As children start their lessons in the 24 Free Schools opening this week, a new experiment begins in English education. The founders and staff will have been working hard for this day over many months and no doubt all will wish the pupils and staff well. There has been a lot of political passion on both sides of the debate, but what is the significance of the Free Schools experiment likely to be?
It is not an experiment because of the “free” part – Free schools will enjoy essentially the same freedoms as Academies do. It is an experiment because now anyone can propose to set up a school, and attract the same per-pupil funding from the state as other local schools. Anyone can propose … but the vetting process to check whether the applicants are “fit and proper” to run a school is (rightly, in our view) so stringent that around 90% of the original applications were turned down.
Will the experiment work? Work for whom? The focus cannot just be on the 5000 or so pupils starting now, less than a twentieth of 1% of their cohort. If so, this would be an incredibly expensive and high profile way to change the education chances of such a small number.
The focus has to be on the systemic impacts of the reform. What impact will the new option of opening a Free School have on the local schools?
The main idea, the big hope for the policy, is that this will raise the “competitive threat” to under-performing local schools. Economists talk about market entry: the theory that the threat of new suppliers entering a market works to keep the incumbents efficient and responsive. Any un-served niches or excessive costs would draw in new companies to offer a better deal. Arguably, the prime function of the Free Schools experiment is to make the schools market “contestable”. This might sharpen the efforts of schools to raise their game to avoid losing their pupils to the new schools.
This might work. It is conceivable that really poorly performing schools may not wish to see their school supplanted by a new entrant, and push to raise attainment standards. But the UK evidence on the effects of competitive pressure on school effectiveness is not encouraging. Put at its best, there is only very weak evidence for small effects of competition. This is not true in every country, nor in every service in the UK, and there appear to be important market structural reasons why competition does not work to raise standards in schools.
This is not at all to say that all competitive mechanisms are pointless. Clearly any school system needs a strong policy for dealing with failing schools, something to act as a discipline on the performance of Headteachers and school governors. Free schools offer this in principle: in extremis, parents very dis-satisfied with their state school can opt out and set up their own school. But there are two reasons why Free Schools are unlikely to be the best answer to this. First, there are very significant set-up costs, both in time and energy from the founders, but also in the straightforward sense of acquiring premises. While currently these are being funded (very generously) by the government, this simply cannot continue if the policy matures and spreads. But secondly, it seems inconceivable that any local area with one Free School would be offered the resources for any others. So as discipline device, this is a one-shot game, not an on-going continued pressure on low performing schools, which is what is needed.
Despite the first rush of enthusiasm, it seems unlikely that Free Schools will act as a major stimulus to systemic higher performance. Perhaps the main problem was not the barriers to entry preventing many highly competent and motivated school-starters from setting up. In which case, we will need other policies to address poorly performing schools in disadvantaged neighbourhoods.
There are other important potential systemic effects, interactions between Free Schools and their neighbouring schools.
The first is school admissions. Free Schools will take part in coordinated admissions and, although, like Academies, being their own admissions authority, they will have to adhere to the Admissions Code. The big issue is whether the founders have any special rights to get their own children into ‘their’ school. It may be that this is one aim of the clause in the new draft Admissions Code allowing schools to give admissions priority to the children of “staff”. This is a difficult issue: it is quite understandable that those putting in time and effort to set up a new school want to get their own children in. But there are dangers: if this becomes the norm, we have a new route for particular individuals to get their child into a good school, by founding or acquiring a stake in, a Free School. In any case, the admissions criteria and the highly related question of how to pass on a “stake” in a Free School need to be clarified.
Another important spill-over effect concerns funding. The on-going current expenditure of Free Schools is funded on the same basis as Academies, so is not much different from maintained schools (just differentiated by not having to pay the local authority for central services). But the capital funding is another matter. Funding the acquisition and refurbishment of premises, estimated by the DfE at £120m, has been a substantial drain on a much-reduced schools capital budget. The funding for Free Schools’ buildings is a direct diversion of funds from other capital funding projects, many in far more needy schools and disadvantaged areas than the Free Schools.
But finally what Free Schools are likely to offer is scope for pedagogical innovation. This will not always be the case – some of the groups setting up schools may have rather conservative ideas on what constitutes a good education. But others will likely be set up by innovators with radical ideas on how to teach.
So perhaps that’s what Free Schools will end up being for: not really working as a spur to higher standards but acting as incubators for radical new teaching ideas. Very valuable indeed, but perhaps not what everyone was expecting.
I have to admit to being stunned by the level of misinformation that is currently accompanying the Health and Social Care bill as it is introduced into Parliament. On Jan 19th alone, the shadow health secretary John Healey stated that “the changes would make the health service profit centered rather than patient centered”, the health secretary Andrew Lansley said “competition would be on quality and not cost” and that as the health service is free at the point of delivery patients obtain the best medical outcome rather than the cheapest option, while Karen Jennings, head of health at Unison, stated that the only survivors will be the private health companies which are “circling like sharks” and MPs say “the reforms have taken the NHS by surprise”.
None of these statements has much basis in either fact or evidence. The reforms being introduced in the bill are essentially a continuation of the reforms started under the previous administration, albeit at an increase in pace and scale. The Labour reforms introduced competition between hospitals for patients and patient choice of hospital and a system of regulated prices. Lansley has changed the buyers of health care from local PCTs to General Practitioners, but under Labour the PCTs were supposed to act on behalf of their local GPs anyhow. Why Healey believes that increasing the pace of reform and replacing the PCTs with GP consortia should mean the NHS switches from being patient to profit centered is completely unclear. GPs have not been seen by politicians as profit centered previously. In fact, perhaps because GPs see so many voters each week, most politicians studiously ignore the fact that GPs are private contractors and not NHS employees. In addition, the new GP consortia will probably employ a fair number of ex-PCT staff. So it seems unlikely there will be a radical shift in values on the purchaser side.
On the other hand, the current secretary of state is also being somewhat disingenuous in his statement that because the NHS is free at point of delivery, competition will be in terms of quality. He rather forgot to mention one key change he has made, which is to abolish the fixed price tariff introduced by the previous administration, The fixed price tariff was introduced on the basis of evidence from the UK (as well as elsewhere) that competition in health care with fixed prices avoids a potential race to the bottom in which sellers of care compete to attract patients on the basis of lowest cost at the expense of quality In his desire to push forward his competitive model, Lansley has thrown away the fixed price tariff and does indeed risk such a race to the bottom. The fact that care will be free to the patients is irrelevant, as the gains from price savings will accrue not to individual patients but to GP consortia that face the pressures arising from fixed budgets and a tight financial settlement. Exactly the same model was employed in the NHS in the internal market of the 1990s and in that market, cash constrained buyers focused on price and reducing waiting lists, at the expense of quality.
Karen Jennings’ statement appears to have even less basis in fact. Again, the plans to allow any willing provider to supply care to the NHS were actually introduced by the Labour administration. There has been relatively little entry of non-NHS providers not because these providers were not allowed to enter but because they didn’t find it profitable. There is no a priori reason why in a more constrained financial era that supplying care to the NHS should make private providers large profits. It is true that GP consortia will probably seek the help of the private sector to carry out their commissioning function but given the poor performance of some PCTs, this may simply allow an increase in talent on the purchasing side of the NHS. And given that the purchaser side has been weaker than the provider side for a long time, this is probably a good thing for patients.
Finally, the fact that MPs think that these reforms have taken the NHS by surprise seems to suggest that MPs don’t notice things until they come to the House. These reforms were trailed very soon after the Coalition government came to power and again in a White Paper this autumn. All the major components of the bill were in that White Paper and many NHS bodies have been preparing themselves for another period of rapid change. What many MPs may not like is the speed of change - and there are good reasons to question this. For example, it is not clear that replacing PCTs with GP consortia at this stage in the reform process will help develop the gains that competition between suppliers has been shown to have had and creating a new set of purchasers will undoubtedly take a lot of attention, resources and time. In my view these resources would have been better used in developing choice to a greater degree, getting the rules of the game right and then introducing, at a later date, a large role for those GPS that want it.
The Government has announced is to permit hospitals to compete on price. This dramatic shift towards a more commercial market in the NHS is announced in a single paragraph in the NHS Operating Framework for 2011-12 published last week. Paragraph 5.43 says: “One new flexibility being introduced in 2011-12 is the opportunity for providers to offer services to commissioners at less than the published mandatory tariff price, where both commissioner and provider agree.” It adds: “Commissioners will want to be sure that there is no detrimental impact on quality, choice or competition as a result of any such agreement.”
This problem with this innocuous sounding paragraph is that we have been here before and it doesn’t work. The reason is that in health care, prices are easy to observe, whilst quality is not. In these circumstances, health care commissioners and sellers, despite the hopes of the Operating Framework, end up focusing on price, raising the prospect of two-for-one deals on surgery and cut-rate consultations for certain specialties. At the same time, in order to provide services at these prices, quality suffers.
Research from CMPO showed this is precisely what happened in the internal market of the 1990s, where price competition was allowed. Hospitals facing more competition focused on bringing down costs and lowering waiting list in order to provide what their commissioners wanted. But this was at the expense of quality and the consequence was that patients in hospitals located in competitive markets were more likely to die after an admission following a heart attack. These kind of unforeseen consequences are ikely to happen again – especially now when budgets are tight.
In making this move, Andrew Lansley is ignoring all the evidence on the impact of price competition in the hospital sector and is potentially endangering patients lives.
Governments faced with rising costs and growing demand are constantly searching for methods of delivering higher productivity in health care, or put more simply, ways of getting higher quality without increasing expenditure. One currently favoured mechanism is to encourage competition between the suppliers of care. But will this work? The appeal is simple – competition works in the rest of the economy therefore it should work in health care.
Unfortunately for politicians, the simple appeal does not necessarily translate across sectors of the economy. There is, in fact, no strong theoretical support for competition in healthcare leading to better outcomes: the predictions of economic theory on this issue are quite ambiguous (1). However, under certain condition, theory models do support competition: this is when prices are fixed by government and hospitals compete in terms of quality.
Testing this theory is often difficult, because competition in health care markets is endogenous to quality. The presence of a high quality hospital may mean that competitors stay out of its market. Or hospitals which are cutting edge tend to be located in urban areas and also attract sicker patients. In both of these situations it will appear that competition is associated with lower quality. Dealing with this statistically is not easy without some kind of natural experiment, as case mix is very difficult to measure precisely.
The English National Health Service (NHS) is subject to frequent policy change as politicians use health care as part of their drive to win supporters. The last Labour administration introduced competition between health care providers as part of its drive to increase productivity in health care. In 2006 the government mandated that all patients must be offered the choice of five, and by 2008 any, hospital in the NHS for their treatment. In addition, the prices that hospitals could charge were fixed by the government in a ‘yardstick competition’ type regime.
This policy change provided a natural experiment that researchers could exploit. Hospitals compete in geographical markets because patients prefer to be treated, inter alia, closer to home. Hospitals thus vary in the extent to which they face competitive forces simply because of geography. Exploiting this fact allowed researchers to look at outcomes pre- and post- the competition policy across different markets.
The research looked at all admissions to hospitals in the NHS – around 13 million admissions – pre- and post-policy. It found that hospitals located in areas where patients have had more choice since the NHS reforms had higher clinical quality – as measured by lower death rates following admissions – and shorter lengths of stay than hospitals located in less competitive areas. What’s more, the hospitals in competitive markets did this without increasing total operating costs or shedding staff. These findings suggest that the policy of choice and competition in health care can have benefits – quality in English hospitals in areas in which more competition is possible has risen without a commensurate increase in costs (2).
One reason that the policy may be having this impact is the fact that prices are externally fixed. Research for the UK showed that when competition was introduced in the early 1990s in a regime that allowed hospitals to negotiate prices as well as quality there was a fall in clinical quality in more competitive areas. Waiting lists, however, declined for these hospitals. This is supported by economic intuition. Where quality is hard to observe, the elasticity of demand will be low. Waiting lists on the other hand are easy to observe. Buyers of health care will therefore have a greater elasticity of demand with respect to the latter than the former and so suppliers will tend to compete on the latter and on price, whilst shaving the less well observed clinical aspects of quality (3).
These results suggest that the details of the policy matter. Competition under fixed prices appears to have beneficial results whilst competition where hospitals bargain over price and quality do not. This in turn has policy implications for governments who are keen on market forces in health care. If competition is to work, price regulation has to be retained. A free-for-all in prices would mean a return to the “internal market” of the 1990s, a regime in which hospitals competed vigorously on waiting times and ignored aspects of quality that are more difficult to measure. In addition, the tendency of the UK government to merge failing hospitals needs to be looked at carefully. Mergers are popular with finance ministries in NHS type systems because they remove what is often seen as ‘excess capacity’. However, while there are gains from removing poor managers when a hospital fails (4), removing capacity by merger will limit the extent of competition and may stifle the impetus given by competitive forces to improve outcomes for patients.
- Gaynor, Martin. 2006. “Competition and Quality in Health Care Markets.” Foundations and Trends in Microeconomics, 2(6): 441-508.
- Gaynor, Martin, Moreno-Serra, Rodrigo and Propper, Carol. 2010. Death by Market Power: Reform, Competition and Patient Outcomes in the National Health Service. NBER w16164.
- Propper, Carol, Simon Burgess, and Denise Gossage. 2008. “Competition and Quality: Evidence from the NHS Internal Market 1991-9.” The Economic Journal, 118(525): 138-170.
- Bloom, Nicolas, Propper, Carol, Seiler, Stephan and Van Reenen, John. 2010. The impact of competition on management quality: evidence from public hospitals. NBER w16032.