The economic theory of markets asserts that that, under certain assumptions, resources will be most efficiently used and allocated when individuals are allowed to choose what to consume from a range of competitive providers. While it is unnecessary to reproduce detailed arguments about market efficiency here, I will follow Burchardt (1999) and briefly outline some of the benefits that markets are claimed to bring:* Competition – producers compete against each other for business, so the most expensive or poorest quality lose out to competitors* Contestibility – if outside agents see that there are possibilities for profit (through increased efficiency or improved quality) in a market, they can establish a new firm and enter the market. The threat of this happening exerts pressure on existing firms to keep quality high and costs lows* Decentralised decision-makers – consumers choose the mix of goods that they would prefer, rather than having that set of goods dictated to them by some social planner* Cost-Benefit analysis – rational agents acting in a market compare the costs and benefits of each transaction, and only act when the benefits outweigh the costs.

This might be unlike the public sector, where costs can be hidden by government subsidy or free provision.However, as stated before, markets are only held to be perfectly efficient when certain assumptions hold – namely that all agents in each market are perfectly rational, that they all have all of the information they need to make decisions, that there are no externalities, and that there are perfect and complete markets, i.e. there are markets for every type of good which are all filled with numerous competing firms.If we are to challenge the ability of the market to act effectively to deliver social policies, then we must find circumstances where these assumptions are unreasonable.

First, and most obviously, the term ‘social policy’ infers that, rather than allowing each individual to decide their own level of welfare consumption, the government formulates policy to encourage, or even force, individuals to consume welfare. Consider compulsory education – the government favours it as a social policy because of efficiency reasons (it encourages national economic growth), because of equity reasons (it eliminates some instances of low skills that result in low incomes) and also because of paternalistic reasons (the government does not believe that under-16s can make a rational decision about whether or not to consume education).It is clear that leaving the provision of welfare services solely to the market, without any regulation making consumption of such services compulsory, necessarily entails the abandonment of all paternalist motives involved in social policy. Giving families money which they could spend on education rather than collectively providing a free school for them to use means that, they could use this money in whatever way they like – possibly in ways that may be considered non-socially desirable, e.

g. on drugs. This also brings us think of externality issues. If a family spends its money on drugs, then we might suggest that this would cause negative social externalities as there might be needles in the street, or an increased crime rate. On the other hand, if a family spent the money on education then there might be positive externalities to society, as the entire country benefits from having an economically and culturally vibrant youth. It is therefore clear that most social policies also seek to enforce consumption behaviour not just because of paternalism, but also because of the externalities resulting from this behaviour.It is clear therefore, that allowing the market to entirely determine the level and mix of welfare consumption is an option considered by no serious political movement. However, surely there are other ways have the market deliver of social policies while maintaining some control over these factors? Consider the case of car insurance – the government regulates to say that everyone must have car insurance, but it neither finances nor provides this service.

Compulsory consumption through regulation provides a way in which government can exercise paternalism and control of externalities without getting otherwise involved – we could suggest that one person’s consumption of car insurance has positive externalities for others (as it makes sure they get paid compensation without delay if they are involved in an accident) and also that without compulsion, myopic behaviour might occur (individuals might underestimate the risk of being involved in an accident, and fail to save/insure for that possibility). The system seems to work quite well for car insurance, so can social policies not be effectively provided through this system, of private finance and provision, but government regulation/compulsion?The first possible objection to this system is one of income – what happens if the poor are so poor that they cannot afford to meet the burden of compulsory education for their children placed upon them? This can be dismissed quickly by an assumption that the government would provide minimum income provision (through cash-benefits such as child-support or some sort of education voucher) for the poor.More viable objections to the provision of welfare services through market mechanisms can be made with regard to efficiency and equity considerations. Many believe that the use of private markets to provide education fails because the characteristics of education mean there are many imperfections in the assumptions needed for market efficiency. Barr (1998) argues that in most situations, there can never be anything approaching perfect competition for schools, because the transaction costs of changing school are high and because transport to school is a major issue. As a result, he suggests that there would need to be price controls imposed by the government on any education market in order to limit monopoly power.

In addition, he points to failures in capital markets – because students (or their parents) cannot always make contingent loans based on future earnings, he suggests that there would be a tendency to under-consume education (if education is compulsory, that is to consume an inefficiently low quality of education).To rectify this (and under-consumption problems that also arise from the positive social externalities of education), then there would need to be state intervention to regulate the quality of education that is provided. In addition to these problems, Barr also raises the question of imperfect information – that consumers do not have perfect information about the nature of the product, i.e. whether the education provided is of good quality, and whether the type of education provided is suitable for the individual child – which any cost-benefit decision is a difficult one to make.As a result, he suggests that there are limits to consumer rationality in education markets, and that at the very least, there will be a need for state intervention in order to regulate educational information (by creating standardised test results, teacher qualifications, guides to best schools etc). Equity problems may also arise under a situation of significant imperfect information – it has been demonstrated that information asymmetries disproportionately affect the poor, and that they tend to be more myopic in their decisions than the well-off. If the state were concerned about such issues, then they would have to provide extra resources to the poor in order to compensate for the informational disadvantage they suffer.

Given the problems with pure-market provision of education, we have suggested a system in which there are price controls, quality controls and information regulation in order to ensure the efficiency and equity of private market arrangements. On consideration, one may suggest that large parts of this system of intervention are not dissimilar to the current U.K. school system. All parents must ensure their children receive education (stopping quantity under-consumption), they can in theory choose which school they attend (between a range of public and private schools – depending on location constraints), all schools are inspected for quality (stopping quality under-consumption), and test results are standardised with and comparable in published league tables (ensuring quality information).

The only difference is that the majority of schools, instead of being privately funded and provided, are instead provided by the public sector. Why is this so?Barr (1998) asserts that public provision simply motivated by equity considerations is not defensible – equity issues could be addressed by cash redistribution without state intervention. However, given the existence of market inefficiencies that require state intervention anyway, he argues that the best way to address equity considerations while causing minimal additional market distortions may be through a system of state provision. Essentially, he recognises that because of the market and information imperfections that characterise many welfare goods, a first-best outcome is often not possible. In this case, the second-best outcome often involves state intervention in the form of funding and provision.While I have just looked in depth at education, it is clear that there are also significant problems allowing the market to deliver other welfare goods. Consider unemployment insurance – there is almost no private provision of unemployment insurance because of the massive principal-agent and adverse selection issues that exist. Therefore, if a social policy decision is made that unemployment insurance should be available, the onus is on the state to provide a collective social insurance scheme to achieve this aim.

Less obviously, many have argued that the private market cannot efficiently provide healthcare, pointing out that in the U.S. the average life expectancy is 77.2 years (World Bank – 2001 figures), similar to the average of 77.4 years in the UK, despite the fact that the U.S spends $4499 per capita on healthcare compared to the $1747 spent by the U.K – apparent proof that for health-care, fraught with principal-agents problems, private markets are poor at containing costs and providing efficient levels of healthcare provision compared to publicly-provided systems.It is for these stark economic facts that, despite the period of welfare state ‘retrenchment’ that was identified during the last twenty years of the 20th century, instead of looking to move responsibility for welfare services directly to the private market, ‘new right’ governments instead looked to increase the efficiency of the public sector by introducing market-like mechanisms, so called ‘quasi-markets’.

Put simply, the state was not ‘rolled back’ to the extent that it lost its role as the ultimate arbiter of the mix and extent of welfare provision (the principles of the UK welfare state – free, universal access – were upheld), but it focused more on an ‘enabling’ role, making social policy decisions and then funding the resulting welfare programmes, whilst looking to the market to be the principal providers. In this way, the New Right hoped to increase the efficiency of welfare-delivery mechanisms whilst maintaining control of social policy direction. As Johnson (1999, 88) states, ‘The state is in a position to take an overall view of the direction of social policy and to set priorities for its future development.

This necessarily involves a degree of planning which allows the individual priorities of voluntary and commercial providers to be coordinated or regulated to serve public policies or purposes’. The increasing application of market principles to public provision, and the use of contracting out responsibility for service delivery has led some to suggest that we have moved from an institutional to a ‘contractual’ welfare state.Though we have so far concentrated exclusively on efficiency and equity considerations surrounding the final product of social policies, it is interesting to take a brief detour to think about other factors relevant to social policy provision that may be affected by decisions made between state/market delivery. Specifically, the means through which social policies are delivered may have differing effects on conceptions of citizenship and the accessibility of welfare services. Devolving welfare service provision to markets may decrease the conception of equal citizenship that some postulate comes with the shared experiences of universal, identical, state-welfare. Similarly, while recently, state-run services have made conspicuous efforts to broaden the diversity of their staff and their attractiveness to minority groups, companies providing specifically-defined contracted services are likely to have much less concern for these matters unless they comprise part of the contract. Johnson (1999, 89) notes that ‘Markets and voluntary organisations exhibit a middle-class and racial bias’. This may have adverse effects on the experience of some users of welfare-services beyond the realm measurable by performance indicators.

Moves towards increasing use of market mechanisms were driven by a firmly-held ideological belief that market systems were more efficient and responsive than institutionalised bureaucratic systems that suffered from principal-agent problems (staff were more interested in preserving their own jobs / responsibilities than delivering value-for-money, efficient services). While the economic arguments against pure market provision of key welfare services, as laid out above, were recognised, successive governments have tried to simulate market conditions within the public sector in order to make efficiency gains.One of the most fundamental changes caused by the move towards quasi-markets was the introduction of compulsory competitive tendering (CCT) for local government services. The thinking behind this scheme was that even if welfare services as a whole could not be provided by the market, then splitting off many of the component inputs of each service (for example school cleaning is a necessary part of school provision) and subjecting them to market forces would in turn increase the efficiency of the entire service.

The Thatcher government therefore made it compulsory that local authorities, rather than just relying on the status-quo of in-house services, had to invite bids from entities to provide these services.The existing in-house cleaning departments were separated from the rest of the council structure and formed into separate bodies, so called direct service organisations (DSO). It was only infrequently that such DSOs did not get contracted to do the job they had previously been doing inside the council. However, the relationship governing the service provision had fundamentally changed. Whereas before the education department would expect the cleaning department to perform the relevant services, now the education department commissioned the DSO (which used to be the cleaning department) to work on a specified contract at a specified price. This move, and the threat of the entry of an outside, private bidder in the next round of contract allocation, was supposed to force the DSO to be responsive to economic incentives and thereby improve performance.

While in theory the private provision of cleaning services should not be controversial (there is no compelling reason to suggest that cleaning has externalities, that agents have different levels of information about the service, or that it is impossible for competitive markets for cleaning to exist). However, problems that did arise, as Vincent-Jones and Harries point out (in Bartlett, 1998), were often as the result of the loss of trust caused by the change in institutional arrangements, and the significant accompanying rise in transaction costs.Examining CCT schemes in Eastmet, London, they compare ‘hard’ quasi-markets, where previously joined departments are strongly separated and experience a reduction in cooperation and trust, which leads to efficiency losses and adversity over the terms of contract, with ‘soft’ quasi-markets that are more successful in fostering the continuing the cooperation and teamwork between client (education department) and agent (cleaning services DSO) and that therefore accordingly function more efficiently. This analysis can be read in terms of Niskanen’s X-efficiency model, which posited that efficiency is best achieved as a result of a well motivated workforce (Niskanen, cited in McMaster in Bartlett, 1998).

In a ‘hard quasi-market’, efficiency could in fact decrease, not just because of the lack of staff-motivation causing a fall in X-efficiency, but also because rigorous and adversarial monitoring and enforcement of a contract requires significant information costs on both sides.Markets can provide welfare services, but significant state intervention is often required, both to ensure efficiency and to allow considerations based on paternalism and equity to be made. Given the nature of many welfare goods (e.g.

unemployment insurance), it is difficult to envisage markets ever efficiently providing the service provided by the state. It is not surprising therefore, that recent attempts at welfare state reform have not sought to move state responsibility to the market, but rather have attempted to mimic market-style incentives to improve the efficiency of welfare provision.While empirical studies have shown that such measures are capable of increasing efficiency, it seems that in some cases the policy-planners have, in their commitment to market-ideology, forgotten the strengths of public provision – namely the existence of long-standing institutions with strong institutional memory and degrees of cooperation and trust between workers. Quasi-markets are most likely to be able to make a meaningful contribution towards the successful provision of welfare services when the advantages of market pressures are integrated into, rather than imposed instead of, existing public-sector frameworks.