حسابرسی و مناقصه رقابتی در بخش دولتی

This paper investigates the impact of a ministry's budget size on the choice between auditing a (Niskanen)bureau and employing competitive bidding in the provision of a publicly funded good. The ministry's marginal expected payoff increases fastest with budget size for auctioning as opposed to auditing. However, the ministry is shown to switch from a purely public provision to a competitive provision as the budget size increases even if expected fixed costs favor the audit. The study contributes to the literature by extending the Niskanen framework and by endogenizing the institutional arrangements for the provision of public sector goods.

Just as the private sector, in recent years governments have looked towards outsourcing and privatization for the provision of goods and services by the public sector (De Fraja, 1993 and Feenstra and Hanson, 1996). An increasing number of prison facilities in the USA are operated by private contractors and reportedly have 10% lower costs than their public sector counterparts (Hart et al. (1997)). Significant cost advantages for privatization over public provision have been found for refuse collection services in cities with 50,000 or more residents, while no significant difference was found for cities with fewer residents (Stevens (1978), Savas (1977)). For policing, it has been found that cities with a population of 80,000 or more had more costly and less efficient policing services than smaller cities, even in the presence of monitoring or auditing activities by citizens (Davis and Hayes (1993)). The key economic problem facing a ministry responsible for the provision of a public good lies in the asymmetric information about costs between the producer of the good or service and the ministry. This is true both if the producer is a profit maximizing private firm and if the producer is a government bureau. As a result, the ministry has to pay informational rents to each type of provider. If the provider is a profit maximizing firm, informational rents are minimized through the use of an auction mechanism, which introduces competition at the bidding stage (Osborne and Gaebler (1993)). McAfee and McMillan (1987) suggest that such competitive provision tends to involve first-price sealed bid auctions. Public sector bureaus have been characterized as Niskanen budget maximizing institutions (Niskanen (1975), Mueller (1997)). Bendor et al. (1985) and Breton and Wintrobe (1975) suggest that monitoring and management compensation packages may lessen the informational rents of such Niskanen bureaus. Monitoring has been defined by Mookherje and Png (1992) as the commitment of resources towards the detection of an offense prior to receiving information about an occurrence. They term the commitment of resources after an offense has been reported as investigation. Auditing, as analysed by Baron and Besanko (1984) in a principal-agent model with asymmetric information about a firm's realized costs, involves the expenditure of resources in order to detect misreporting after receiving a signal about costs. This paper investigates the impact of budget size on the choice between auditing and competitive bidding in the provision of a publicly funded good. Auditing and competitive bidding raise several questions. First, what is the impact of the budget size on the funding source's choice between competitive bidding and auditing? Second, what cost distributions will result in the ministry having a non-trivial choice between competitive bidding and auditing the bureau for a given budgetary allocation? Third, what are the implications of the number of firms capable of providing the public sector good on the ministry's choice of competitive bidding or auditing? In order to address these questions, the equilibrium outcomes of two mechanisms are compared: one is an auditing mechanism offered by the ministry, as principal, to a bureau, the agent. As in Baron and Besanko (1984) and Lacker and Weinberg (1989), the first order approach to the principal-agent problem is used. In this approach, the global incentive compatibility constraint for the agent is replaced by a local incentive compatibility constraint, allowing the use of optimal control techniques. The second mechanism under investigation is an auction mechanism. Due to its greater analytical simplicity, a second-price sealed bid auction is used to characterize the outcomes. The revenue equivalence theorem assures us that this auction will lead to the same (expected) utility for the ministry as the first-price sealed bid auction observed in practice.1 It is well known that an agent could be induced to truthfully report private information in the presence of infinite penalties for misreporting. The cultural norm that the “punishment fit the crime” however implies that there is a finite maximal penalty which the courts will uphold. This level is imposed exogenously. Auditing, thus, must occur with positive probability since even if the principal detects misreporting the penalty is bounded. Hence the auditing mechanism, just like the auction, is not capable of extracting all informational rents. Auctions, of course, are well known to be efficient mechanisms, i.e., to minimize the informational rents of the agent. Since in practice both auditing and auctioning are costly to implement, it is nevertheless not obvious which mechanism is better under what circumstances. The ministry thus faces a non-trivial choice between providing the publicly funded good either via competitive bidding or an audited bureau. The main findings are as follows: most generally, the optimal way to contract for the provision of the publicly funded good depends on the fixed costs of the mechanism, the number of potential bidders, and the distribution of costs. More specifically, we confirm that the informational rents of the provider increase faster with budget size under auditing than in the auction mechanism. This informational rent advantage of the auction mechanism can offset a fixed cost disadvantage relative to the audit mechanism. Hence, as the budget size increases, switching from auditing to the auction mechanism may occur at sufficiently high budget sizes even if auditing enjoys a fixed cost advantage over the auction mechanism. In the absence of this fixed cost advantage, the auction mechanism strictly dominates the audit mechanism for all budget sizes. An increase in the number of firms participating in the auction results in a decline in the budget level that induces switching from auditing to the auction, since the ministry's expected payoff from the auction is increasing in the number of bidders. We can also show that for small enough budget levels the ministry will choose to produce the good without auditing or conducting an auction, having some able provider do so at the highest possible cost, thereby avoiding any fixed costs associated with either mechanism. These results are consistent with the norm for many organizations that do not conduct audits or auctions for sufficiently small procurement contracts or activities, as well as the examples at the beginning of the introduction. The resulting important role of the allocation of fixed costs in government is further discussed in the conclusion. The paper is organized into five sections. In Section 2, the formal model is developed. Section 2 also characterizes the equilibrium outcomes for the auditing and competitive bidding mechanisms. In Section 3, the ministry's equilibrium choice is determined. Section 4 provides conclusions. All mathematical derivations and proofs are in a series of Appendices.

The paper analyses a ministry's choice in the provision of a public sector good in the presence of asymmetric information. The ministry copes with asymmetric information about provision costs by either auditing a Niskanen budget maximizing bureau, or by contracting with private firms. In the case of auditing, the penalty which can be imposed is exogenously limited, limiting the ministry's ability to extract informational rents from the bureau. In the contracting case, a second-price sealed bid auction is used in order to minimize informational rents. The paper extends the Niskanen model of government bureaucracy, and endogenizes the choice between auditing and competitive bidding in the public sector. We show that at very low budget levels it may be optimal not to invest any resources towards the reduction of informational rents, but to provide the good without any supervision. This result is consistent with the common practices of many public institutions that permit the issuance of procurement contracts to a single firm without tender when these contracts are sufficiently small. It is noteworthy that the implementation of government regulations to require the use of auctions independent of the budget level may therefore not improve welfare. For higher budget levels the optimal choice involves an attempt to limit informational rents. The approach that is best will depend on the costs associated with each and the budget size. An auction is the optimal choice independent of the size of the budget if the fixed costs for competitive bidding are low relative to those expected for auditing. If, however, the fixed costs for competitive bidding are high relative to those for auditing, the optimal choice can depend on the budget level. In that case the institutional arrangement can involve switching from public sector to private sector production as budget size increases. This is driven by the fact that auctions are the optimal way to limit informational rents, and so will ultimately outperform any other mechanism (e.g., auditing and unsupervised production) even if it has an initial cost advantage. Both auctions and audited provision may involve a reservation level of cost, above which the ministry will not fund the provision of the good. The optimal level of this reservation cost depends on the distribution of costs, as is usual for auctions and shown here for auditing. Indeed, it is the same for both procurement choices. As is common practice for public sector procurement, it is found that the auction mechanism requires more than three firms to participate. The size of government is reportedly positively correlated to overall economic activity as measured in terms of gross domestic product per capita (Alesina and Wacziarg, 1998). Holsey and Borcherding (1997) also suggest that government budget rises faster than income in the short run while its long-run income elasticity is unity (i.e., a restating of Wayne's Law). While Wayne's Law is not supported by all empirical studies, the size of government is generally found to be positively correlated to per capita income and the income of the median voter.14 It follows that the results predict that there will be growing reliance on private provision of public sector goods during periods of economic boom and expanding government budgets even when fixed costs favor auditing over auctions. Also, developed countries tend to have larger government budgets in absolute and as a percentage of GDP than developing nations. Again, if fixed costs, including the availability of technical ability, favor auditing over auctions, then it can be anticipated that the private sector will be more appropriately involved in the provision of public sector goods in developed countries as compared to developing nations. In general, the implementation of institutional arrangements in government without regard for the fixed costs of maintaining these structures or the size of the government budgetary resources may not be welfare improving. This study predicts that the government copes with information asymmetries by aiming to minimize informational rents. The optimal institutional arrangement of government is found to vary depending on the costs of each arrangement. However, in a government environment with multiple products, the allocation of fixed costs between programs is endogenous. Since it also is a significant determinant of the choice between public or private sector provision, this study suggests that the administrative rules associated with the allocation of these costs are important in predicting and determining the institutional arrangements of government. Bureaucrats can influence these fixed costs through the allocation methodology, and may therefore be instrumental in determining the fixed cost of conducting public sector auctions or public sector supervision. Hence, the institutional rules surrounding the allocation of these costs are an important determinant of the institutional arrangements established for the provision of public sector goods.