less well-known–that ideas championed by the political right also provide reasons to reject Pareto Improvements.
Nobel Prize-winner Amartya Sen has demonstrated that respect for liberty can block Pareto Improvements. The reason is simple: what benefits someone depends in part on what he wants. And people often have desires about what other people do. Satisfying these other-regarding desires can then come into conflict with respecting the liberty of the object of the desires. As Sen showed, this can mean that we have to choose between making a Pareto Improvement and respecting individual liberty.
This shows that even when we can achieve a benefit to some at no cost to others, ideals espoused by many on the political right and on the political left imply that we might have other reasons for not doing so. These ideals attach importance to considerations other than benefit–considerations such as fairness, or respect for individual choice. Given that efficiency savings in the real world of public policy are not likely to be pure Pareto Improvements, we should remember that a simple appeal to efficiency is not necessarily as decisive in justifying political decisions as we might first think.
Unless I’m misunderstanding Sen then isn’t his characterisation of Pareto Improvements too restrictive? If Person X gains a benefit at no physical cost to Person Y, but seeing another gain in this way causes a decrease in Y’s happiness then that doesn’t seem a Pareto Improvement, because there is some cost to Y, namely a decrease in his wellbeing, even if he hasn’t physically lost anything, i.e. nothing was taken from him and given to X. Does the concept of Pareto Improvements strictly concern only tangible benefits and losses?
Thanks for your great comment, Edward. You raise a good question about how to understand benefits and costs. One could try to say that displeasure at someone else’s gain ought not to count as a real loss. But Sen was working with the standard way that economists understand the Pareto principle, namely in terms of any preference that a person has. Sen’s argument relies on the idea that people sometimes have preferences about what others do, or what happens to others (which is surely true). If one wanted to depart from this standard understanding, one would have to be able to draw a distinction between preferences that count and those that don’t, and defend its ethical significance. That might be possible, but it’s certainly not easy. For example, it doesn’t seem to correspond with the distinction between preferences that are self-regarding and those that are other-regarding — at least, not if we think that (for example) the preference a parent has that his child fare well should count for the purpose of applying the Pareto principle.
So, though you raise a good question, (1) Sen’s argument uses the standard interpretation of what counts as a cost, and (2) defending an alternative interpretation is no easy task.