MA Chapter 2

Chapter 2: Why introduce markets for schools?; marketisation of schools in a Lakatosian scientific research programme framework

2.1 The Lakatosian Methodology of Scientific Research Programmes (MSRP) framework

In a radio broadcast on 30th June 1973, Imre Lakatos (1978a) succinctly introduced his methodology of scientific research programmes (MSRP) to the public. This methodology was to be a framework for the analysis of developments in science. He argued that scientific achievements should be viewed as sets of programmes rather than solitary theories and hypotheses. For example, Lakatos (1978a) argued that Newtonian science consists not only of the ‘hard core’ of the three laws of mechanics and the law of gravitation, but also a ‘protective’ belt of auxiliary hypotheses that guards the ‘hard core’ against refutation. ‘Heuristics’, in either positive and negative forms, then guide scientists in the Newtonian SRP, to solve anomalies which threaten both the ‘protective belt’ or the ‘hard core’. At any stage in any research programme, Newtonian, Einsteinian, Marxism, or Copernican, etc., there are always unsolved anomalies and problems. Therefore, as Lakatos (1978a) asserted,’(a)ll theories, in this sense, are born refuted and die refuted’.

MSRP has since been used as a tool to analyse the history of economic thought and to examine the development of economic theories and programmes. For instance, Blaug (1980) argued that the neo-classical human capital model has the characteristics of the Lakatosian SRP and defined its ‘hard core’ to be “People spend on themselves in diverse ways not only for the sake of the present enjoyment but also for the sake of future benefits.” while the various human capital theories served as its ‘protective belt’. Elsewhere, Latsis (1972) and Simon (1978) described and compared competing research programmes related to the theory of the firm: the former dealt with ‘situational determinism’ versus ‘economic behaviorism’, while the latter juxtaposed the ‘substantive rationality’ of economics with the ‘procedural rationality’ of psychology.

‘Markets for schools’ can and will similarly be explored by scrutinising it under the microscope of the MSRP. Before going on to place the school marketisation programme into the SRP framework, it is necessary to define and explain the various elements of the SRP.

The heart of each SRP is its hard core. This hard core is a set of embedded assumptions that all workers in the SRP consider to be irrefutable, according to Lakatos (1978b). The hard core’s protective belt however is refutable, and is indeed expected to bear the brunt of any refutations. The protective belt is adaptable and continually changing, and modifies itself, such as the replacement of its composite theories, to explain away the refutations and account for any anomalies. Anomalies and refutations exist in each and every SRP. These are phenomena which have not been accounted for or explained.

As mentioned earlier, there are two types of ‘heuristics’: negative and positive. These are both methodological rules. Negative heuristics prescribe paths of research to avoid, and ‘involves the stipulation that the basic assumptions underlying the programme at the hard core must not be rejected or modified’ (Chalmer, 1983). Negative heuristics .forbid any questioning of the hard core and redirect any refutations to the protective belt. On the other hand, positive heuristics stipulate what paths to pursue and how the research programme should be developed. In Lakatos’ (1978b, page 49) words,

‘The positive heuristic consists of a partially articulated set of suggestions or hints on how to change, develop the “refutable variants” of the research programme, how to modify, sophisticate, the “refutable” protective belt.’

Also, the positive heuristic

‘(D)efines problems, outlines the construction of a belt of auxiliary hypotheses, foresees anomalies and turns them victoriously into examples… (and) (i)t is primarily the positive heuristic of (the scientist’s) programme … which dictates the choice of his problems ’ (Lakatos, 1978c, page 111)

SRPs may be progressive or degenerating. In progressive SRPs,

‘theory leads to the discovery of hitherto unknown novel facts. In degenerating programmes, however, theories are fabricated only in order to accommodate known facts’ (Lakatos, 1978a, page 5)

Competing and conflicting SRPs may co-exist simultaneously, and a scientific revolution then comes about when (the degenerating) one is abandoned for (the progressive) other. Refutations, given that they permeate all SRPs, are not responsible for the revolution. Instead,

‘(w)hat really count are dramatic, unexpected, stunning predictions: a few of them are enough to tilt the balance; where theory lags behind the facts, we are dealing with miserable degenerating research programmes’ (Lakatos, 1978a, page 6)
2.2: ‘Markets for schools’ as SRP

Having defined the components of the SRP, I will now suggest what they are in the programme of ‘markets for schools’. Clearly, as Leijonhufvud (1976) pointed out, the definition of any SRP is likely to be contestable, because different workers inside, as well as analysts outside the relevant field are likely to have different perspectives of it. For example, he raised the example of Milton Friedman, who opined that monetarists and neo-Keynesians share essentially the same theory, and that all their differences derived from disagreement over the magnitude of particular empirical parameters; other monetarists and neo-Keynesians would clearly feel that their differences are far more fundamental than that. Also, the economic liberal Jevons, coming from a different ideological perspective from the libertarian Friedman, would probably support school markets for different reasons, and see different school choice programmes as useful. What they see as the hard core, positive heuristics, etc. therefore may vary, at least in degree.

The hard core
Hard core preposition 1: Parents and pupils maximise their utility by maximising educational opportunity and outcomes; choice is inherently utility enhancing.
Hard core preposition 2: Agents, with access to sufficient knowledge, can and will make rational decisions.
Hard core proposition 3: Parents and pupils maximise their utility by choosing the school which will maximise the child’s educational opportunities and outcomes.
Hard core preposition 4: Teachers, headteachers and other educationalists working in schools maximise their utility by maximising their income (by maximising pupil number).
Hard core preposition 5: Education is a commodity that may be traded just like any other.
Hard core preposition 6: The market is a useful tool, with which social justice may be achieved in schools.

The first five of these prepositions are first principles, from which the sixth preposition may be derived. Liberal and libertarian economists may fervently believe that like hammers and nails, the market is a tool that may be used to achieve a given goal, not just as a matter of irrefutable logic, but also ideology.

Social justice, if defined according to Konow’s (2003) integrated justice theory, would include not only Need and Equity but also the principle of Efficiency.

The Need principle is with reference to basic needs, and deems an allocation socially just if basic needs are provided for equally across individuals. In education, for instance, an allocation that allows for a socially and politically determined minimum level of literacy and skill sets across all individuals may be considered just according to this first principle. Included under this broad Need principle are ideas of egalitarianism and Marxism, as well as Rawls’ Liberty and Difference Principles.
The Equity principle, that an allocation is fair if individuals’ outcomes are proportionate only to inputs that they control, stems from Aristotle’s distributive justice theory and Locke’s desert theory (Konow 2003). Distributive justice is the doctrine that a decision is socially just if all parties receive what they need or deserve. It is often contrasted with procedural justice. The former concentrates on just outcomes, while the latter concentrates on just processes. Desert theories essentially identify factors that might be considered fair to use in the determination of economic distribution. For example, Buchanan (1986) distinguishes between luck, choice, effort and birth. He considered the distribution of economic outcomes according to effort as the least controversial and believed that conflicts with common notions of justice would only come about with inequalities that are caused by serendipities of birth. Therefore, for Buchanan, it would be socially just if educational outcomes were such that those consumers of education (i.e. pupils) who put in the most effort would derive more benefits, and it would be grossly unjust if certain consumers derive more benefits from the educational process merely as a consequence of social class, gender or colour. Linked to Buchanan’s desert theory are Musgrave’s (1959) concepts of horizontal and vertical equity. Horizontal equity refers to the equal treatment of equals, i.e. if individuals A and B are socially deemed to be identical in pertinent and essential characteristics, they should have the same access to the same educational opportunities, to attain the same educational outcomes. Vertical equity is the unequal treatment of unequals. Therefore, if individual A puts in less effort in his education (and gets lower GCSE grades for example), it is perhaps vertically equitable for A to receive a smaller financial option return (among other benefits captured by students directly) and have a relatively restricted access to higher education. Also, it might be considered socially just if children with special educational needs (SEN) receive more funding. Inequality, therefore, might be equitable.
Konow’s third principle, Efficiency refers to a relationship between inputs and outputs, and is achieved by either a) maximising the value of outputs given the value of inputs or b) minimising the value of inputs given the value of output. When a situation is said to be inefficient, this means that the desired means could be attained with less inputs, or that the means utilised should be able to produce more of the desired ends. People are assumed to be homo economicus, seeking to maximise surplus, or output over inputs, and this goal, that of achieving the greatest good for society, is deemed to be a form of fairness. In this sense, efficiency is in itself a component of justice, and not necessarily at odds with social justice as a whole.
The positive heuristic

These positive heuristics generate theories and ideas supportive of markets that go into the protective belt:

Positive heuristic 1: Explain how markets can be used to improve social justice.

Positive heuristic 2: Propose mechanisms with which markets may be used to improve social justice.

Positive heuristic 3: Explain why and how the state monopoly has failed to attain social justice.

Positive heuristic 4: Provide empirical evidence in support of the first three positive heuristics above.

The positive heuristic may be viewed as a ‘recipe’ or set of ‘recipes’, with which to concoct the protective belt ‘broth’.


The protective belt

Protective belt preposition 1: Markets give the working class the option of exit, which improves equity.

Proponents of markets have also claimed that contrary to intuition, there may not be the conventional economics efficiency-equity trade-off and equity may actually be increased instead of decreased in a school market system. Using Hirschmann’s (1970) concepts of ‘voice’ and ‘exit’, Tooley (1995) argued that markets reduce the power of the middle-class vis-à-vis the working class. In any organisation, customers may express their dissatisfaction with its performance either ‘exit’, i.e. by leaving and stopping their custom of the organisation’s products, or by ‘voice’, i.e. by ‘express(ing) their satisfaction directly to management or some other authority to which management is subordinate or through general protest to anyone who cares to listen’ (Hirschmann, 1970). In an educational setting, this means that parents or pupils may express their displeasure with a school either by leaving it for another school, or by complaining to the head teacher, government or to the press. Under a non-market, bureaucratic system, ‘voice’ tends to be the only available recourse, while in a market system, both ‘voice’ and ‘exit’ are viable options. In the former system, the middle class can be expected to dominate, since it is ‘politically influential, skilled and adroit, …organised, … more articulate… (and) endowed with cultural power’ (Seldon, 1990) and are more capable at manipulating ‘voice’. The denial of markets to the working classes means that ‘exit’ is not permitted to them, and they are stuck with only ‘voice’, in which they are seriously disadvantaged. Markets, by reducing bureaucratic procedures, such as LEA admissions rules, will enable them to make choices previously not available. Though Gewirtz et al. (1995) and Willms & Echols (1992) may argue that the middle class are relatively more adept at working the market, at least markets give the option of ‘exit’, where there was none before, to the working class, and allow them to ‘vote with their feet’ (Tiebout, 1956) and to leave poor schools for better ones.

Protective belt preposition 2: Government failure is pernicious.

Government failure, it has been argued, can be as pernicious, if not more so than market failure. According to Barr (2004), the public-choice literature points to two distortionary factors that may explain governments’ failure to maximise social welfare. Firstly, there is the distortionary response of government to coalition groups of voters and pressure groups, which may come about because, for example, the middle class has an inordinately strong electoral power as the median voter (Tullock, 1970), or governments may be elected with only a minority and therefore only need to pander to a small but allied coalition of voters. Secondly, bureaucrats themselves exert distortions, since they may run public agencies at least partly for their own benefit and with their own goals, and with considerable ‘organisational slack’. Also, bureaucracy has been closely linked, arguably not unfairly, to inflexibility in decision-making, un-necessary duplication, red tape, high internal transaction costs and inefficiency.

Protective belt preposition 3: Markets, competition, and choice improve the efficiency, effectiveness and standards in schools

Markets are believed by supporters such as Friedman & Friedman (1980) to be the key means to improve standards in education. They are thought to provide the right incentives: successful and therefore popular schools are rewarded with increased enrolment and funding, while unsuccessful schools are forced to change and improve or face closure. Increasing the range of types of schools and enhancing diversity may encourage education to break free, innovate and push for greater effectiveness. As Gorard (2003) pointed out, the desire for greater diversity stems from ‘the perceived failure of state-funded monopolies of schools, and the differential effectiveness of sectors and school types’. Markets are thought to be efficient, while state monopolies are bureaucratic, slothful and wasteful.

The anomalies

Anomaly 1: There has been little evidence that efficiency has indeed been improved since the introduction of markets in the English school system.

There appears to be very little clear-cut evidence that efficiency has indeed been improved as a result of the market based reforms. Instead, the autonomy and flexibility given to the grant-maintained schools has neither led to a significantly different pattern of resource allocation (Fitz et. al, 1993), which could indicate the presence of innovative behaviour, nor to significantly better educational outcomes (Levacic and Hardman, 1999). The relationship between schools’ quality of resource management and their educational outcomes has also been shown to be only weakly positive (Levacic and Glover, 1998).

DfEE (1998) statistics may show that the percentage of pupils getting 5 good passes at GCSE and its equivalent level has not only increased year-on-year since 1975 to 1998, but also that the increases have been larger since the late 1980s, i.e. the implementation of the ERA. But as Gorard (2001) argued, it is impossible to simply attribute this raw-score improvement to market forces per se, given that there were other policy changes happening at the same time. Any of these could have been equally plausible as the, or one of the, causes of GCSE improvements: replacement of the CSE and the GCE in the 1986/87 academic year with the arguably easier GCSE, and the abolition of strict norm-referencing which has previously served to keep results relatively constant (Foxman, 1997). Moreover, as efficiency refers to a relationship between inputs and outputs, it is insufficient to examine just the outputs of educations, but also the inputs. If educational output has increased as a result of a proportionate, or even more than proportionate increase in the usage of inputs, then efficiency may hardly be considered to have improved.

Anomaly 2: Economists and sociologists have identified many mechanisms through which social inequality has been argued to have been negatively impacted by the market.

A major concern among economists and sociologists is that of cream skimming (Glennerster, 1991). Since a school’s performance in the league tables is determined largely by socio-economic background and schools in greater demand can attract more pupils and hence more funds, a school may attract more funding by maximising the examination results of its pupils and it can do that at minimum cost by cream skimming students from higher socio-economic backgrounds. Evidence suggests that worries in England about cream skimming are real. Schools have been shown to seek out and select particular types of students, e.g. those from the middle class, or from South Asian backgrounds, based on their abilities to boost test scores, at the expense of the ‘less able’, such as children with Special Educational Needs (SEN) (Gewirtz et al, 1995), etc. A significant minority of schools, especially those with autonomy over admissions, i.e. voluntary-aided and foundation schools, had criteria which appear to cream skim, despite Labour government reform attempts since 1997, such as the Schools Standard and Framework Act 1998, and the introduction of the Code of Practice on School Admissions, to counter this phenomenon (West et al., 2004). More specifically, schools have been using interviews meant to establish the pupils’ religion to determine the presence or absence of other desirable characteristics, which might partially explain the different percentages of pupils eligible for free school meals in different types of schools: by religion, 11.4% for Church of England schools, 15.6% for Roman Catholic schools, 6.2% for Jewish schools and 6.5% for Sikh schools, compared with 16.1% for all other maintained secondary schools in England (House of Commons, 2001).

Social inequity may also be negatively influenced because privileged middle class parents, with their cultural capital and educational knowledge, are better than their working class counterparts at discriminating between schools, evaluating teachers, and interpreting league tables and Office for Standards in Education (Ofsted) reports and therefore come up trumps in the schooling markets (Ball et al., 1996). In England, the government, under New Labour’s ‘Fair Funding’ for example, does provide higher per capita funding for schools with a higher proportion of socially disadvantaged pupils, but it is difficult to tell, especially given the complexity and opaqueness of the formula used, how much this additional funding constitutes, and whether it is sufficient to compensate for the working class’ relative disadvantage.

Yet another concern is that with the development of markets in education, schools may become increasingly pressured to restructure teaching, such as by introducing setting, in order to attract white middle class school children, who are viewed as ‘valuable commodities’ (Reay, 1998). Research has shown that there are negative consequences for pupils allocated to bottom sets since, for example, teachers for bottom sets tend to be less experienced and junior staff (Ball, 1981). Because pupils in bottom sets are predominantly black or white working class, and those in top sets may be ‘uniformly white’ (Gillborn and Gipps, 1996), this suggests that social inequalities based on race and social class are being exacerbated here. Indeed, when schools are ranked according to their percent of pupils achieving A-C grades, Gillborn and Youdell (2001) have observed an ‘educational triage’ syndrome, whereby schools ration their time and effort, and focus on pupils on the borderline between the D and C grades, at ‘the cost of judging some pupils (disproportionately Black and working class young people) as without hope’. Schools do not deliberately set out to further disadvantage the disadvantaged. Instead, to keep their place in the pecking order, schools have to maximise their measured output from limited resources, and therefore have no choice but to decide where best to concentrate their efforts upon. This decision is based on judgements on ‘ability’, which unfortunately, like IQism, is a ‘loaded, fallacious and highly dangerous concept … (that) … offers a supposedly fair means of condemning some children to second class educations’ (Gillborn and Youdell, 2001). Since pupils who are black, or who receive free school meals, for example, are considerably more likely, and possibly more unfairly to be judged as lacking in ‘ability’, they are also arguably more likely to be consigned to lower sets and lower tiers. If this postulation bears true, then it may well be expected that under a market setting, the educational divide can only widen with what some sociologists consider to be a misplaced belief in ‘ability’. These sociologists (and many others) are almost definitely against educational markets on an ideological basis, and are therefore not working within the ‘markets for schools’ SRP. However, their (and similar) findings with respect to equity would still certainly come to the attention of economists who are working within the SRP as ‘anomalies’, and who would then try to defend the SRP, perhaps by discrediting the sociologists’ methodology or questioning their partiality.

The negative heuristics

Negative heuristic 1: Blame failures of the market on perversions of the market.

For example, Tooley (1995) has asserted that the full potential of markets in education cannot be accurately assessed from studies of quasi-markets, or what he called ‘so-called market(s)’. The state educational marketplace, as it exists in England, is not a free market, since there is no price mechanism in operation, there is heavy regulation from the state in form of the National Curriculum and league tables, etc., and while the demand side has been liberated through per capita funding, the supply side, i.e. the provision of education, has hardly been freed from state control. As such, he dismissed criticisms of markets in education by writers such as Ranson (1993), whom he feels mistakenly uses arguments against the Conservatives’ ‘so-called market’ against markets in general. Just because the schools quasi-market as it is today is inequitable does not mean that education markets in general are inequitable.

Negative heuristic 2: Strike at the method of the work of opponents.

According to Blaug (1980), Becker poured ‘scorn on the “considerable ad hocery” that is required in the conventional approach …’, including the forms of utility functions. This attack on the methods of his opponents may be viewed as an attempt to simultaneously promote and defend his own research programme.

Similarly, in the school choice debate, where empirical work has been involved to ascertain the impact of the ERA, especially with respect to equity, there has been much disagreement over method. There are many different methods that can and have been used to measure educational equity in different contexts: the Gini coefficient, the inter-quartile range, the desegregation index, etc. Gorard and Fitz (2000) used the desegregation index and data on the percentage of pupils with free school meals in every school in England in Wales, to show that socio-economic desegregation between English and Welsh schools in term of student intake has actually decreased significantly since 1989. This may arguably be taken as key evidence that equity has improved.

However, opponents of school markets have produced work with contrary conclusions. For example, multilevel modelling (Goldstein and Noden, 2002) and the isolation index (Noden, 2000) have been used to show that schools have become more socially segregated.

Gorard’s (2003) response to these findings has been to question the approach and method. Gorard (2003) maintained that Noden committed a ‘terrible mistake’ in using invalid arithmetic, asserted that ‘whatever it is that (Goldstein) & (Noden) believed they were doing they were not, in fact, measuring segregation at all.’ and concluded that ‘the approach proposed by Goldstein and Noden (2003) is retrograde, erroneous, and of no clear practical value.’


2.3 Monopoly power as refutation and ‘anomaly’

Economists and sociologists both have recognised that the English schooling market is not a free market, and that schools are not the perfectly competitive agents of neo-classical theory. The acknowledgement may be implicit, including when their work revolves around different levels of competition, such as Bradley et al.’s (2001) and Levacic’s (2004) examination of the impact of competition on efficiency. Stating that different school markets have different levels of competition or that schools have different degrees of market power is akin to stating the fullness or emptiness of one’s glass. School market A has a moderate level of competition: the glass is half full. School Alpha in market A has a moderate level of monopoly power: the glass is half empty.

There is not just the continued intervention by the state, but also the non-fulfilment of the necessary assumptions and conditions for a free market, such as perfect information, freedom of entry and exit, etc., as would be outlined in any basic university economics textbook. More appropriate as an economic model for the schools market, especially at a local level, could therefore be one of imperfect competition, oligopoly or even monopoly, depending on the local context: the number of schools, the types and quality of schools, the selection procedure of the schools, etc. Consider village X in rural Shropshire, with only one school in miles; consider town Y in Kent with one selective Roman Catholic girls’ college, and one mixed comprehensive; consider London Borough Z, which has seven foundation schools, some of which have different specialisms. The local school market in X, Y and Z may rightfully be considered to be a monopoly, an oligopoly and a monopolistic competition respectively. To a degree, the selective Roman Catholic girls’ college may also be viewed as a monopoly to parents who strongly desire their children to go to their nearest single-sex school, since this is the only school in the vicinity with such characteristics.

It follows that these hypothetical schools have market power, and that with the ERA, instead of empowering parents with real choice, economic power and control might simply have passed from the LEA to schools. Furthermore, as suggested by Adnett and Davies (2002), schools may actively strategise to seek to increase their market or monopoly power by further differentiating themselves from their rivals, for example by acquiring beacon school or specialist school status. An irony is that the introduction of all the myriad new types of schools under both the Tories and New Labour is meant to increase competition and choice, but because they differentiate the schools, the market power of particular schools may actually have gone up and competition may paradoxically be stymied.

Market power in the hands of schools may be undesirable, crucially because schools may abuse their monopoly power to do well under the pressures of the market system, and also because the management of schools that are de facto local monopolies could well be as inefficient and inflexible as the bureaucratic LEAs that they were meant to replace. This abuse is manifested, I argue, in schools’ admissions polices, and in schools participating in cream-skimming as a key strategy to maximise school performance. When schools have monopoly power, school choice, I argue, will ironically indeed be school choice, in that schools instead of parents will be in a position to choose which pupils they want. Such schools will be oversubscribed and indeed, a clear indicator that a school might have a degree of monopoly power would be when there is chronic over-subscription year-on-year. Then, admissions policies, whether set up by the LEA or the school’s own governing body, can and have been used to select in and select out pupils.

Efficiency arguments for the market in schools tend to rely on the presence of competition as an impetus for continual improvement, innovation and dynamic efficiency. The incidence of schools with monopoly power would mean that there could be insufficient competition in particular school markets. Then, if headteachers and educational professionals are motivated solely by self-interest, the lack of competition would result in X-inefficiency (Levacic, 2004).

Therefore, the case for school marketisation could be weakened on the grounds of both equity and efficiency. In the framework of the Lakatosian MSRP, monopoly power would be an ‘anomaly’, a problem that school market enthusiasts would have to solve.


2.4: Introducing countervailing power into the protective belt to negate monopoly power

The existence and pervasiveness of monopoly power held by schools may arguably pose a serious challenge to proponents of school marketisation. Monopoly power would emerge for reasons discussed in the earlier section, as a phenomenon that occurs naturally in markets, and therefore, writers like Tooley would hardly be able to blame state intervention for its emergence[1].

Pointing fingers in blame is hardly productive, and if monopoly power in schools as an anomaly and refutation is to be dealt with in a satisfactory manner, it may best to look again at economic theory, to rediscover rationalisations that may perhaps argue either the good of monopolies or that monopolies may be restrained by a system of checks and balances. Certainly, Baumol et al’s (1982) contestable market theory is one that springs to mind, whereby monopolies are driven to behave like perfectly competitive firms, not because there is competition, but because there is a real threat of competition. Unfortunately, schools may have high setup costs[2], and the necessary freedom of exit and entry is perhaps not easily attained for schooling institutions in England. Therefore, contestable market theory may have little to offer as a solution here.

On the other hand, countervailing power, as a concept developed by Galbraith (1956), led him to remain optimistic despite the growth of huge corporations, American Capitalism and corporate monopolies. There is good reason to believe that instances of countervailing power may also be in the English school market, both when parents are choosing schools for their children and when their children are already students in their schools, and therefore provide concerned policy-makers with a tool with which to ensure social justice in its myriad conceptions. In the framework of the Lakatosian MSRP, countervailing power will lie in the protective belt, as a proposition to negate the anomaly of monopoly.

[1] Please see Tooley’s (1995) claims in the negative heuristics section, page 20.
[2] Of course, set up costs for schools vary. It may be prohibitively expensive for schools to be constructed or even expanded in densely built up urban areas, while it may be very cheap to do so in rural areas of particular developing countries, where space, cheap labour and raw materials are very much in abundance.