When I first read the title of David Lipsey's recent Prospect article ("Too much choice," December 2005), my heart sank. I thought it would be yet another of the many recent polemics against extending choice in public services—but this time written by someone whose views I generally respect.
In fact, it turned out to be a thoughtful and balanced piece. Far from making a blanket condemnation of all forms of choice, Lipsey says choice in public services "may have an important role to play," and he endorses collective choice by primary care trusts, choice in social care through the direct payments scheme and individual choice of GP.
But of course, the general thrust of his argument is against more choice. In particular, he appears to dislike key aspects of current government policy, especially choice of hospital for elective surgery and parental choice of school. Here, as one of the architects of these policies, I have to take issue with him.
Although he later calls for experimentation and evidence, Lipsey's opposition to extending choice is formulated mostly on theoretical grounds. He has three basic arguments. First, the areas concerned are characterised by market failure, meaning that owing to externalities and poor information, individual choice cannot work as it does in a supermarket. Second, to extend choice properly requires the provision of excess capacity, thus creating waste and inefficiency. Third, the exercise of choice will increase inequality, with expert choosers choosing the better schools and hospitals, thus rendering them even better off while the worse off are left with sink providers.
The fact that there are long-standing private sectors in healthcare and in education suggests that the specific kind of market failure that he adduces cannot be too onerous a problem. Most healthcare does not generate externalities, and such that it does is best dealt with not by denying patients choice of hospital, but by providing the care free at the point of use—a principle which remains at the heart of government policy. Education externalities are best addressed not by denying parents a choice of school, but by compulsory school attendance and a national curriculum; both of which again are long-standing government policies.
Poor information is a genuine issue for choice—although no more so than with other methods of giving power to users, including their ability to exercise "voice" through measures such as complaints procedures or voting for school boards. Here it is worth drawing on the experience of the choice pilots in hospital care. These gave patients waiting for specific treatments at a particular hospital the chance to choose an alternative; patients would be given assistance with travel costs and a patient care adviser to assist them in making the choice.
The choice pilots were successful. Patients from all social groups particularly valued the information and advice role of the patient care adviser; and this idea is likely to be extended as the choice programme in health rolls out. The scheme has also prompted the proposals in the recent education white paper for choice advisers to help the least well-off make school choices.
As Lipsey points out, whether or not there are choice advisers, the information requirements of choice programmes will add to costs. But the amounts will be small compared with the gains in efficiency, equity and responsiveness that are likely to flow from having properly informed patients, parents and pupils. Take one illustration of those gains from the health pilots. In England—excluding London, where a choice pilot was in operation—in the six months ending March 2003, ophthalmology referrals fell by 2 per cent and waiting times by 6 per cent. But London saw a decrease in waiting times of 17 per cent: nearly three times the national figure, despite the fact that referrals in the capital rose by 5-6 per cent.
The excess capacity argument crops up a lot in the anti-choice literature. But as it stands, it is wrong. The healthcare choice pilots operated at a time of long waiting lists with no overall excess capacity; so does the current choice-based lettings scheme for social housing. In fact, what the health pilots demonstrated was how inefficiently the no-choice system uses the capacity that exists. One study of that system showed that 98 per cent of the population lived within one hour's travel time of 100 available and unoccupied NHS beds, and 76 per cent within an hour's travel time of 500 such beds. Making choice available permitted a much more efficient use of capacity.
What the excess capacity argument might be getting at is that choice does require there to be alternatives from which to choose. It is often argued that choice is an obsession of the metropolitan elite: that while it may be possible for people to have a reasonable choice of schools and hospitals in London, monopoly in the rest of the country is inevitable.
But again, this is a fallacy. The capacity study mentioned above showed that 92 per cent of the population had two or more acute NHS trusts within 60 minutes' travel time. And the same story is true for schools. Eighty per cent of mainstream state secondary schools in England have two or more schools within three miles of them. The average length of the journey to school for 11-16 year olds in England is three miles, and over half of English schoolchildren already do not attend their nearest school. So there are at least two alternatives to four fifths of English schools, attendance at which would entail little if any extra travelling.
What of the inequality argument? The first point is that there already exist substantial inequities in the no-choice world. Inequalities in intake and performance between schools where selection is based on catchment areas are well documented. In the no-choice NHS, studies have shown that instances of surgical intervention following a heart attack were 30 per cent lower in the lowest socioeconomic groups than the highest, while hip replacements were 20 per cent lower among the lower groups despite a 30 per cent higher need. This suggests that the middle classes are very effective at manipulating monopoly systems, where whether you get what you want depends on your ability to argue with the appropriate bureaucrat or professional—or, if necessary, on your ability to move.
So will choice make things better or worse? There is the danger of cream-skimming, especially in education: of popular schools choosing the more able pupils. But there is evidence from the US that school choice programmes improves performance all round—especially in previously poorly performing schools. So the incentives for quality improvement that choice and competition create appear to outweigh any effects of cream-skimming.
Lipsey discounts this work, saying that "the researchers found what they wanted." Well, they obviously didn't find what Lipsey wanted, but that hardly seems sufficient reason on which to base what is close to an accusation that they falsified their results.
Moreover, if cream-skimming does become a problem, there are well established ways of dealing with it. One possibility is a stop-loss insurance scheme whereby the government picks up the bill for very expensive patients or pupils. Another is not to permit providers any discretion over admissions. Yet another, and the most attractive, is to adjust the payment system so that providers have a positive incentive to take on difficult or expensive users.
Towards the end of his piece, Lipsey calls for the assessment of policies favouring choice on a case-by-case basis, using experimentation and evidence. But in the cases that he specifically criticises, the choice of hospital and school, there has been experimentation, and there is evidence. The results of the experiments, the evidence and indeed the theoretical arguments all point to the same conclusion. Policies designed to extend users' choices within public services are far from perfect. But they are likely to do a better job of empowering users and promoting quality services in an equitable fashion than a no-choice system where providers have a monopoly, people are chained to their local school or hospital and where their only "choice" if they are unhappy is to go private or to move house.
