The effect of treating public services as commodities

IMPACT

The case for competition and marketization of public services, though widely accepted in government, has been made through the application of formal economic reasoning rather than practical experience. Efficient market production relies on a process of defining services in terms which allow for competition, choice and the substitutability of tradable products. The evidence for this theoretical position is mixed at best. This article provides policy-makers, those commissioning services and practitioners with support in arguing for public services to be judged by different criteria.

ABSTRACT

Within the frame of orthodox economics, only market allocations can be efficient, and markets can achieve any desired outcome. Public services, however, operate by criteria which are not satisfied by market allocations, including the requirements of policy (such as targeting, universality and equity), cost-effectiveness, and conformity with the requirements of democratic government (such as accountability and prior authorization). The efficient delivery of commercialized services depends on commoditization—standardizing commodities so that they can be traded on equivalent terms. That process changes the nature and character of what is provided, and compromises the effectiveness and quality of public services.

KEYWORDS:

• Commoditization
• effectiveness
• efficiency
• marketization
• public services

Economic policy in the UK has been captured by an economic theory that asserts the intrinsic superiority of market provision over public service delivery. The first part of this article explains what that theory is, and why it is not applicable to public services. The second part examines what happens when governments attempt to put the theory into practice. The process of commoditization—converting activities into a form which can be subject to marketization and competition—changes the nature of what is being provided, and that is liable to undermine attempts to achieve appropriate provision through the action of markets.

While arguments for ‘free markets’ have a strongly ideological and political character, they are rooted in economic theory. Economic theory mainly works by constructing models of the consequences if people were to behave according to certain criteria, such as the pursuit of rational self-interest, or the maximization of utility. More recent developments in the discipline, such as game theory and ‘rational choice’ models, have reinforced that approach, particularly in the discipline’s emphasis on methodological individualism. This kind of theory does not in most circumstances generate ‘hypotheses’, as Milton Friedman claimed it did (Friedman, 1953), because processes such as maximization lie at the limits of conceivable behaviour (and sometimes beyond them). It offers, rather, idealized norms by which economic behaviour can be judged.

The Green Book, the authoritative guidance for assessing policy and projects published by HM Treasury, explains (albeit with qualifications) that their guidance ‘is based upon the ideas of welfare economics and concerns the optimisation of social welfare’ (HM Treasury, 2022, p. 4.22). ‘Welfare economics’ is a theoretical field which, as the Treasury document attests, has had substantial influence on public policy. Despite the name, it has very little directly to do with welfare, or at least welfare as most specialists in social policy would understand it. The arguments for the intrinsic superiority of markets have been given their fullest formal expression in the ‘Fundamental Theorems of Welfare Economics’. The First Fundamental Theorem claims to demonstrate that the operation of a free, competitive market maximizes the ‘welfare’, utility or satisfaction of producers and consumers. This has been presented (questionably) as a confirmation of Adam Smith’s concept of the ‘invisible hand’ that guides an economy (Blaug, 2007). The Second Fundamental Theorem is supposed to prove that any desired state or outcome can be achieved through such a market. This, Starr writes:

 … is the basis of the common prescription in public finance that any attainable distribution of welfare can be achieved using a market mechanism and lump-sum taxes (corresponding to the redistribution of endowment). On this basis, public authority intervention in the market through direct provision of services (housing, education, medical care, child care etc.) is an unnecessary escape from market allocation mechanisms with their efficiency properties (Starr, 2011, pp. 213–214).

The Fundamental Theorems are not sound. They depend on a series of implausible claims, based on muddled terminology (for example ‘utility’ or ‘choice’), inconsistent assumptions (rational self-interest versus aggregate behaviour, equilibrium versus creative destruction) and outlandish criteria—for example, extreme individualism and Pareto optimality (Spicker, 2013a). Regardless of the flaws, however, their influence has been considerable. Arguments for competition, free exchange and choice have become a standard part of the lexicon of government. The Public Sector Efficiency Group in the Cabinet Office treats markets and competition as one of the ‘key broad drivers of public sector efficiency improvement’ (Aldridge et al., 2016). The Competition and Markets Authority explains that the introduction of competition and markets is part of its mission: ‘Where elements of market competition are present, they can help drive service improvement and value for money in markets for public services’ (CMA, 2014, p. 2.8). It seems, then, to be accepted in government that the allocation and distribution of public services can be achieved through the introduction of market mechanisms, and the way to make public services do things better is to make them more like commercial markets.

The case for markets

The arguments for markets have been advocated vigorously by the think tanks of the ‘New Right’—bodies such as the Institute for Economic Affairs or the Adam Smith Institute (see for example King, 1987; Glennerster & Midgley, 1991). Eamonn Butler writes:

Markets are amazing. They unite the populations of the world in peaceful trade, co-ordinating the efforts of millions of diverse individuals. They enable us all to swap things we don’t want for things we do. They steer resources to where they are most valued. They discourage waste and encourage fresh ideas. And they do all this without any governments or authorities needing to tell them how. It’s amazing, but it’s true (Butler, 2009, p. 17).

The case is overblown, but it is a difficult argument to reject wholeheartedly: it reflects what we do in the provision of basic commodities, such as food and clothing. When there are problems, that is taken to be an argument for offering cash benefits, rather than providing a service. There is a formal theory of the ‘second best’, which shows that once the conditions for a free market fail, that might not be true (Lipsey & Lancaster, 1956), but it tends to be broadly accepted in economics and politics that this is not conclusive: the starting point for analysis is an ideal model of a perfect market. The Treasury Green Book defends this approach: ‘the main value … lies in providing an abstract thinking tool used by economists to trial economic propositions under a range of market imperfections ‘(HM Treasury, 2022, p. 4.20).

Much of the economic literature, and much of the guidance, focuses on ‘market failures’—a class of issues where the ideal operation of the market is frustrated, for example because of monopoly supply, or because the market is unable to distribute costs and benefits appropriately, as happens with externalities such as pollution. These failures are widespread, but they do not go to the root of the problem. One of the most basic political arguments for markets extols the merits of choice, both as a mechanism for maximizing personal utility and as a process which makes it possible to reconcile production with demand. But there is a simple corollary of this point. Producers have choices, too. They decide what they are going to produce, and how much they will do. Markets leave gaps.

Health­­care, for example, is often taken as an example of imperfect competition, reflecting imperfect information and constrained choice. That underestimates the problems. In the provision of health insurance, we find that some problems are too difficult to respond to, some conditions are too expensive to treat, some consumers are liable to be excluded. Private providers make their selection, in principle, on the basis of the return that provision to that section of the market can offer. ‘Cream skimming’ in private markets is commonplace: insurers sell their policies to people who are healthier and more profitable (Hunter, 2008, p. 111). The process of selection can lead either to elevated prices for special treatments, or to effective exclusion from the market. These are not market failures. They are what happens when markets work as they are meant to do. So if we want, as a matter of policy, to ensure that some conditions are treated at all, it makes sense to look to alternative, non-market arrangements.

Public services

Public services operate by different criteria to commercial markets. They are said to be ‘services’, not in the economic sense, but to the extent that they depend on a relationship between the public agencies and service users. The classic ‘social services’ are health care, social housing, social work, education and social security; they depend on a relationship that is in some sense personal, bringing people into contact with the provider. Not all publicly-provided services are personal in the same way—the police serve the general public, not necessarily the people they most make contact with—but the fact that we refer to them as ‘services’ says something substantive about them; we think of them as establishing a continuing link between a provider and the service user.

The ‘public’ element in public services is not necessarily there because they are owned or delivered by public authorities; a wide range of such services is provided by other bodies, including voluntary, charitable, religious, mutualist and non-profit agencies. Bozeman & Bretschneider attribute ‘publicness’ to political authority (Bozeman & Bretschneider, 1994), but that is not quite right either: charities and non-profits commonly have their own public purposes. Services are said to be ‘public’ because their objectives—objectives such as social protection, citizenship, or the common good—are the objectives of public policy (Spicker, 2009). They operate by publicly-defined criteria which are rather different from those that are supposed to be satisfied by market allocations: examples are tests of comprehensiveness, inclusiveness, targeting and equity. The values which are extolled by neo-liberals—values such as individualization, choice and adaptability—are worth considering, but they are not necessarily to be preferred to those public values.

Whether the principles and methods of the market are preferable in specific instances has to be decided according to the circumstances. Individualizing rubbish collections would lead to differentiated pricing according to location, for some people to opt not to pay, and for some dumping of rubbish that might otherwise be expensive to clear—in the UK, that happens now with bulky rubbish. Individualized, market-based water supplies may lead to reduced demand, but they also lead to people having their water supply cut off and potential problems for public health. It is perfectly possible to imagine an education system where access to schooling would depend on parents paying—that was the case for English secondary education before the 1944 Act—but it would not meet the objectives of universal education or of child protection. The Treasury has come, over time, to accept that there are ‘values that economic markets are either unable to fully capture, or are unable to register at all’: they include ‘environmental, cultural, health, social care, justice and security effects’ (HM Treasury, 2022, p. 4.22; 2.3). The idea that public services might operate on different criteria from private provision is also recognized in European law, where there are exemptions from market rules for ‘social services of general interest’, or SSGIs. The characteristic features of SSGIs are that they are solidaristic (or redistributive), not for profit, protective, and have an ‘assymetric’ relationship between producer and consumer, requiring finance from a third party. SSGIs include social security, which in much of Europe also means health insurance, as well as a range of services intended to ‘facilitate social inclusion and safeguard fundamental rights’, such as advice, rehabilitation, employability, personal social services and social housing (European Commission, 2007; 2011). Education is explicitly not covered, but the same rationale seems to apply there.

Beyond the objectives, there are also major differences in the processes by which public services work. Public services generally have to conform with the requirements of democratic governance, such as public accountability and prior authorization. Sheaff et al. (2023) draws a distinction between ‘commodified’ and ‘collaborative’ arrangements with voluntary, community-based and social enterprises; that might also be seen as a distinction between organizations that work to the objectives of commercial providers, and those that see their role as the operation of public policy. Boyne summarises the differences in terms of the organizational environment, goals, organizational structures, and the values that apply—the ‘public sector ethos’ (Boyne, 2002). Hood points to three core sets of values in the public services. There are ‘sigma-type values’, emphasising frugality and the reduction of waste; ‘theta-type values’, emphasising rectitude, fairness and legitimacy; and ‘lambda-type values’, emphasising resilience, robustness and security (Hood, 1991). It is easy to see the ‘public sector ethos’ as the application of different priorities or values from commercial exchange, but the difference is much more profound than that. The values reflect the particular character and purpose of the public services—theta-type values emphasising accountability and legitimacy, the others being concerned with the conditions necessary for optimum delivery of policy objectives. Those values can conflict with each other: saving money can work against robustness, accountability can lead to delay or close off courses of action, and resilience is not enhanced by strict conformity with procedural rules.

Efficiency

Much is made, in the theoretical material I have referred to, of the ‘efficiency properties’ of private markets. ‘Efficiency’ means different things in different contexts—the Gershon report discussed five alternatives (Gershon, 2004). In relation to public services, it often refers generically to the reduction of waste, and to the sigma-type values referred to in the previous section. In economic theory, however, there are two rather different concepts. Allocative efficiency is a situation in which no party can be made better off without making someone else worse off. This standard, sometimes called Pareto efficiency, is widely accepted in the economic literature, but it is deeply flawed. As a moral position, it sets aside all distributive considerations, a stance that Ryan calls ‘intolerable’ and ‘repulsive’ (Ryan, 1989, p. 47). As a tool for economic analysis, it is hopelessly inadequate, because it disregards the economic implications of inequality. The effect of disparities of purchasing power is that people with more can exclude poor people from access to resources, either directly (as happens in land ownership) or indirectly, because when some people have greater command over resources, they bid up prices (see for example Filmer et al., 2018). Even if free markets do maximize allocative efficiency (as argued by Kaplow & Shavell, 2001), it does not follow that they should be welcomed.

Productive efficiency is found where the unit cost of production is as low as possible. The connection between this idea and allocative efficiency depends on some complex assumptions and constructs (Spicker, 2013a, pp. 102–106) but, for the present purposes, it is the emphasis on productive efficiency which shapes policy most directly. A competitive market is supposed to improve efficiency in three ways. First, it pushes firms to deliver goods and services at the lowest possible unit cost. Second, it encourages entry to, and exit from, the market, so that only firms that are able to deliver this low unit price can succeed. Third, it encourages firms to find new ways to achieve great efficiency, encouraging innovation.

Figure 1 shows a conventional production function. The point where the unit cost of production is at its lowest—the point where marginal cost is equal to average cost—is said to be the point of productive efficiency. Competition drives producers towards that point—either increasing their production, or limiting it, so as to minimize the costs of producing a unit and so to optimise their profit.

Figure 1. The model production function.

There is an unwarranted assumption underlying these claims: that the kind of efficiency that is being sought in the public sector is the same sort of thing as efficiency in the private sector. Efficiency, in the sense of low unit costs, is not a primary value for public services. When public services try to avoid waste, they are not aiming for the lowest possible unit cost of production, but for cost-effectiveness—meeting the (public) objectives of a service at the lowest possible cost. The difference, shown in Figure 2, is easy to demonstrate. Productive efficiency depends on balancing outputs against productive costs. Cost-effective services aim to minimize costs for a given level of output. A rationally self-interested, private producer will seek to find the optimal balance between quantity and cost, because that is the point at which profits are maximized.

Figure 2. Efficiency and cost-effectiveness.

A public service works to different criteria—typically, in public policy, distribution to a target population or universal coverage. If private firms are delivering goods or service at the lowest possible cost per unit, they are doing it by exercising their choice—delivering what they do best, and—crucially—avoiding activities which offer lower returns. Elements of service or production that are notably expensive, difficult or require firms to extend too far—such as the provision of water supplies, mail to remote areas, domiciliary care for older people, and many health treatments—will be avoided. That kind of selection is liable, fatally, to compromise the objectives of public services. It is often claimed that private provision is more ‘efficient’ than public service, because their unit costs are lower. That may well be true—but it is irrelevant. Public services are supposed to be doing something else.

Commoditization: markets in practice

Sooner or later, a delicate and refined economic theory has to come into contact with a coarse and brutal reality. Before the supposed advantages of the market can be realized, the interactions of actors in economic markets have first to be translated into terms that allow goods and services to be exchanged. Competition in the production of goods and services depends on processes of commodification and commoditization. Commodification is the process of identifying goods and services that can be traded as commodities. The discussion of commodification in public services has often been concerned with the commodification of ‘public goods’. If goods are not divisible and not excludable, there is no practical way of matching any economic demand for the facility to its consumption. A deal of ingenuity has been devoted to finding ways of commodifying public goods, so that they might be divisible and excludable. That might be done by introducing charges for service, restricting access to those who pay, individualizing charges for collective provision such as rubbish collection, and so forth.

Commoditization goes beyond commodification; it is about what happens next. ‘Commoditization’ refers to a process of removing the distinctions between commodities so that they can be the subject of direct economic competition. Conceptually, competition is central to arguments for markets, because competition constrains producers to make efficient choices. In a perfect market, with multiple providers in competition, all producers are price-takers—that is, they have to accept the price for which their goods or services can be sold. are pushed to produce on terms where they can maximize profits at that price, and so at the lowest possible unit cost. Producers who cannot match the costs of their competitors are driven out of business; producers who make excessive profits will find their profits diluted by new entrants to the market.

For competition to be possible, the output of each producer has to offer an alternative to the product offered by its competitors. Producers compete, in short, by selling commodities that are sufficiently alike to be presented as alternatives to each other. This model of competition depends implicitly on the principle that commodities produced by one producer can effectively be supplied by other producers. This is based on the process of commoditization. The word favoured by the OECD to describe the same process is that commodities have to be ‘homogenous’ [sic]: ‘Homogenous products are considered to be homogenous when they are perfect substitutes and buyers perceive no actual or real differences between the products offered by different firms’ (OECD, 2006, ID:3230). The commodities do not have to be identical, or completely standardized, but they do have to be similar enough to be capable of substitution. That tends to favour, in the production of goods, the reproduction of common features in competing products—for example, the position of the controls on a car, the use of common controls and commands in computer operating systems, or the standard sizes of paper. There are whole classes of virtually interchangeable consumer commodities—hatchbacks, smart phones, kettles—distinguished only by marginal differences in style or function but where, nevertheless, consumers will be presented with a wide-ranging set of options. This reflects, of course, on the claim of private sector producers to be ‘innovative’. Innovation, in the sense of doing things differently, is necessarily constrained; if a product is highly differentiated, it will not be substitutable by others. Commoditization implies not true innovation, but a degree of conformity.

The theory of how production comes to be commoditized is heavily geared to the manufacture of goods, not the provision of services. Economic production happens when a good is made; goods like food or clothing can sit on shelves until they are bought. There are financial products which can be conceived of in the same way, and they are called ‘services’ too, but for the most part this is not how services work. The key element in most services, Osborne et al. (2013) suggest, is that production and consumption happen at the same time—so it is generally not possible to offer someone a service without that person being part of the process. Effective services depend on access to the person served (and, therefore, on the participation of that person).

There are commoditized services, in this sense, that are widely traded in the private market, but they tend to be limited to specific classes of service. Some services allow consumers to hop between service providers—insurance, hotels, catering and taxis are examples. Hairdressing is more personal but it might be done on an ad hoc basis. These are services that can be itemized, costed and dealt with distinctly from other actions.

In other cases, however, the principle is not easily extended. Corcoran and Albertson (2023) point to a shift, implied by the process of marketization, from relational to transactional forms of provision. Many services are intrinsically personal, and personal relationships are basic to quality of service. Lack of continuity detracts from that quality. This has become a major problem for the provision of domiciliary services to older people. Allocating services according to the need for specific tasks to be performed—such as brushing teeth or combing hair, both specifically provided for in the Scottish legislation on personal care—suggests a level of fine discrimination which is hard to sustain in practice. Treating domiciliary care as a series of 15-minute packages of time has made it possible for firms behave as if they were renting out hotel rooms, but the approach also means that people can be faced with a bewildering series of care workers coming through their doors. This could be done differently. Isaacs and Neville (1975) make a distinction between long term, short term, and critical (that is, irregularly occurring) need; in principle it should be possible to allocate service in different patterns according to the shape of a person’s need, rather than specific task-based activity.

A different, but related, problem, has emerged in residential care, where the process of providing care has come to work on a standard pattern based, a CRESC report argues, on the model of a budget hotel:

An unintended consequence of the chain business model is a future in which care homes are increasingly alike. By default, society must then accommodate its older people in large, full service hotels of single rooms with en-suite, in a setting which is more institutional than domestic (Burns et al., 2016, p. 10).

The idea of ‘personalization’ in social care is heavily dependent on the commoditized production of services. The Griffiths report (1988) argued for a system where there would be multiple purchasers, each with their own budget, and multiple providers—duplicating, to the greatest extent possible, the structure of a market. Personalization in social care uses a series of methods intended to make services more responsive individually to service users, including personalized assessments, mechanisms for exercising personal choice, and individualized budgeting. The arguments have been applied to elderly people, psychiatric patients, people with learning disabilities, people with addictions, offenders, school pupils and homeless people; the fields of operation include social care, education, health, housing, social security and criminal justice (see Spicker, 2013b). Personalized responses rely on there being several options from which a choice can be made. In theory, what should happen is that the service user is offered a range of outcomes. Shortages of provision, competition for resources and limited options make it difficult for choices to be exercised effectively, and people have to accept second-, third- or fourth-best options. It is easy to blame this on the under-resourcing of services, but choice inevitably involves compromise; people can only choose from a range of options only when there is surplus provision and alternatives exist. The commoditization of personalized services implies a paradox. Personalization is supposed to be responsive to individuals. However, individual responsiveness in a competitive market is not done by tailoring services to people’s needs. It is achieved by putting together homogeneous commodities in different ways, and homogeneous commodities are not personalized. The services that are being delivered will be standardized to some degree, and in order to compete providers have to manage the delivery of commodities by making treatment more uniform.

There are many tensions between the criteria applied by public services and the approach of private firms, and commoditization is not necessarily foremost among them. A degree of uniformity in housing units, the general adoption of ‘wheelie bins’ for domestic rubbish collection or mass vaccination are not beyond controversy, but they are not intrinsically incompatible with public principles. By contrast, there are commoditized services which have been controversial—where the standardized, mechanistic responses delivered by commercial contractors have been found to be inappropriate to the character of the service being delivered. Malin (2018) points to a trend to de-professionalization. The idea of professional intervention—where a trained specialist assesses what is required, makes a judgment and delivers the appropriate service—sits uneasily with the delivery of a standardized commodity. In various fields of activity, including health care, probation and personal care, the shift to commoditization has favoured a focus on tasks to be undertaken rather than relationships or process (Malin, 2018). Some of the other practices which have invited criticism include, for example:

• Probation services which have reduced the average time input and consequently accepted higher risks (BBC, 2018).

• The standardization of the terms of calls for emergency health care: a private contractor undertaking out-of-hours calls for access to health care instructed its staff to cut down the number of emergency referrals being made, so that service could be held within the anticipated norms (Lawrence, 2013).

• Employability services that throw people at the job market in the hope that some of them will stick (House of Commons Work and Pensions Committee, 2013).

• The over-zealous imposition of benefit sanctions, reflecting a structure of incentives where the contractor was rewarded for the numbers of benefits withdrawn (Cowburn, 2016).

It would be easy to dismiss this kind of thing as just another example of cutting costs, or cutting corners, to make bigger profits. But the issues go beyond costs: public services hold back costs too, they just don’t do it in the same way. Nor, despite the withering criticisms that have been made about competence and quality, is it a question of whether the private sector is capable of delivering services. In each case, what is happening is that individual services have been commoditized—treating probation or emergency responses as specific instances of production, conceptually equivalent to portion control in a fast food restaurant. This is what happens when competitive firms attempt to maximize returns on ‘units of service’.

This reflects on the validity of the economic theory that has been used to justify these transformations. The attempt to inject the spirit of private enterprise into the public services has failed to grasp the central point: private businesses do things differently. The processes of commodification and commoditization change the character and nature of the service being provided. This is liable to frustrate any attempt to achieve the objectives of public services by means of the market.

Marketizing public services

‘Marketization’ is the process of developing a market in which producers and consumers can interact in order to produce and allocate commodities. When the market comes into contact with actual practice, however, the effects may be unanticipated—something that is simply not captured by the theory.

Imposing the model of the private market on the public services has taken three main forms: sub-contracting public service activity, marketizing public services, and developing a mixed economy. The first of these has been based in sub-contracting to private firms, in the belief that those firms will optimise the corners of public service activity which they are undertaking to provide. If a public service agency wants a specific, identifiable, time-limited activity or the provision of a set of goods, that can done through a sub-contract and, in many cases, it will be uncontroversial. For example, the provision of desks for a local government office is something that is specific, identifiable and time limited: no-one has argued that local authorities should manufacture their own desks. A private firm can meet publicly-defined needs without having to jump through the same hoops as the public sector, and walk away afterwards. That is not true, however, of many of the private contracts for public service, and certainly not of the sorts of contract that are being developed in the name of marketization, commercialization, public-private partnerships or competition. These are contracts where a private firm undertakes to participate in the provision of a public service, or indeed any slice of work that otherwise would have been undertaken by the public authorities. In contemporary societies, punishment is exclusively the province of government; a ‘private prison’ has to depend on the authority of the state. This is a public activity, where government contracts with an independent provider to fulfil the government’s obligations as its agent. The process of sub-contracting ‘services for child protection’ or ‘work related activity’ is not a private matter between two contracting parties. And the process of contracting out adjudication, which is the subject of another article in this Public Money & Management theme (Machin & Reynolds, 2023), requires the independent provider to perform a public function, as opposed to transferring that function into the private market.

Given the public nature of the activity, the authorities issuing the contract have to account for their actions, and so do the contractors. This is currently an area of dispute: public officials have taken to recusing themselves from Freedom of Information requests on the basis of ‘commercial confidentiality’. The line cannot hold. Wherever public policy is at stake, wherever services are operated in trust, the contracts that are issued are governed by the criteria of the public services. The nature of the activity has to be capable of being specified in a contract in terms which are compatible with the objectives of the public service and the requirements of public accountability. There has to be someone charged with the management of public policy who is capable of commissioning the work. Private contractors are sometimes engaged to help policy-makers develop objectives, but they cannot be engaged to set the objectives themselves—that would not be consistent with democratic governance. And this means that, once a private provider undertakes to provide a service for reasons of public policy, it needs to be to subject to the structures, mechanisms of accountability and public criticism associated with public service.

The second approach consists of attempts to make the public services behave more like the private market. The key elements of internal markets have been the presumption of competition, the use of price and ‘market signals’ to indicate success or failure and the provision of incentives for performance judged in those terms. It is difficult to see what there is about these arrangements that could improve the quality of service delivery. Public services cannot respond to competition by entry or exit from the market. Rewarding success tends to imply the denial of resources to failing agencies, making it rather difficult to remedy the situation; without free entry, there will be a concentration of resources, and without free exit from the market, poor performers will be trapped in a vicious cycle. It seems, rather, to have been a marker for a different kind of system, more in tune with market thinking; but advocacy for an internal market as something of value in itself it seems to represent another example of a failure to distinguish public services from free markets. This has taken the form of a ‘quasi-market’, an attempt to mimic the operation of an ideal market within the structure of public services—the ‘internal market’ of the NHS, abolished only in name, was an example—but a quasi-market does not operate like an economic market. What happened in the NHS internal market was not that the system became markedly more commercial, but that service providers came to lean on the methods of partnership and collaboration that were characteristic of responses to the preceding scenario; the model of GP commissioning groups, arguably the most successful aspect of the internal market, was developed by practitioners with a strong public sector ethos, who were looking for ways to make the system work despite itself.

The third approach—and, to a large extent, what happens by default—has been the creation of a mixed economy—marketizing what can be marketized, inviting private and independent operators to participate, and relying on residual public services to cover the rest. In practice, just about every system is mixed (Béland & Gran, 2008). Wherever we look, there will be some services which are provided on a commercial basis, while others will be provided by the principles of public services—not necessarily by government, but by independent, voluntary or mutual providers. The mix is likely to be complex, reflecting a range of historical experiences, influences and institutions. The question is not so much whether to choose between markets or public provision, but rather to ask what the balance between them is going to be.

Mixed systems, however, present problems for the public services. Wherever there is diversity in provision, public and private systems coexist, but necessarily they adopt different roles. The effect of selection, producer choice or exclusion is that there will be something left over—a residuum—and government, as the provider of last resort, has to make up the difference. Government has to accept a degree of uncertainty and risk; the private sector makes its own decisions about quantities, and the public services have to deal with any outstanding variations and fluctuations in demand. The scope of government activity is determined by the hole that is left after other services have been delivered; that is not necessarily well-defined, or well-chosen, or convenient. Residual provision commonly goes to people who do not have the resources themselves, and have to be paid for in some other way; to people with complex, multiple problems, requiring a range of different kinds of support; to circumstances bearing greatest risks, where the return is not commensurate with that risk. Typically the residuum is complex, dispersed and difficult to deal with. So it is hardly surprising if the public services appear at times to be expensive, with higher unit costs, and fraught with difficulty: it goes with the territory.

Conclusion: public services or markets?

While there is now extensive experience of the introduction of competition and private sector imperatives into public services, the evidence of benefit is inconclusive. Showing that the propositions associated with welfare economics are true in some cases can usually be countered by showing that they are not true in others. It would not be unreasonable to argue that the diversity of approaches introduced by housing associations, or the growth of private pensions, have been beneficial; rail privatization or the UK market for domestic energy might be taken to show the opposite; but any of those examples, when the detail is examined, become complex, ambiguous and uncertain, subject to contradictory argument and motivated reasoning.

There are certainly elements of ideology, on both sides, in judgments about the welfare mix. Some would argue for public provision, for example in health care, as something that is simply good in principle. On the other side, in the course of the coronavirus crisis, the outsourcing firm Serco sought to establish a stake in a field where they had no background or competence, not because there was reason to think they could do it better than public providers, but because they wanted to ‘cement the position of the private sector’ in public provision (Marsh, 2020). Where there are policy choices to be made, any decision is liable to be pragmatic and—because the range of actors and institutions extends beyond the capacity of governments to determine—provisional.

The objection to injecting market principles is not that they are always inappropriate, but that the argument for markets is over-generalized. The problem is not just that markets sometimes fail to deliver in practice: it is that commercial markets do not work to the same principles as public services. The choice between market and public services depends on the objectives of services, the type of activity being undertaken, the experience of performance and the criteria being applied. That might all seem obvious, but the statement that ‘it depends’ stands in opposition to any model that assumes that market provision is an intrinsic good. Markets are never going to work to the same principles as public services. That does not mean that markets never work at all. They may well be a good option in some circumstances; in others they may not be. The purist advocacy of markets does not help us to know which is which.

The theory discussed at the start of this article makes some heroic assumptions about the terms on which providers and service users engage with public services. If the aim of those assumptions was, as the Treasury suggests, to test which ones fit, then the straight answer is that they hardly fit at all. The processes of commodification and commoditization are ways of hammering reality into shape, so that it can fit the ideology. The market approach does not lead to services becoming more like a market. It leads somewhere else, somewhere which is neither clearly public, nor clearly private; it is not even certainly in-between. Treating public services as tradable commodities changes the character and nature of the issues that are being dealt with. The process matters.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Paul Spicker

Paul Spicker is Emeritus Professor of Public Policy, Robert Gordon University, UK. He has written extensively on social policy and administration, poverty, social security and housing policy, and the services of the welfare state.