2000年-世界发展银行全球_Decentralizing_the_Provision_of_Health_Services___An_Incomplete_Contracts_Approach_32页_1mb

Summary of "Decentralizing the Provision of Health Services: An Incomplete Contracts Approach" by William Jack

Core Content

This paper explores the optimal allocation of authority between central and local governments in the provision of health services using an incomplete contracts approach. The key insight is that in the absence of a complete and binding contract, the ownership of health facilities and the distribution of financing authority play a critical role in shaping incentives and overall service quality.

The model considers three types of inputs required for the delivery of health services:

• K: Physical capital (e.g., clinics or hospitals)
• e_G: Central government efforts (e.g., drug procurement and distribution)
• e_A: Local authority efforts (e.g., physician monitoring and incentives)

These inputs are combined to produce health services, and the model assumes that cooperation leads to higher service quality and lower costs compared to non-cooperation.

Main Viewpoints

1. Ownership of Facilities

• The ownership of the physical asset (K) is identified with the responsibility for service delivery.
• Ownership should be assigned to the party that values the well-being of local residents the most, as this maximizes the joint surplus of the two governments.
• If ex post bargaining fails, the party with more value for residents still benefits from the other's efforts, which enhances the incentive to cooperate.

2. Financing Authority and Spending Responsibility

• Financing authority and service delivery responsibility should be negatively correlated.
• It is generally optimal to assign tax authority to the party that values residents' well-being less, thereby separating spending responsibility from financing authority.
• This separation ensures that the financing burden is not too heavy, which can reduce the incentive to provide effort.

3. Costs of Transferring Assets

• If the cost of transferring ownership is high (due to human capital embodiment), it may be optimal for the party with higher construction costs to have planning authority.
• Higher asset transfer costs increase the likelihood of separating ownership and provision, as the costs of reallocation are too significant to justify integration.

4. Incentive Distortions

• Incomplete contracts lead to incentive distortions, as the returns to effort are affected by the bargaining power of each party, which depends on their outside options (i.e., the value of the asset they own).
• Ex post Nash bargaining is assumed to result in a 50:50 split of the gains from cooperation.

5. Government Objectives

• Both central and local governments care about consumer well-being and effort costs.
• The weight assigned to consumer well-being by each government is denoted by β_G and β_A, respectively, with β_G + β_A = 1.
• A higher β_G indicates the central government values residents more, while a higher β_A indicates the local authority is more responsive to local needs.

Key Information

• Model Structure: The model is based on a three-input framework (K, e_G, e_A), with cooperation assumed to yield higher quality and lower costs than non-cooperation.
• Consumer Welfare: Quality is used as a proxy for consumer benefits, and tax bases are assumed to have different distortionary costs (λ_G and λ_A).
• Total Surplus: The total surplus is the sum of the weighted consumer welfare and effort costs, and the optimal allocation is determined by maximizing this surplus.
• Ownership Impact: Ownership of the physical asset increases the return to effort for the owner, thus shaping the effort levels of both parties.
• Empirical Context: The paper is part of a broader effort to understand incentives and institutional design in the provision of social services, with a focus on healthcare.

Conclusions

• Optimal Ownership: The central government should own the physical asset if it places a higher weight on resident well-being (β_G > β_A), and vice versa.
• Effort Levels: Under optimal ownership, effort levels are sub-optimal compared to the first-best scenario, but incentives are better aligned.
• Robustness: The ownership assignment rule is robust to technological changes, as it is primarily driven by preference parameters rather than the specifics of production functions.
• Policy Implication: Decentralization should be designed to separate ownership and financing, and to align incentives with the value placed on resident well-being by each party.

Structure of the Paper

1. Introduction: Discusses the challenge of allocating authority in health service provision.
2. Technology and Preferences: Outlines the production functions and government objectives.
3. Optimal Asset Ownership: Analyzes the trade-offs between central and local ownership.
4. Allocation of Tax Authority and Financing: Examines how tax and financing powers should be distributed.
5. Conclusion: Summarizes the findings and policy implications.

This paper provides a theoretical framework for understanding how decentralization can be structured to improve the efficiency and equity of health service delivery in developing countries.