Theoretical Issues: Markets, Hierarchies, and Quasi-Markets
What is different about social services, especially education and health, that justifies the intense involvement of the public sector in their provision? Why are those sectors not treated like many others including public utilities in which state intervention is either absent or limited to some light forms of regulation? One reason is equity: the consensus in most societies is that everyone should have access to a minimum threshold of security, healthcare, and education, regardless of their income level. In this context, the state can play a unique role in taxing and redistributing resources to those who lack the funds to finance those services on their own. Another reason is market failure: because individuals do not appropriate all the benefits of consuming these services, if left alone they will underconsume them. A typical case is immunization against contagious diseases. In the extreme, some of these services are public goods—that is, it is costly to prevent any individual to consume them once they have been produced— and will not be produced privately—for instance, the security provided by policing a certain city or neighborhood. The state may not only subsidize the consumption of services with positive externalities, but also enforce universal participation in consumption or finance of these services. A third reason is the lack of insurance mechanisms against income losses accruing from serious illness and old age. Problems of adverse selection—in demand and/or supply—usually complicate the existence of such schemes, and the end result may be a large part of the population remaining unassisted. State intervention can not only enhance the scale of such schemes, in some circumstances lowering costs and allowing for greater risk-diversification, but also force universal participation and cross-subsidization.
In developing countries, all these problems tend to be more acute. Low and ill-distributed income leads to low consumption of education, preventive care, and insurance instruments. Inefficient or absent institutions raise transaction costs and asymmetries of information, and weaken contract enforcement, making it difficult for markets for social services to expand beyond elite groups. Mistrust of private insurers discourages participation in voluntary insurance schemes, be they in health or pensions. In such circumstances, ‘consumers’ either lack the resources or are unwilling to spend as much as they should to directly reward service providers, and governments intervene to assure a proper consumption level.
These arguments are not, though, unique to social services. Indeed, most are also relevant in public utilities (electricity, telecom, water, sanitation, and transportation infrastructure) and the financial sector, and have been used to justify a similar degree of public sector participation in those sectors. Equity concerns justify policies geared to universalize the consumption of public utilities and access to the payments system, usually through some form of direct or cross-subsidization. There are also important positive externalities in the consumption of sanitation services and in connecting to networks, as in telecommunications, roads, and the payments system. In many institutional settings, savers’ mistrust of private banks discourages financial intermediation and causes the state to establish its own banks. The construction of roads may demand state intervention to force all drivers to pay tolls. Finally, increasing returns to scale, asset specificity, asymmetry of information, high dependence on third-party contract enforcement and collective action problems, among other factors, may prevent the existence of financial markets and public utilities based exclusively on private service providers. Again, these problems are particularly acute in developing countries.
Historically, public intervention in both social services and public utilities occurred through hierarchical structures, in which a public provider subordinated to the sector minister was responsible for supplying these services. Hierarchies offer many advantages: they facilitate the transfer of subsidies to low-income consumers, they allow for the introduction of social goals to correct for the presence of market failures, and they overcome the problems stemming from weak institutions, in particular the risk of administrative expropriation of sunk private investment.
Consider, as in Fig. 2.1, the alternative of public intervention through regulation of a private service provider. In hierarchy, the policymaker (e.g., a minister) can simply command the public provider to operate according to social objectives, directly transferring public subsidies and having the treasury implicitly guaranteeing insurance schemes. The alternative is to rely on subsidies and regulations that encourage the private provider, while trying to maximize its profit, to behave according to the social goals. Whereas in hierarchy the policymaker could rely on good information on costs and demand to fix the volume of public subsidies, when contracting a private provider, information would be incomplete, easier to manipulate, making it difficult for the policymaker to determine a fair value for the subsidies. The private provider would be the residual claimant of cost savings, and thus would have an incentive to cut on quality, a less likely problem with a public provider. Finally, in hierarchies, operations can be easily adapted to reflect changes in technology or market conditions, whereas in a market type relation this would require renegotiating the contract.
Thus, the traditional approach in social welfare policy has been the direct provision of services by the public sector. The drawback of this alternative is that the public sector, and hierarchical administration in general, is usually less efficient than the private sector in delivering services and less responsive to changes in technology, consumer needs and preferences, and other external conditions. For one, most public providers start with the perverse incentives of monopolists and face little pressure from service consumers, whose power to demand better service is weakened for they do not pay directly for it (and thus cannot threaten to reduce the provider’s revenues by exit) and lack information and mechanisms to adequately voice their dissatisfaction (Sect. 5 returns to these long and short routes of accountability). For another, administrators in typical bureaucracies lack the necessary degrees of freedom to operate efficiently, notably regarding labor management and the acquisition of consumables which often result in high levels of absenteeism and corruption. Other important problems are irregular and inflexible budget allocations, conflicting or unclear goals established by public sector principals (e.g., local politicians and the ministers of health and finance), and the difficulty of securing resources for new investment (and often also for maintenance).
Differently from what took place in public utilities and the financial sector, marked by outright privatization of state-owned enterprises and concessions to private investors, in social services the preferred options have been either to strengthen incentives—through decentralization or NPM,
Fig. 2.1 Make or buy?
for instance—or to resort to more complex governance schemes that transfer some of the operational activities to the private sector, but with close regulation and oversight by the public sector. Complete privatization is usually a good option when service provision can be entirely financed by consumers and the government can withdraw to regulate markets from a distance to correct market failures, but is much more complicated when equity and other social objectives are higher priorities. Several factors contributed to this differentiation between social services (especially education and healthcare) and public utilities:
In social services, the disjuncture between those who control resources and pay for the services, those who provide the services, and those who consume them complicates, at a minimum, the governance structure supporting these transactions. When governments have to pay for the services, public procurement problems change but do not necessarily diminish. There is still a need to select or accredit providers, define fair prices for a multitude of services (as in health) and situations (e.g., in large metropolis and small communities), and monitor providers. Moreover, depending on the means of finance adopted, the pressure applied by consumers may be weak, especially if they do not have a choice of provider, which may often be the case in small communities and when transportation costs are relatively high (Greener and Powell 2009).10 These relations are less complicated in public utilities where consumers pay providers directly, but regulating prices can be a more complicated and contested process.
- • For social services, the reliance on resources from the government budget introduces a number of uncertainties. There may be delays in public sector disbursements, difficulties in negotiating adaptations to non-contracted contingencies, and changes in priorities and policies that lead to a reduction in the scale of operation or simply cause a discontinuity in the transfer of resources altogether. Although private providers can count on a contract to demand payment, contract enforcement and actual collection of payment may take a long time and cost much when the public sector is the defendant. This may encourage private providers to shy away from certain types of contracts and use inefficient technologies that nonetheless minimize asset specificity and hence dependence on government. This uncertainty will also encourage private providers to mobilize for, and invest in, politics (see Sect.5). Resources for public utilities come more from consumers than government budgets, but the political incentives are similar because most prices are regulated.
- • Although complex, it is relatively easier to assess efficiency and quality in the case of public utilities than in social services where pronounced asymmetry of information makes contracts more difficult to write and monitor. Moreover, social services more closely resemble the case of multitask principal-agent relationships examined by Holmstrom and Milgrom (1991) in which strong incentives to save on costs and the difficulty to contract on service quality may result in compromising on quality (Delbono and Lanzi 2012). To prevent service interruption and declines in quality, the public sector may decide to forego much of the savings it could obtain as a result of the efficiency gains accomplished by the private provider.
- • There is usually greater political and ideological opposition to bringing the private sector to operate public health and education facilities.11 The efficiency gains derived by bringing the private sector in largely stem from its profit-orientation, which is anathema to many stakeholders in the health and education sectors. In contrast, the fact that public utilities were already provided by commercial companies (though state-owned) greatly facilitated their privatization. Moreover, different from public utilities, which are capital-intensive, social services are labor-intensive, counting with more influential labor unions, which can more easily mobilize against reforms that reduce job security and penalize absenteeism.
• To address these problems and complexities, and others, governments introducing quasi-markets often pursue a range of complementary and compensatory measures, the focus of the next sections.