Public versus private ownership: Quantity contracts and the allocation of investment tasks
Introduction
This paper offers a new perspective on the boundaries between public and private firms. Suppose the government wants a certain good or service to be provided. According to the property rights approach to the theory of the firm, ownership structures (i.e., the allocation of control rights over the non-human assets that are required to provide the good) matter when only incomplete contracts can be written. However, in the light of the high complexity involved in the provision of public goods or services, focusing on property rights only might be too narrow. Instead, a richer set of contractual arrangements encompassing various task assignments and hybrid governance structures seems to be relevant. Indeed, in the public debate, privatization frequently refers to providing private parties with governing authority or managerial responsibilities, but it is not necessarily associated with transferring asset ownership.1 For instance, in the context of museum privatization, Schuster (1998, p. 128) argues that “it is highly unlikely that privatization will entail a choice between pure forms of public institutions and private institutions; rather, it will entail a complex set of arrangements, each one slightly different from the next, with responsibility for certain elements of the cultural institution's operations being vested in private hands but with responsibility for others vested in public hands.” Similarly, contracts on the privatization of infrastructure projects do not only specify the transfer of ownership rights, but also assign the responsibilities regarding design, construction, maintenance, and modernization (see e.g. Gilroy et al., 2007, Samuel, 2005).
In order to study how the assignment of investment tasks and the decision between private and public ownership interact, we extend Hart et al.'s (1997) seminal work on incomplete contracts and privatization. Society, represented by the government G, writes a contract with a manager M who is supposed to provide a good or service. After the contract which specifies the characteristics of a basic version of the good has been written, two kinds of observable but unverifiable investments can be made. First, it is possible to invest in the development of a quality innovation which (if implemented) increases the government's benefit from the good. Yet, the quality innovation also has the side effect of increasing the costs of provision. Second, it is possible to invest in a cost innovation which (if implemented) reduces the manager's costs, but also lowers the quality of the provided good. Hence, the implementation of an innovation leads to an (ex ante non-contractible) modification of the good actually provided. Both kinds of innovations are efficient. This means their desired effects outweigh their negative side effects; i.e., both innovations increase the total surplus (defined as the sum of the two parties' payoffs). However, the implementation of innovations requires access to the essential assets. Under public ownership, the government controls the assets, so that it has the right to approve or veto the implementation of innovations, while under private ownership the manager can decide whether or not innovations are implemented.
It has been pointed out by Hart (2003, p. C71) that the model of public versus private ownership by Hart et al. (1997) differs from the standard private-sector property rights model in two ways. First, the focus is on investments affecting the cost-versus-quality trade-off that has often been emphasized in the debate on privatization.2 Second, a long-term contract between government and manager is of crucial relevance, while in the standard property rights model contracts are assumed to be sufficiently incomplete to be useless. In our model, in accordance with the complex arrangements governing the provision of public goods and services in practice, we develop these two specific aspects further and argue that their implications may have been seriously underestimated.
In particular, we extend Hart et al.'s (1997) analysis in three important ways. First, we follow their assumption that “G and M are able to write a long term contract specifying some aspects of the good or service to be provided” (p. 1132). Yet, while they assume that the quantity specified in the contract is always the ex post efficient one, we show that ex ante the parties may actually prefer to agree on a smaller quantity, which will later be renegotiated to its first-best level.3 Indeed, in practice it is often observed that the government systematically ends up paying a larger price due to an increase in the number of items being procured, a larger project size, or additional features required by the government ex post. For example, referring to the transportation sector in a number of countries, Flyvbjerg (2004, p. 39) states that “an important cause for cost increases is that the scope or ambition level for a given project changes significantly during its development and implementation.”4
Second, we depart from the standard property rights approach by assuming that the investment tasks are “contractible control actions” in the sense of Aghion et al., 2002, Aghion et al., 2004; i.e., while the investment levels are non-contractible, it is possible to contractually specify who is in charge of which task and hence bears the investment costs. This assumption is in the spirit of the privatization debate where, as has been pointed out above, it is a central issue which responsibilities remain in the public hand and which ones are delegated to the private sector. Schuster (1998, p. 130) argues that “what is most important is the recognition that the management and operation of the museum can be separated into its constituent parts and those parts can be vested separately in a wide variety of responsible authorities.” For instance, in the context of museum privatization, it is plausible that in order to achieve a cost-efficient provision, the responsibility to maintain and operate the museum may be assigned to a private party, while due to quality concerns the care for the designated collections and the restoration of old paintings may remain in public responsibility.5
Third, in addition to public and private ownership, we consider two kinds of public–private partnerships where the parties are in more symmetric positions. In particular, either both parties can veto the implementation of innovations (i.e., there is joint ownership), or they have no veto power; i.e., the government has the right to implement quality innovations and the manager has the right to implement cost innovations.6 Indeed, according to the European Investment Bank (2004), a public–private partnership is a generic term for relationships formed between the private sector and public bodies which includes a wide variety of working arrangements from formal joint venture companies to loose and informal partnerships.7
As we combine four different ownership structures with four possible task assignments, there are altogether 16 governance structures which we analyze. It will turn out that only five governance structures can actually be optimal.
Under private ownership, the manager should always be in charge of the investment in cost innovation, while it depends on the parties' relative bargaining powers who will be responsible for the quality investment. Similarly, under public ownership the government should be in charge of the investment in quality innovation, while the responsibility for the cost investment depends on the bargaining powers. Finally, if the parties agree on a partnership, then no one will have veto power and they assign the cost investment to the manager and the quality investment to the government.
Different allocations of ownership rights imply markedly different patterns of investments. In particular, under private ownership the investment in cost reduction is at its first-best level, while there is underinvestment in the development of quality improvements. Hart et al. (1997) argue that private ownership leads to overinvestment in cost reduction, because the manager disregards the side effect of the cost innovation (as she has the right to implement the innovation without having to compensate the government for the reduced quality). Yet, the overinvestment result crucially relies on their implicit assumption that the ex ante specified quantity of the basic good is the ex post efficient one. In general, the manager's investment in cost reduction is an increasing function of the quantity specified in the initial contract. If the ex ante specified quantity is small, then the investment yields only a small return in the absence of renegotiation. When the parties renegotiate, the manager will get only a fraction of the additional total return (cost savings net of benefit reduction) generated by her investment. Hence, if the parties initially specify a sufficiently small quantity, there is underinvestment, while a suitably chosen intermediate quantity leads to the first-best investment level. In contrast, the quality investment is independent of the ex ante specified quantity, because under private ownership the manager would not implement the quality innovation in the absence of renegotiation. This means that the quality investment is carried out only because successful renegotiation of the implementation decision is anticipated (and as a consequence, the incentives to invest in quality depend on the parties' bargaining powers in the renegotiations).
Similarly, under public ownership first-best investments in quality improvement are attainable by specifying the appropriate quantity of the basic good in the initial contract, while there is always underinvestment in cost reduction.
In a partnership where no party has veto power, both kinds of innovations are implemented even in the absence of renegotiations, because the manager chooses to implement the cost innovation and the government implements the quality innovation. Hence, renegotiation will occur only with regard to the quantity. The manager's incentives to invest in cost reduction are the same as under private ownership, while the government's incentives to invest in quality are the same as under public ownership. This means that now both investments depend on the ex ante specified quantity. However, in general the quantity that induces first-best cost investments will be different from the one that induces first-best quality investments. In other words, there are two goals that the parties must try to attain with only one instrument. It will turn out that the parties then agree on a quantity which induces overinvestment with regard to one task and underinvestment with regard to the other task.
Which governance structure is the optimal one depends on the importance of the cost innovation, its adverse effect on quality, the importance of the quality innovation, its cost-increasing side effect, and the parties' bargaining powers. Roughly speaking, if one party has a sufficiently large bargaining power, then the extent of the underinvestment in one task (which is the only disadvantage of ownership by a single party) becomes arbitrarily small, so that a partnership cannot be optimal. In contrast, if the parties' bargaining powers are almost equal, then the underinvestment problem under single ownership will turn out to be most severe, which makes ownership by a single party less attractive. Whether private or public ownership will be preferred depends on the relative strengths of the side effects caused by the innovations. If the side effect of the cost (quality) innovation becomes very large, so that inducing cost (quality) investment is relatively unimportant, then public (private) ownership is optimal (because then the first-best investment in the important innovation is induced). In contrast, if the side effects of both innovations almost disappear, then a partnership with no veto power must be optimal, because overinvestment (which is the partnership's drawback compared to single ownership) is no longer a problem.
Our paper brings together different strands of literature. First, there is by now a huge literature on the pros and cons of privatization. For surveys, see e.g. Vickers & Yarrow, 1988, Shleifer, 1998, Martimort, 2006. Second, our paper contributes to the growing literature on the property rights approach to the theory of the firm based on incomplete contracting as pioneered by Grossman & Hart, 1986, Hart & Moore, 1990; see Hart (1995) for a comprehensive survey.8 While several authors have investigated the issue of privatization from an incomplete contracts perspective, many of these papers rely on informational asymmetries (e.g., Shapiro & Willig, 1990, Laffont & Tirole, 1991, Schmidt, 1996a, Schmidt, 1996b). In contrast, following Hart et al. (1997), we assume throughout that there is symmetric information between the government and the manager. Third, to the best of our knowledge, the present paper is the first one that analyzes an incomplete contracting model of privatization when the investments are “contractible control actions” in the sense of Aghion et al., 2002, Aghion et al., 2004. In standard models of the property rights approach, only ownership rights (i.e., the control over ex post decisions) can be allocated ex ante, while in addition we consider how the investment tasks should be assigned to the parties.9
The remainder of the paper is organized as follows. In the following section, the model is introduced. In Section 3, we analyze the investment incentives for all assignments of the investment tasks and for all ownership structures, taking into account that quantity contracts are feasible. In Section 4, we characterize circumstances under which different governance structures are optimal. Concluding remarks follow in Section 5. Most proofs have been relegated to Appendix A.
Section snippets
The model
Consider the following model of the public or private provision of a good (or service). At date 0, the government (G) and a manager (M) write a contract that specifies a quantity q ∈ [0, 1] of a basic good (whose characteristics can be described ex ante) and a payment P0 from the government to the manager. When the manager provides the basic good, she incurs costs qC0, while the government's benefit is given by qB0, where B0 > C0. The parties also agree on an ownership structure and on an
Only the manager has investment tasks
Suppose first that it is exogenously given that the manager is responsible for both investment tasks, A = M. Then the date-1 investment levels under ownership structure o are characterized by
Consider first the case of privatization (o = M), where xM = 1 and yM = 0. In this case, the manager's date-2 payoff reads
Hence, at date 1 the manager chooses eMM and iMM to maximize UMM(e, i) − e − i, so that the investment
Optimal governance structures
In this section we characterize the specific circumstances favouring each of the possibly optimal governance structures. Under sole ownership by one party, the optimal investment task assignment depends on who has the larger bargaining power, as is stated in the following proposition.
Proposition 5
(i) Consider o = M. If α > 1/2, then both investments should be assigned to the manager (A = M). If α < 1/2, then the manager should be responsible for the cost investment and the government for the quality investment (A =
Concluding remarks
The remarkable shift away from public to private firms in the last two decades has evoked a vital and controversial discussion concerning the issue of privatization. At the same time, the property rights approach to the theory of the firm based on incomplete contracting has been developed and is now widely accepted as a standard model to analyze the effects of ownership rights. When privatization is discussed from the perspective of the property rights theory, the central focus is on private
Acknowledgments
We would like to thank the editor, two anonymous referees, Odilon Camara, Maija Halonen-Akatwijuka, Andrea Leuermann, Wanda Mimra, Andreas Roider, Anja Schöttner, Dirk Sliwka, Achim Wambach, seminar participants in Heidelberg, Cologne, and Munich, and participants at the IIOC 2009 in Boston, the EARIE 2009 conference in Ljubljana, the EALE 2009 conference in Rome, and the GEABA 2009 conference in Vallendar for helpful comments. Moreover, we are grateful to David Kusterer and Florian Gössl for
References (47)
- et al.
On partial contracting
European Economic Review
(2002) - et al.
Building and managing facilities for public services
Journal of Public Economics
(2006) - et al.
Quality and outside capacity in the provision of health services
Journal of Public Economics
(2002) - et al.
To build or not to build: normative and positive theories of public–private partnerships
International Journal of Industrial Organization
(2008) Incomplete contracts and privatization
European Economic Review
(1996)- et al.
The management of innovation
Quarterly Journal of Economics
(1994) - et al.
Transferable control
Journal of the European Economic Association
(2004) - et al.
Delegation of contracting in the private provision of public services
Review of Industrial Organization
(2006) - et al.
What should governments buy from the private sector — assets or services?
- et al.
Government versus private ownership of public goods
Quarterly Journal of Economics
(2001)
Sources of Weapon System Cost Growth
Cooperative investments and the value of contracting
American Economic Review
Public–private partnerships: task interdependence and contractibility
An Evaluation — Major Highway Program
Business 2010: The Public Sector
The EIB's Role in Public–Private Partnerships
Procedures for dealing with optimism bias in transport planning
Control rights in public–private partnerships
Building new roads through public–private partnerships: frequently asked questions
The costs and benefits of ownership: a theory of vertical and lateral integration
Journal of Political Economy
Firms, Contracts and Financial Structure
Incomplete contracts and public ownership: remarks, and an application to public–private partnerships
Economic Journal
Cited by (54)
-
Private equity market development, state ownership, and firm innovation: Evidence from China
2024, International Review of Economics and Finance -
The proper scope of government reconsidered: Asymmetric information and incentive contracts
2023, European Economic Review -
Open price contracts, locked-in buyers, and opportunism
2022, International Journal of Industrial OrganizationCitation Excerpt :Assigning residual decision-making rights to one party or the other can improve the outcome (Aghion and Holden, 2011; Hart, 2017), but the assignment of rights also meaningfully affects the distribution of surplus. A large empirical literature has examined various aspects of incomplete contracts, such as contract design, efficiency, competitive effects, and others, in a variety of real world settings (Hoppe and Schmitz, 2010; Antrás and Staiger, 2012, and Bajari et al., 2014 and others). In spite of the interest in incomplete contracts generally, one very prominent type of incomplete contract has received less attention in the literature.
-
Optimal ownership of public goods under asymmetric information
2021, Journal of Public Economics -
State ownership and efficiency characteristics
2017, Russian Journal of EconomicsCitation Excerpt :Due to limited competition and incomplete contractual relations, as seen from theoretical models of contracts with limited opportunity for prompt renegotiation, the optimal strategy for the management of state-owned firms would more often (as compared with managers in private companies) be related to making uncreative, routine, and/or “approved” decisions. Managers in such firms rely relatively less on innovations to reduce production costs (Estrin et al., 2009; Hoppe and Schmitz, 2010). The monopolistic power and administrative leverage possessed by state-owned companies may strengthen this trend significantly.
-
Pandemic Crisis, Power and the Role of the State
2024, International Journal of Public Administration