W%V I :Ixi POLICY RESEARCH WORKING PAPER 2727 Hospital Governance Representation of community and government interests on and Incentive Design hospital boards can balance the competing concerns of The Case of Cdrd Public reducing costs and increasing the quality of service provision Hospitals in Lebanon in corporatized hospitals. Florence Eid The World Bank Operations Evaluation Department Country Evaluation and Regional Relations Division November 2001 POLICY RESEARCH WORKING PAPER 2727 Summary findings There are three potential levels of government activity in To improve the understanding of the role of the health sector: regulation, finance, and direct governance and incentives in corporatized hospitals, Eid provision of services, with the government owning and explores the design of corporate boards of public managing hospitals and primary care clinics. Eid focuses hospitals, the institutional linchpin of such systems. She on service provision. shows how principal-agent theory, particularly the In recent years corporatization has been introduced as multitasking and common agency approaches, can an institutional design for public hospitals-as a means provide a useful analytical lens in understanding hospital of improving efficiency and reducing transfers in a board design in the case of Lebanon. She also shows the publicly owned, decentralized health system. Eid treats implications of corporatization for health policy and decentralization as a reallocation of decision rights to management. lower levels of the public sector. She shows how such a strategy creates new needs for monitoring and control of decentralized units. This paper-a product of the Country Evaluation and Regional Relations Division, Operations Evaluation Departm 't- is part of a larger effort in the department to evaluate the performance of public sector institutions. The study was f ; ided by the Bankis Research Support Budget under the research project "Analyzing Problems in Public Hospital Corporatization Using Information Economics." Copies of this paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Agnes Santos, room H3-306, telephone 202-473-1675, fax 202-522-3124, email address asantos((gjworldbank.org. Policy ResearchWorkingPapers are also posted on theWeb at http://econ.worldbank.org. The author may be contacted at florenceid@aol.com. November 2001. (51 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get tPe findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Policy Research Dissemination Center Hospital governance and incentive design: The case of corporatized public hospitals in Lebanon Florence Eid* * This is one of series of papers prepared under the World Bank Research Project, "Analyzing Problems in Public Hospital Corporatization Using Information Economics". The research project is financed by a grant from the World Bank Research Committee and is directed by Anwar Shah, Operations Evaluation Department, World Bank (ashah@worldbank.org). I am also grateful to Professors Bengt Holmstr6m, Karen Polenske and David Cutler for their excellent guidance. I have benefited from discussions with Susan Athey, William Cleverley, Lucy Goodhart, Pere Tberm, April Harding, Jamie Robinson, Nancy Kane, Nancy Rose, Meredith Rosenthal, Anwar Shah, and Steve Webb. I also wish to thank participants in the MIT Industrial Organizations Seminar for their helpful comments. 2 1 Introduction Corporatization is a hybrid organizational forn, between governrment ownership and privatization that seeks to improve efficiency and reduce transfers (and costs) in a publicly owned agency. Corporatization is a brand of decentralization in that it reallocates decision-making authority from the central administration to lower levels of the public sector. After a brief overview of decentralization and corporatization in public health, this paper analyzes the role of governance and incentives in corporatized hospitals. The analysis focuses on the design of public hospital corporate boards, the institutional lynchpin of such systems. Drawing on Dixit's (1996) multitasking common- agency model as a conceptual lens, I propose a manner of assessing the institutional design of corporatized hospital boards. I analyze the extent to which the Dixit model explains factors salient to such boards, and point to other factors that come into play. I conclude with some policy implications for the reform of the public hospital law in Lebanon. 1 In part 2, I discuss the role of decentralization and corporatization in public health reform to indicate the institutional structure that has recently been promoted in a number of countries. In part 3, I introduce the multitasking common agency model and map it onto the problem of hospital board design. I also describe the initial data collected to carry out the analysis. I introduce the case of Lebanon in part 4 and discuss the objectives of the reform as well as the principal features of the institutional structure governing corporatized hospitals. In parts S and 6, I discuss the coordination and agency I Passed in 1996, analysis of this law (Eid 1998) revealed that its design is weak in some key areas that make it difficult to implement. In a policy note addressed to the Ministry of Health, I recommend its amendment - a project currently underway. See Appendix A. 3 problems emanating from the design of corporatized hospitals in Lebanon and offer some ideas for the reform of the system. 2 Decentralization in public health care provision There are three possible types of government involvement in health: regulation, finance and service provision. Regulatory functions include decisions on the rules of system configuration and the definition of respective roles for the public and private sector. Finance functions determine the extent of universal health coverage using public funds. The government may also be involved in direct provision of services, as the owner and manager of hospitals and primary care services. The limits of private initiative in the delivery of public goods and political constraints on privatization are the two main factors behind public health provision. Two principal schools of thought have developed in answer to the question of how to increase efficiency in public service provision, health included. On the one hand, it is argued that efficiency and performance are more important than ownership, and that good management is key (e.g., Moore 1996; Barzelay 1992). As such, hiring innovative managers with the right technical and leadership skills and introducing the appropriate ("private sector-like") management systems improve efficiency. On the other hand, it is argued that the public sector has inherent inefficiencies due to the nature of the goods it provides and to the limited power of incentives it can offer, and that the size of the public sector is better reduced to a minimum through the transfer of responsibilities to the private sector where possible (Wilson 1989, Kikeri, Nellis & Shirley 1992; Schleifer 1998). From this perspective, privatization is the preferred option for better service delivery. Advocates of corporatization take a middle ground as a point of departure, 4 namely that both the public sector and the market are capable of failure, necessitating the search for organizational forms that reduce inefficiencies on both sides. In designing such organizational forms, incentives and coordination are key levers. Public ownership implies, in practice, various constraints on the management of facilities. Personnel are usually civil servants and procurement procedures are subject to system-wide rigid rules. Therefore, an inevitable effect of public ownership is less flexibility in adapting to local conditions, and 'low powered' incentives. Tirole (1994) considers four reasons why the 'power' of incentive schemes tends to be 'low' in public sector agencies: a) the multiplicity of goals and the difficulty of their measurement; b) the unavailability of benchmarks for comparisons; c) the heterogeneity of owners; and d) property dispersion. Holmstr6m (1994) arrives at similar results with respect to large organizations: in developing systems to manage diverse sets of activities, they tend to damnpen incentives and quell innovation. Coordination issues are also extremely important when agents have low-powered incentives, since discretionality and autonomy can lead to poor performance, such as shirking. Resolving coordination issues relates to institutional design.2 A central element of design is the allocation of residual control rights (or decision rights), between centralization and decentralization.3 An inevitable trade-off exists between centralization 2 Throughout this paper, I use North's (1990) distinction between "institutions" and "organizations". Institutions are the formal and informal rules that shape interaction. They range from constitutions, to laws, to common practice to corporate culture (Kreps 1993). Organizations are groups of individuals bound by some common purpose to achieve a given set of objectives. They include political, economnic, social and educational bodies. In this proposal, a hospital is an organization. The law and decrees governing the operation of the hospital are a set of institutions. 3 'Residual control rights' over an asset are defined by Hart (1995) as "the right to decide all usages of the asset in any way not inconsistent with a prior contract, custom, or law ... possession of residual control rights is taken virtually to be the definition of ownership ... in contrast to the more standard definition of ownership, whereby an owner possesses the residual income from an asset rather than its 5 and lack of efficiency, and decentralization and lack of monitoring. Coordination seeks to minimize this trade-off. Studies of organizational boundaries consider two elements in the decentralizalion of decision rights (Holmstrom 1995; Hart 1995; Milgrom and Roberts 1992; and Kreps 1992). First, those with authority must also bear the responsibility for their decisions because the alignment of authority and responsibility creates incentives for optirnal decision making. Second, coordination is important in ensuring that organizations allocate the authority to make decisions to the agents best informed to make them. 'I'he benefits and costs of decentralization have been well studied. Table 1. Benefits and costs of decentralization Benefits Costs Better use of information at the local level Agency costs Lower response time in adapting to local Coordination costs conditions Increased motivation of managers Costs of communication between central and local units Source: Brickley, J. et al. 1997. Managerial Economics and Organizational Architecture, Irwin. In health service delivery, organizational boundaries are in flux throughout the world, because of changes in medical technology, know-how, and costs, resulting in differential changes in transaction costs (Robinson 1996). Organizational boundaries have also been in flux because policy-makers have deliberately experimented with new residual control rights" (pp.30). Residual control rights are also referred to as 'decision rights' by Holmstr6m (1995), Milgrom and Roberts (1992), and Kreps (1992). The latter, shorter term is used more frequently in this paper. 6 organizational forms to solve the agency and coordination problems outlined above. In the US private health sector, for instance, vertical disintegration and horizontal integration have been the two prominent trends in managed care (Robinson 1999). Numerous industrialized and developing countries are experimenting with the separation of funding from provision functions, with the aim of improving efficiency (Govindaraj & Chawla, 1996). One of the main institutional responses to this effort has been the corporatization of public hospitals. 2.1 Corporatization and its implications in public hospital reform. Corporatization seeks to retain public sector ownership of hospitals, but to reduce their cost by: (a) granting them revenue-raising capacity, and; (b) changing the incentive structure at the local level, including the level of risk incurred by hospitals. By transferring decision rights over finance and management to the level of hospital managers, corporatization also seeks to improve the quality of public health provision. However, unlike what happens in private health provision, corporatization cannot achieve a complete transfer of risk to the provider (hospital). Because financial risk continues to be consolidated at the level of the national public sector, among the difficult issues in the design of corporatization is the decentralization of decisions rights in a way that transfers a sufficient degree of financial risk to the corporatized entity, to improve performance. Under corporatization, public hospitals are generally required to develop a revenue-raising capacity through user fees. However, the incentive to raise funds depends on the role and structure of health insurance coverage. Under universal coverage, hospitals receive a transfer from the public budget. The design of hospital 7 finance options ranges from (a) an allocation estimated based on transfers made in previous years, and (b) a performance contract. In the former case, the hospital is designed as an administrative unit similar to any arn of the central administration. In the second case, when establishing a performance contract, the central administration or sector aims at setting the goals and expected budget and empowers decisions and responsibility at the level of the hospital (Harding & Preker 1999). The impact of performance contracts has been mixed (World Bank 1995; Shirley 1999). Since there is no significant transfer of risk, the real effect on incentives depends on multiple factors that go beyond the definition of the contract. As an example of the range of options under this arrangement, the hospital manager reports to the Minister of Health in some cases, while he/she reports to a board of directors in other cases. Further complicating this sort of arrangement is the difficulty of monitoring hospital directors and board members in the presence of political intervention. As a result, especially when performance contracts are present, the design and effectiveness of hospital governance institutions are key, and depend on the following types of factors, currently being 4 grappled with. * Consistency between the proclaimed objectives of corporatization and organizational design, i.e., where on the gamut between administrative units and performance contracts the system lies; 4 These questions are of special interest given the international trend toward decentralization and corporatization of the public sector, and given the apparent difficulty of designing effective public hospital boards in both industrialized and developing countries (Govindaraj & Chawla, 1996; Barnum & Katzin, 1993; Shonick & Romer, 1983, Savage et. al., 1997; Schleifer & Vishney, 1997; Gertner & Kaplan, 1996). It is curious that despite the wide interest in this topic, there is little theory that informs it. 8 * Whether key stakeholders/principals are represented on hospital governing boards and how much influence they wield; * Requisite alignment of the incentives of the agent, or hospital manager, with those of the principals, and, by extension, alignment of the objective function of the hospital with that of the sector - a coordination problem that impacts both the quality and cost of service provision; * Adequacy of the power of incentives given intended outcomes. In what follows, I discuss the relevance of some agency models to understanding incentives and coordination in the institutional design of corporatized public hospitals.5 I then evaluate the incentives that the system in Lebanon has provided for hospital boards of directors to be responsive to the objectives of their stakeholders, of which there are at least two sets -- the health sector's regulators of public hospitals and community members/hospital users in their areas. 3 Conceptualizing the corporate governance ofpublic hospitals Dixit's formulation of the problem of governance in the public sector builds on two seminal models in the field of organization economics. The first is the multitasking model, developed by Holmstrom and Milgrom (1991). In this model, an agent has several tasks that compete, at least partly, for the agent's attention and effort. Because 5 I use the terms institutions, institutional structure, institutional design, laws and their decrees of application interchangeably in this paper. A comprehensive treatment of institutions would normally cover problems of implementation, enforcement, and monitoring in addition to issues of design (structure) (Polenske 1999). For the sake of narrowing and deepening the scope of this research, I focus on issues of design, which are most amenable to the analysis of decrees - an important component of my data and policy problem. I will bring in issues of implementation, enforcement, and monitoring insofar as they enlighten the problem of design, but they will not be the focus of the discussion. 9 the agent's priorities over tasks are not certain to correspond with the principal's, the latter devises an incentive scheme to influence the agent's allocation of effort. The choice of incentive scheme depends on the degree of observability of inputs and outputs, and on the differences in values between the agent and the principal. Two important results derive from the Holmstrom/Milgrom model: (1) If the output from one task is poorly observable, compared with output from a competing task, then the incentive scheme for the competing task must have lower power (i.e., the reward must be less) to avoid excessive diversion of effort from this task to the more observable one(s); (2) If some tasks are primarily of value to the agent (as compared with the principal), and can be controlled by being prohibited altogether, then it may be preferable for the principal to prohibit them, rather than attempt to provide stronger incentive schemes for the performance of other tasks. Bemheim and Whinston (1986) consider the problem of one agent with more than one principal. The agent may work on the basis of explicit delegation by principals, or by intrinsic assignment (when the agent takes decisions that affect several principals). If principals cooperate, or agree on goals and coordinating incentives, the result is similar to having a single principal. If principals do not agree on goals, then actions by the agent may be biased to those principals providing greater incentives, otherwise the mean behavior by the agent would be to satisfy all principals at the same level. Dixit combines the two models to show that the combination of multiple principals and multiple tasks results, perforce, in low-powered incentive schemes. His model is based on the intuition that in such situations, each principal will try to free ride on the incentives provided by the other(s). The multitasking common agency model 10 predicts that given unobservable effort, an agent will exert second best effort if the principals are united and third best effort if the principals do not act cooperatively. Under non-cooperative arrangements, even though a given principal j may not be concerned with any other components of the agent's output but those of interest to j, principal j would prefer that the agent exert less effort in other dimensions because that would induce the agent to make more effort in the dimension that benefits j. In equilibrium, a situation with multiple principals and multiple tasks yields low-powered incentive schemes because some of the incentive provided by principal j to the agent results in benefits to other principals as well. This "leakage" makes it much less desirable for principal j to offer a powerful incentive scheme. Given unobservable effort, improving on this outcome involves better coordination of principals, an important potential lever in the design of public sector organizations, especially given the difficulty of providing high-powered incentives. 3.1 Mapping the model onto the problem of hospital board design In applying the multitasking common agency model, we consider the hospital manager or CEO as the agent. This agent has several principals (stakeholders) such as the MOH (tutelage sector) on one end and the community on the other, as well as doctors, licensed employees, unions, etc. . some or all of whom can be represented on the hospital board. To simplify, we take a case where the manager has two principals, and assume they are the MOH and the community. The MOH's primary objective is to reduce the costs of the sector given minimum standards of quality -- a goal partly achieved through reductions in transfers for public health provision. The more a hospital gets its financial house in order, through cost-recovery and cost-effective service provision, the closer the I1 MOH gets to fulfilling this objective. The communities dependent on public hospitals have different and potentially conflicting objectives. Public hospital users, or "stakeholders" (Savage et. al., 1997), want the best possible care at the lowest possible price, especially since the previous system provided the possibility of universal coverage.6 Prima facie, the objectives of these two principals are in conflict under the new law in Lebanon. A further dimension is the agent's tasks. To simplify, we assume that the hospital manager under the new law has two main tasks: to control costs and to improve the quality of health care provision. The former of these tasks is easily measurable while the latter is not, but has important equity implications. A similar question about the incentive tradeoffs between prospective payment and cost reimbursement systems in the United States has been analyzed by Ma (1994) using the multitask agency approach (Holmstrom & Milgrom 1991). In this model, the hospital allocates its efforts between cost reduction and quality enhancement. Along similar lines, this approach allows for an analysis of the extent to which hospitals in Lebanon, in having to internalize their production costs once corporatized, risk resorting to excessive cost reduction, and compromising quality. A desirable objective of design would be for corporatized hospitals to internalize the benefit of quality as well. For any given public hospital, it is clear that controlling costs will be a more measurable task than the improvement of the quality of health care provision. It remains to be established whether principals are united in their demands on the agent or not. In a 6 The idea of considering hospital users and/or the "community" in general as "stakeholders" or principals is fairly prevalent in the healthcare literature. Among the possible hospital stakeholders 12 micro-organizational setting such as a public hospital, this task is more difficult than for the macro-policy-making example of GATT, illustrated by Dixit (1996). Deterrnining the degree of principal coordination can be done by looking at the principals' channels of influence, in terms of (1) appointment rights, i.e., rights principals possess because of the manner in which they came to occupy their positions, and; (2) decision rights or the formal and informal prerogatives of principals once they are appointed to a board, defined by law and convention. To simplify, the main difference between rights (1) and (2) is that the former yield power that emanates from the person, while the latter yield power connected with the position. Empirically, this difference is important as I will illustrate. Take the example of appointment rights. There are cases, where the strength of appointment rights granted, differs. For instance, a local political appointment to a hospital board will enjoy a more powerful appointment right than a politically un- connected community member. A political nominee to the board would also enjoy a more powerful appointment right than a physician who sits on the board representing medical staff in the hospital, but who is not affiliated with the local political leadership in the area. There are other cases, where a principal is not granted a decision right all together, as the following example illustrates: By definition, any public agency has at least two sets of principals, the governmental body (or sectoral tutelage) in charge of it, and its taxpaying beneficiaries (or community). If both principals are present on the enumerated by Savage (1997) and Tucker & Burr (1990) are patients and local commnunities, state and local governments, health plans, professional/trade associations, physicians, and employers. 13 board, they may or may not be coordinated. When a public hospital board does not contain a member of the community, it cannot be representative of it. Therefore, by virtue of the fact that an important principal (in this case a community representative) does not sit on the board, the board would not embody the interests of both principals. In such a case, principals can be considered to be un-coordinated because an important principal does not enjoy an appointment right at all. Such cases from Lebanon shed interesting light on how the manner in which stakeholder representatives come to sit on a board influences the decisions they are able to make. This case also offers an opportunity to analyze the currency of influence behind the differential capacity of principals to provide incentives to the agent, and the circumstances under which a given principal may choose to exert influence7. I analyze the institutional design implications of the Lebanese system in section 5 of this paper. By analyzing the problem of public hospital board formation in Lebanon using this approach, I try to answer the following questions: . If the key principals of public hospitals can be considered to be "uncoordinated", what sorts of outcomes can be expected, and how well does the empirical evidence to date corroborate predicted outcomes? . What can be done about the structure, prerogatives, and manner of appointment of a board to increase coordination among principals? * To what extent is better coordination of principals likely to improve the system? * Dixit's model assumes equal power on the part of the principals to influence the agent. Empirically, we observe significant differential powers ro 7 Aghion and Tirole's (1997) work on the difference between formal and real authority in organizations describes simnilar empirical outcomes. 14 influence the agent, both through appointment rights and (post-appointment) decision rights allocations. Can the design of governance institutions (boards) account and correct for skewed distributions of power? 3.2 Data The conclusions of this paper are based on open-ended and structured interviews, analysis of documents, and draft and published legislation. Hospital budgets, accounts and strategic plans (where available) were also drawn on in the analysis. Between March and September of 1998, I benefited from permission to take part in weekly meetings of the Ministry of Health (MOH) Task Force on Public Hospitals as a participant observer.8 My presence in these meetings was crucial to understanding the sectoral and macro dimensions of public hospital reform in Lebanon, and the day-to-day obstacles encountered in implementation. During the summer of 1999, I benefited from permission to accompany the MOH Ratings Commission to inspect hospitals and assess their standards. Because these visits included public and private hospitals, they were central to understanding the uniform vision for quality and performance that the MOH has for both types of hospitals under the new, corporatized regime. During the summer of 1997 and the Spring of 1998, two rounds of introductory, then open-ended interviews were carried out. These were with the Director General of the Ministry of Health, the Minister's advisor in charge of legal matters, four middle managers in the MOH in charge of public hospital management and finance (the Directorate of Medical Care), procurement (the Procurement Division) and accounting (the Accounting Division), and a total of 6 directors and board members of the three first hospitals slated for corporatization - Nabatiyye, Tannourine and Qartaba. Along with 15 many other things, the tradition of serious research on the public sector disappeared during the war in Lebanon. Introductory interviews were crucial in explaining my professional affiliations, and establishing a rapport with my interviewees. Substantive discussions would typically begin with a second meeting. Another series of interviews was carried out during the Spring and Summer of 1999, with 11 board members and directors of newly corporatized hospitals, this time using a specific set of questions developed based on Provision #14 of the Decree on Finance and based on Provisions #9-11 of the Decree on Personnel. These interviews lasted two hours on average, and began with an explanation of the approach, including definitions of decision rights and decision rights allocations to ensure that interviewees had a uniform understanding of both the approach and the questions. Some of the interviews were carried out in two parts or supplemented with an additional interview for clarifications. Also interviewed were the current Minister of Health, Karam Karam and his advisors in charge of public hospitals. Implementation of Law #544 began in 1998, once the decrees were drafted. To date, only four out of 17 public hospitals have begun to function under the new regime. These cases form the empirical evidence this paper is based on. Because of the dire need for public health provision in Lebanon, the nomination of further boards of directors for public hospitals is underway, and more hospitals are expected to adopt the system in the next year. However, the intention of the MOH is to amend the hospital corporatization decrees. In the meantime, some of the information I have obtained on the weaknesses of the system and used as empirical evidence in this paper, constitutes criticism of sectoral 8 See Pomper (1991) and Jorgensen (1989) for a review of the benefits and constraints of participant observation as a qualitative research method. 16 structures and policies by people employed in the sector, and can compromise the professional positions of its sources. As a result, the names of both individuals and their affiliations are kept confidential in this version of the paper, as the reform proceeds. The hospitals corporatized to date are Nabatiyye, Qartaba, Dahr el-Bachek, and Tannourine. The eleven board members and directors interviewed are from these hospitals, but their names are referenced in this paper as numbers (1-11), and their affiliations are omitted. The objective of this paper is not to arrive at incontrovertible conclusions about the system in Lebanon, nor are such conclusions possible given the limited sample of hospitals corporatized to date. Instead, this paper seeks to explore ways of understanding the problem of board design, in anticipation of a time, in the near future, when the empirical evidence from Lebanon and elsewhere will be richer and both the application of models and the conclusions can be more definitive. 4 The case of Lebanon: Background and policy reform The Lebanese public hospital sector experienced a period of deterioration in coverage, quality of service and financial management during the war from 1974-1990. By 1990 the sector was providing a set of perverse incentives. For instance: . Incentives for uninsured patients to seek expensive private care because the quality of care at public hospitals was low and provision was erratic. The Ministry of MOH had begun to reimburse uninsured patients who sought private care during the war in order to ensure that all those in need of health care were able to get it without having to travel during battles. Given that the uninsured constitute 44% percent of the population, this policy resulted in a rapid escalation of public health expenditures, 77% of which 17 went toward the purchase of medical services from the private sector in 1994, when the reform was launched (MOH reports and data). * Incentives created by the cost reimbursement system, for physicians to choose to hospitalize patients for interventions that could be provided on an outpatient basis, and for hospitals to use high-cost interventions when lower-cost treatments would be sufficient. Not surprisingly, cost reimbursement also created incentives for over- billing, especially given expected and actual arrears on the part of the MOH. . In the public hospitals, eroded public-sector wages and compressed pay scales. These created incentives for public hospital staff to absent themselves from their positions, and seek employment in the private sector in order to supplement their income. * Weak incentives and meager means for hospitals to gather and use information that would improve their performance, and an even weaker regulatory capacity at the level of the Ministry of Health to oversee the operation of public hospitals. * No consumer protection policies, and therefore weak incentives on the part of hospitals to ensure that they were satisfying community needs and equity considerations. Despite the possibility of government reimbursement, poor patients have difficulty accessing private hospital services, and when they do receive care, they are often asked for significant co-payments. Those who were most politically connected benefited most from the cost reimbursement system. 4.1 The declared objectives of the reform As part of the effort to restructure the public health sector, a law was drafted to corporatize public hospitals by granting them a degree of fiscal and managerial autonomy 18 Corporatization grants public hospitals their own governing board, thereby delegating some of the regulatory authority of the MOH, but retains the MOH as residual claimant on the hospitals. As part of their autonomous status, hospitals have the right to charge patients for their services to develop a revenue base that would gradually replace transfers received from the MOH. The objective of the law on public hospital autonomy is to provide: * Incentives for hospitals to improve the quality of care they offer while keeping costs under control, thereby satisfying the health sector's equity objective of providing good quality affordable health care for low-income and uninsured patients; * Incentives for hospital management to be responsive to the sector's cost reduction priorities. Making hospitals financially autonomous reduces (and eventually stops) the need for transfers; O Incentives for hospitals to be more attuned and responsive to specific local needs, especially in preventive and basic health care. Central to how well hospitals achieve these objectives is hospital board effectiveness in regulating the activities of their hospitals. The design and prerogatives of hospital boards, discussed later, are therefore key. Under the new, corporatized system, hospitals sign a service contract with the MOH, civil service bureaus (e.g., the army and internal security administrations), insurance companies and other private purchasers. Hospital own-source revenue is raised through private sector purchases and through patient contributions to the price of treatment partly covered by the MOH. Under the new cost-sharing rules, uninsured (MOH) patients are required to pay 5% of the price of treatment at public hospitals, while the MOH contributes the remaining 85% - effectively "purchasing" services from its own 19 hospitals.9 The new system continues to provide universal insurance for the time being. Eventually, benefits (or MOH contract privileges) will become means-tested in the sense that public hospitals that do not break-even will cease to operate.") Hospitals are to prepare and agree upon an annual Strategic Plan with the MOH, which constitutes a basis for the MOH's continued purchasing of services from the hospital. To encourage use of public hospitals, the MOH insurance scheme is available to only 15% of private hospital bed capacity, while it covers 75% of public hospital bed capacity. Today, hospitals are receiving a one-time transfer ranging from 300 million to three billion Lebanese Pounds (USD199,000.00 to USD1,989,000.00) depending on their size, to help jump-start their autonomous operations. The years 1999 and 2000 are being considered by the MOH as trial periods for the reform, with the objective of reaping lessons of experience and improving the system (Interview with Roger Sfeir, Advisor to the Minister of Health). 4.2 Describing the institutional design: Principal features of the legal structure governing corporatized hospitals Law #544 mandating the "Establishment of Public Enterprises for the Management of Ministry of Public Health Hospitals" was promulgated in 1996. The simple three-page document outlining this law is followed by five Implementation Decrees that lay out the technical details and instructions for applying the law. Laws are voted on in Parliament. Implementation Decrees are drafted by the ministry concerned, in consultation with legal, administrative, and financial experts in the various sectors, 9 The remnaining 10% is to be covered from the hospitals' profit margins. 10 Given the geopolitical nature of public hospital care provision in Lebanon, the closing down of unprofitable hospitals would be rationally desirable, but politically difficult. 20 including the Ministry of Finance, and then submitted for ratification by the Council of Ministers. Law #544 mandates the followingi I: 1 A public enterprise (also, "public health enterprise", or "public hospital board" in this paper) is to be founded to manage each public hospital in the country. Public health enterprises are to enjoy financial and managerial autonomy, subject to the supervision of the Ministry of Health. Such enterprises are subject to regulation by the Ministry of Finance, the General Accounting Office, and the Central Inspection Office. 2 The revenues of such public enterprises are constituted of: (a) central government transfers; (b) fees for services; (c) other sources. 3 The Ministry of Health's responsibilities include the definition of sectoral strategy, the coordination of health provision at the national level and the rationalization of the sector. 4 The drafting of five Implementation Decrees defining: (a) The Appointment of Boards of Directors and Ministry Representatives; (b) Financial Regimes for Public Hospital Enterprises; (c) Personnel Matters; (d) Compensation; (e) Internal Administration of Public Hospital Enterprises. 5 The determination of fees for services, patient contributions to fees and budgetary matters, including MOH transfers to public hospitals. "1 The following items are translated from the Arabic text of Law #544. 21 6 The Minister of Health's responsibilities and prerogatives in establishing collaborative agreements among public health enterprises, and between public health enterprises and medical schools domestically and internationally. 7 The determination of the size of boards of directors for public enterprises. 8 The establishment of a consultative committee to study the impact and implementation of public hospital autonomy. The law contains two additional Items, 9 and 10, mandating the drafting of the five Implementation Decrees defined in Item 4 above, and activating Law #544 upon its publication in the Official Journal, respectively. 4.3 The lynchpin of the system: The Governance Decree and its implications The decree outlining conditions for The Appointment of Boards of Directors and Ministry Representatives (henceforth, the "Governance Decree") determines the size, composition, prerogatives and MOH representation on/oversight of public health boards. The detailed content of the decree underscores the centrality of the board to the operation of corporatized hospitals, and the importance of its governing mechanisms in advancing or retarding the goals of efficiency and coordination. Some elements of the Governance Decree are important to examine in light of Decree # 4517 (1972) -- the legal underpinning which defines the establishment and operations of all Public Enterprises and Autonomous Agencies. The following discussion draws on both decrees to analyze salient aspects of the institutional design of public hospital governance. The particular elements that are important in this context include: Stakeholder (principal) mix 22 The decree stipulates that board members should have a background in medicine, business administration, finance, law, or public health. However, apart from listing a restricted set of possible specializations, the decree does not ensure that board members have the required skills to represent (at least the most important and obvious) stakeholders, such as the user community, medical staff in the hospital, the MOH, ...etc. Hence, the focus is more on defining eligibility to the board, than on ensuring representativeness on the board. Manner of appointment of board members (principals) The process of selecting board members is highly ambiguous. Provision #2 stipulates that the board is appointed upon the recommendation of the MOH through a decree to be ratified by the Council of Ministers. Among the important issues to clarify are: how the MOH forms the list to be submitted for ratification, what the criteria used are, and how immune from adverse political influence the system is, keeping in mind that responsiveness and accountability to political demands are desirable features. Empirical and implementation evidence to date point to unclear/inadequate criteria in the selection process, as well as politicization in the choice of candidates, which have led to the administrative paralysis of some newly inaugurated and much needed hospitals. One important reason for this paralysis has been the lack of coordination between board members. The system places a large onus on the Minister of Health to select the right people and negotiate their appointment. 23 The hospital manager (agent) The manner of appointing hospital managers is unclear. Although the decree does state that the hospital manager is to be appointed by the hospital board, it contains no further detail on the selection and appointment process.12 Furthermore, the practice has ranged from the board making recommendations that the MOH may or may not accept, to a local political leader submitting one name to the Minister of Health, who then recommends the appointment without consulting with Ministry cadres nor with the hospital board, nor with the MOH division in charge of public hospitals. The hospital manager sits on the board of directors ex officio and is responsible for the day-to-day running of the hospital. Because of his/her informational advantage, he/she has the potential of wielding important influence on the board, despite his/her non-voting position. * The extent of MOH regulatory responsibility decentralized to the board In sharp contrast to the weak structure described above, the responsibilities of the board are fairly significant. The board and hospital manager's responsibilities range from setting the policy and administrative direction of the hospital, to overseeing inpatient and outpatient service provision, quality control, cooperation and collaboration with educational organizations, setting policy and strategy for various departments within the hospital, setting the annual strategic plan and budget for the hospital, and overseeing contracts and collaboration with the private sector (Translated from the Governance Decree). 12 Decree #4517 states that the director is to be appointed by the Council of Ministers upon the recommendation of the relevant sectoral ministry, in this case the MOH. This Decree also stipulates that the salary of the director is set by the Council of Ministers. 24 Although, as this list shows, the board has extensive responsibilities, there are few areas in which the board and management can make decisions without clearance from higher level authorities. In only four out of twenty decision rights/areas of responsibility devolved to the board by the Finance Decree, can the board actually make decisions without clearance from either the MOH or the Ministry of Finance, or both (Mubarak, 1999). These are instances of transfer of responsibility without the transfer of full authority, and they weaken incentives for optimal decision making. * Sectoral oversight, or accountability between MOH and hospital The MOH oversees the day-to-day operation of the hospital through its principal and voting member of the board - the ministry delegate (or representative). This principal's objective is to influence the operation of the board, by aligning the hospital manager's incentives with those of the MOH, thereby ensuring that sectoral standards and priorities are satisfied at the hospital level. The MOH's oversight and regulatory functions, carried out partly through the MOH delegate, are well laid out in Decree #4517. * Appointment of MOH delegate Similar to the ambiguity surrounding the appointment of the hospital director, it is not clear how the MOH delegate is appointed. Provision #10 in the decree only defines two aspects of this appointment: the five-year term and the requirement that the delegate be a MOH civil servant of a certain grade or above. Crucial issues such as how this person is selected, how close to some key functions of the administration such as finance and procurement he/she can be, what his/her relationship to the local and/or political community should or should not be, or at least, his/her area of 25 specialization, receive no mention. As a result, despite the fact that all MOH delegates to hospital boards enjoy the same set of decision rights and one vote on the board, some of them can exert an excessive degree of influence on the hospital, anid others not enough. Risk transfer Similar to the ambiguity surrounding the appointment of board members, a degree of ambiguity surrounds the degree of risk borne by board members, the hospital director and the MOH delegate for the performance of the hospital. Apart from defining board member remuneration per meeting, the decree makes nIo mention of the consequences of bad performance. As a result, both in regards to term renewal and in regards to compensation, the financial risk of hospital insolvency on the hospital board appears to be zero. Given local conditions and the unfavorable reputation of the public sector in Lebanon today, the reputational consequences that hospital managers and board members bear can also be relatively minor. In summary, the institutional structure of corporatization in Lebanon is strong in some areas and weak in others. It is strong (and ambitious) in that it seeks to deconcentrate a significant degree of administrative, fiscal and regulatory responsibility from the central administration of the MOH down to the hospital level. Reallocating decision rights down to the level of agents with the information needed to make decisions is a way of improving organizational output. On the other hand, the design of the Lebanese system is weak because it is replete with ambiguities that allow for much variance in outcomes depending on the personalities in place. This is particularly 26 apparent in the choice of principals, the choice of the agent and the definition of their decision rights. 4.4 Examining partial empirical evidence The following discussions are based on two illustrative provisions from two decrees: the Decree on Finance and the Decree on Personnel. I analyze the provisions to understand whether the extent to which the system in Lebanon can generate Dixit's third best, how much of this is due to lack of principal coordination, and to point to other factors that might be at play.'3 For the purposes of this analysis, we take Dixit's conclusions on the making of economic policy as a point of departure: the difficulty of achieving good perfornance in government is due to the fact that principals tend to be uncoordinated, incentives weak, and outcomes third best. The question then becomes how uncoordinated principals on Lebanese hospital boards are in practice, and what can be done to improve the equilibrium. In Table 2 below, the "expected outcome" listed in the second column, corresponds to the "policy action" in the same row, mandated by the decree. I treat the expected outcomes as hypotheses for how the system can be expected to behave, and provide, following the table, a discussion of the degree to which the empirical evidence to date supports the hypotheses. The Policy Options presented in the first column of Table 2 are taken from Provision #14 in the Decree on Finance, which mandates the possible actions a manager can take in case of hospital budget deficit. The Policy Options (1-5) constitute recommendations that the hospital manager can make to the board, to 13 Using the Decree on Boards of Directors as a baseline, a similar analysis can be carried out on the remaining decrees. 27 cut/control costs (Translation from the Decree on Finance, Provision #14). Provision #14 was selected for this analysis because it touches upon a broad range of management and finance decisions, and because the policy options it offers are amenable to analysis as hypotheses about the behavior of the system. 28 Table 2: Identifying outcomes based on the Decree on Finance Policy Options Expected Outcome and Empirical Findings for Hospital Brief Reasoning Boards 1. Recommending Expected: Excessive fee Outcomes 1. and 2. have occurred in some hospitals, an increase in fees increases. but not in others. They have not occurred where Reason: No community board members are also members of the community, representative on the board. originating and residing in the community. On the other hand, these policy options have been a 2. Recommflendintg Expected: Patient contributions problem where none of the board members are an increase in could be set too high. selected from the community, especially not the pantincrase in culdbe et to hgh.hospital director. patients' Reason: No community contributions to representative on the board. fees 3. Deciding to This coordination problem is resolved at the level of the central administration of the increasefees MOH, which sets rates to be charged to insurance companies by all public hospitals. In charged to practice, because policy option 3 is difficult (and impracticable) to implement at the insurance agencies. level of a hospital board, it is a very weak (or "hollow") decision right. The author of policy option 3 assumed an imperfection in the insurance market that is resolvable through regulation at the level of the public hospital board. The MOH's retention of a central decision right over such an issue, if a market imperfection indeed exists, is a good idea for a small country like Lebanon where regional idiosyncrasies and the need to adapt to differential market conditions are relatively minor. 4. Deciding to Ex-oected: Frequent Decision right not implemented in any of the hospitals to date. Reasons are fear of social sanction increasefirst-class implementation of this option. and the reticence to develop a reputation of being hospital fees. Reason: Lack of community expensive, while the goal is to encourage use of representation on the board. public hospitals.. 5. Recommending Expected: Strong influence of Although none of the corporatized hospitals have to MOH and MOF MOH delegate on the board resorted to this policy option to date, interviewees that the deficit be could rule this out every time, have mentioned and expressed concern for influence covered through even when necessary. Weak in both directions, depending on the MOH delegate reservefunds. influence of the MOH on the appointed. board could ratify such recommendations, when they are not necessary. Reason: Possible randomness with which MOH delegate is selected. 29 It is interesting to note that policy options 1 and 2 did not result in the expected outcomes in hospitals where board members are also members of the community served by the hospital. Social sanction, reputation, and a degree of altruism have prevented board members from raising fees (Interviews with board members 1-7). In the case where the hospital director and board members are not from the community, complaints have been filed by patients that hospital fees are too high. These results shed interesting insights on the way we might think of principals and principal coordination. The Lebanese hospital boards do not include a community representative, while in other countries such as Columbia and France, the boards include an elected or appointed community representative (Discussions with health policy experts from New Zealand and France, 1997/1998). However, as the case of Lebanon illustrates, the physical presence of a community representative is not necessary if community "interests" are represented. This is an especially interesting proposition if the objective is to keep boards small, for reasons I will discuss below.'4 Policy option 3 presents an example where the presence of a principal is not necessary if the coordination problem is resolved at a higher level in the administration. Hence, although insurance companies are important stakeholders in a hospital system (Savage, et al., 1997), the case of Lebanon provides an example of their interests being represented through means other than a principle, further reducing the need for principle coordination on a board. 14 Furthermore, if we are exclusively concerned with coordination, it appears preferable to exclude the community from the board, such that the agent (hospital director) is certain to respond to the MOH's incentive to reduce cost. The agent's problem then becomes that of fulfilling the right social welfare function (because they are not alerted to community needs), but the board would be more coordinated. 30 Policy option 4 presents a decision right that has not been exercised to date, and is unlikely to be exercised because public hospital users are by definition those who seek to pay the lowest possible prices for health care, even if they can afford to pay higher rates (Interviews with board members 1-6 & 8). On the rare occasions when first class service has been requested, the MOH recommendation for first class fees has been applied because the community members of the hospital board have not wanted to be seen as trying to exploit patients who could otherwise afford to seek private sector care. They have also done this in order to encourage people to use public hospitals. Again, social sanction and reputation have played an important role to date, and good business skills have certainly contributed. The reticence to exploit this provision is an indication that in some instances, the manner in which principals are appointed has contributed to controlling the price of care charged to patients.15 Policy outcome 4 raises questions similar to those discussed above, with respect to how one might define who the principals are, and how important it is for all of them to influence the agent directly. For policy option 5, the closer the MOH delegate is to the treasury and finance functions of the MOH, the more influence he/she can wield in this very important area (Interviews with board members 1-11). The amount of finance a hospital has partly determines the degree to which the hospital can pursue aggressive development and capital investment strategies. Access to finance and a tight reign over use of finance are necessary for survival in the face of stiff private sector competition. A similar (predicted) result also applies to two other decision rights not listed in the table: the right to request a treasury loan (option 6) and the right to request a private sector loan (option 7). None of 1 5 The reticence to charge high fees may also have a positive impact on equity considerations, although this work cannot substantiate it. 31 the hospitals corporatized to date have attempted to exercise these rights, however interviewees expected the same type of influence to result from the mix of principals present on the hospital board."6 4.5 Placing the evidence in perspective Over a decade ago, as the health care market became more competitive and accelerated the drive toward organizational forms that split purchaser from provider functions, the literature on hospital board effectiveness in US markets listed a number of challenges (Shortell, 1989). As hospitals moved from relatively benign to competitive environments, they needed smaller, more nimble and risk-taking boards, composed of members that were focused on strategy, specific expertise, evaluation and accountability. These boards are closer to the boards of competitive firms than to the benevolent, community notable-type boards of hospitals in previous, less demanding market environments (Shortell, 1989; Kovner 1985; Delbecq & Gill 1988; Weiner & Alexander 1993). Since the late 1980s, non-profit boards across sectors have moved in the direction outlined then, and the focus continues to be on smaller-sized boards with fewer insiders, and responsibilities related more to the ratification and monitoring of policy, than to direct involvement in specific operations (Taylor, Chait & Holland 1996) Hospitals in Lebanon are facing a similar set of challenges, among them: * Managing diverse groups of stakeholders (principals); * Involving physicians in the management and governance process; 16 The analysis carried out based on Table 2 can be extended to other areas of hospital board decisions using the remainder of the Finance Decree in addition to the Internal Administration, Personnel, and Compensation Decrees, to a larger and more detailed survey design, from which I expect broadly similar findings. Extensions of this work would be helpful in substantiating testable hypotheses, and in developing a method for analyzing the institutional design of hospital boards. 32 * Responding to the needs of hospital restructuring; * Meeting the challenges of diversification; * Understanding and carrying out strategy formulation; * Balancing equity and efficiency considerations. The design of Lebanon's public hospital boards is similar to the "new" hospital board in some ways, and different in other, important ways. It is closer in its small size, focus on strategy, and representation of stakeholders. It is further in its capacity to assume risk and carry out evaluation and in its accountability. The examples discussed in Table 4 bring the capacities of the Lebanese boards to bare in ways that I expand upon below. For example, despite the fact that public hospital boards in Lebanon do not include a member who is officially appointed as "community representative", commnunity representation is not compromised because some board members fulfill a dual function of, for example, "doctor" and "community representative." This manner of selecting board members resolves part of the principal coordination problem while helping keep the board size to a minimum. In moving toward more technocratic boards, it is important to ensure that the community continue to be represented on the board in some manner, without expanding the size of the board significantly. In a similar manner, policy option 3 illustrates that keeping some decision rights at the level of the central ministry serves to unify policy and reduce costs of principal coordination, contributing to the nimbleness of the system. On the other hand, randomness in outcomes based on important policy options 5, 6 and 7 indicates that the system is weaker on the financial management side, perhaps 33 including the financial accountability side, although more evidence will be required to establish this. What is certain is that the politicization of boards has resulted in some loss of transparency and some non-technocratic decision-making and strategy formulation. This suggests that public hospitals in Lebanon today may not be in the best position to respond to the needs of restructuring and adaptation to a competitive market. 4.6 Further evidence from the Decree on Personnel A reading of an example from the Decree on Personnel serves to illustrate the importance of issues other than principal coordination in the design of hospital governance institutions. These include simple agency and information problems that could result in collusion, political pressure and graft. In provisions #9-11, the decree states that applicants for hospital vacancies must be ranked based on performance on an exam. The hospital board holds the decision right to arrange to carry out the exam. By virtue of his e-x officio position on the board, the hospital manager is a co-holder of the decision right. Given that the decree does not specify any details with respect to the manner in which the exam or examiner is to be selected, the ambiguity has resulted in solutions inferior to first best. A first best outcome could be characterized as one where the hospital board, taking into consideration manager(s) recommendations, short-lists a set of possible examiners, and in consultation with experts in the field, selects the best possible one, ensuring proper screening in its recruitment process. Agency and information problems (but not coordination problems) likely to prevent this first best outcome from occurring include collusion between some board members and the manager at the expense of other board members. This could influence the choice of examiner, in the absence of criteria for this 34 choice. The influence could include political pressure through one of the principals on the hospital board to favor applicants from specific political or religious backgrounds. Under this scenario, outcomes inferior to first best include instances where the decision is made to grant the contract to an examiner with a lower benchmark for "successful" performance. In this case, an exam would have been carried out, but the pool of applicants from which the final choice of employee will be made is of lower average quality, and criteria other than performance on the exam will carry larger real weight. Another outcome inferior to first best could be one where both the choice of examiner and the choice of exam are determined in ways that maximize chances of success for less competitive applicant profiles. In this case, the use of an exam as a screening device would have failed. Empirically, there have been three different applications of these decision rights to date. In one case, the hospital manager and some board members agreed to disregard the examination requirement and established their own point system for the ranking of applicants for positions. This system has not served the hospital well, and has resulted in a number of physicians it wishes to dismiss because of malpractice, and one lawsuit as a result of a dismissal. Among those who were hired, there is evidence that the powerful political appointees to the board had an overwhelming degree of influence on the final choice of candidates (Interviews with board members 2-5). In another hospital, the director of the board is wondering how many competitors he is likely to have for the positions the distant rural hospital is looking to fill. When asked about whether and how his board will comply with the requirement to carry out an exam, he said that they would probably put together a pro forma writing and interview 35 exam for those who do apply, to be evaluated by the board (Interviews with board members 6-7). A third hospital has selected an outside screening committee that is likely to achieve an outcome closest to first best. At the time of this writing, the hiring process was just beginning and no further information was available. Mechanisms that would improve this outcome include amending the Decree cn Corporate Boards to minimize the chance of collusion between board members and the hospital manager(s), and to minimize pressure for political appointments. This would affect appointment rights as well as decision rights. However, given the inevitabie presence of some political interference in multi-confessional countries like Lebanon, aid differential powers on the part of principals to influence the agent, minimal criteria ior the selection of the examiner and exam would move the outcome closer to first best. But if a net improvement in the allocation of decision rights is not feasible at this level. a possible solution would be to reallocate the decision right over exams to a regional or central level, where transparent and technocratic selection of examiners can be carried out. Principals represented on the board would then retain decision rights over other aspects of screening that are related to local specificities and needs, even political preferences within technocratically circumscribed limits. The local choice of candidates would be made from a short list compiled at the national level. The short listing of candidates would effectively provide two levels of screening and scrutiny whic'-, combined with visibility and transparency, would prevent egregious errors from occurring. Countries such as New Zealand and the UK have resorted to similar solutiolns for the appointment of staff to corporatized entities."7 17 Yet another solution could be to do away with the examination requirement all together, as is done in the private sector, and replace it with screening instruments set by each hospital individually. The 36 5. Problems emanatingfrom the system 's institutional design In this section, I synthesize the evidence to date in answer to questions the theory can help inform. I underscore the importance of Dixit's model to some areas, and show how simple agency, influence and information costs are inherent to institutional design in other areas. I also offer some preliminary answers to the questions raised in section 2.1 about the design of corporatization. Representation of key principals/stakeholders On the one hand, there is some evidence that the main principals, are fairly well represented on Lebanese public hospital boards de facto, even if they are not represented de jure. The size and composition of the board parallels fairly well what is suggested in the empirical literature on new, strategy-oriented boards in hospitals. There arguably is some room for enlarging the Lebanese boards slightly, from an average of four members to six, which is closer to the average in non-profit hospital boards internationally. Enlarging the size of boards would allow for a stronger presence for some principals, and/or broader representation of principals. On the other hand, the lack of strict criteria in the selection of principals and differences in appointment rights and power can affect outcomes in a significant way, and are problems of design beyond the principal coordination problem. The result is that even when the important principals are represented de jure, the coordination problem is still not necessarily resolved because of the differential capacity of principals to exert influence over the agent. experience with the point system in one public hospital however indicates that the easy politicization of what are still considered "public sector jobs" may preclude the proper operation of market-like hiring practices in a country where the religio-political map is still expected to be reflected in the distribution of public sector jobs. 37 The currency of influence In the case of Lebanon, three currencies of influence appear to determine the power of principals over the agent. Although this appears to be changing today as the new presidential administration accelerates the push for public sector reform and accountability. The order in which the currencies are discussed reflects their relative importance. The first is political. Appointees of political leaders have tended to wield the most significant influence by any measure. The second is informational, a result found by Aghion and Tirole (1997) in private organizations. The third is technocratic, granting those with skills and experience some leverage over the direction of policy on the board.'8 Two aspects of the manner in which currencies of influence work in hospital boards in Lebanon are important. First, combinations of two types of currencies are what tend to empower principals most. As such, political and informational currencies combined have wielded virtually uncontested influence over the agent. Combinations of political and technocratic currencies have also been fairly powerful. The informational and technocratic currencies on their own have yielded fairly low-powered incentive schemes. The relative importance of these two has been a function of the personalities in place. The political currency on its own has been an important source of influence, but this may be changing today. The relative importance of currencies of influence is a good proxy for the relative influence of appointment rights that principals have when they act on the board. 8 The analysis of power wielded through property rights (decision rights) is not a new subject. See Polenske (1999) for an insightful discussion of the relationship between different types of power, property 38 Decision rights Some decision rights are expansive, while others are fairly circumscribed. For example, by virtue of the fact that they are minimally defined, decision rights over hiring transfer all authority over hiring to the hospital board, thereby devolving a significant degree of power to the board. On the other hand, while the responsibility over procurement is devolved to the board, the authority devolved to board members is circumscribed by virtue of the fact that decision rights are co-held with the MOH and Ministry of Finance, through a series of controls, mostly ex ante. Such "weak" devolution of decision rights might have been intended as a mechanism of controlling agency problems when board members do not bear the risk of procurement decisions, but it is not clear that this indirect mechanism will achieve its objectives without compromising others, such as agility and adaptability to demand. Instead, some level of direct financial risk (and benefit) might be transferred to the level of the hospital manager and board, further strengthening the power of the MOH to influence the hospital on the cost control side. Improving the coordination ofprincipals The types of policy measures that can improve the coordination of principals, and consequently the outcomes, have to do both with appointment rights and decision rights. More homogeneous appointment rights can decrease the variability of the power of incentive schemes that principals can exert over the agent. Such measures can range from a more transparent, technocratic and systematic screening and selection of board members, to the development of a public sector corporate culture combining the rights and development strategies. Contributors to this topic, discussed by Polenske, range from Marx (1967 [1888]) to Parsons (1963) to Poulantzaz (1973) to Weber (1978), and Bowles & Gintis (1986). 39 Weberian and Krepsian notions. While it is important not to overestimate the degree to which the "personality effect" can be controlled, some reduction in arbitrariness is clearly possible in the case of Lebanon. This could either be achieved through benevolent and enlightened top-down selection of candidates for positions (as the government is attempting to do today) or through the establishment of institutions that guarantee a minimum degree of continuity across political regimes. In the area of decision rights allocations, withdrawing some decision rights all together from the level of the hospital board, and reallocating them to the level of the central administration or some other third body can serve to decrease the need for coordination among principals. For example, in instances where unanimity (or at least a super-majority decision) is desirable but cannot be guaranteed by the board, decisions are perhaps best taken outside the board. The proclaimed objectives of corporatization From the case of Lebanon, there is evidence that the cost-quality coordination problem is difficult to resolve. In cases where the cost of care has been kept low, there have been complaints about quality. In cases where quality has been improving over time, evidenced by demand for the hospital's services, there have been complaints that costs are too high. This may well be a perennial problem for hospital management worldwide. The data collected for this paper does not allow for stronger conclusions about the situation in Lebanon. What is certain is that the difficulty of achieving this balance in Lebanon is partly due to the lack of transfer of financial risk from corporatization. Although the law does 40 specify that hospitals are to be financially solvent (after an initial transitional period), it does not indicate what the consequences of violating budget constraints are for board members and the MOH delegate on the board. By default, the risk of financial default is assumed by the MOH, the recurrence of which would presumably cause the hospital to be shut down. There are no explicit financial incentives relating, for example, salary bonuses to cost containment that could improve the expected outcome. This is one of many design deficiencies outside the scope of the principal coordination problem, which will need to be handled through the amendment of the Implementation Decrees. The power of incentives The power of incentives is high in terms of the agent's response to some principals in some cases. For example, hospitals whose board members are chosen from the community, tend to have boards that are sensitive to social sanction. This is not the case in hospitals whose boards are selected from outside the community, where increases in fees have been easy to implement, and have resulted in complaints from the community. In terms of the financial solvency of the hospital, incentives tend to be relatively low powered across the board for reasons discussed above. This outcome corroborates the low-risk, low power of incentives conclusion from the moral hazard model. 6. Concluding remarks: from theory to practice and back This paper has proposed the common-agency multitasking approach as an analytical lens to understand the problem of board design in corporatized public hospitals. On the theoretical side, it has shown that while principal coordination is indeed a problem, a more detailed and variegated approach is necessary to understand problems 41 of govemance when the model is applied to a micro-organizational setting. In particular, a closer understanding of differential capacities on the part of principals to influence the agent, and the various currencies of influence appear key to a more detailed modeling of the problem. On the empirical side, the application of the multi-tasking common agency model has raised questions that shed light on some ideas for the improvement of the institutional design of public hospital corporatization in Lebanon. These ideas are the subject of a different, policy-oriented paper. They have been partially included in summary form as part of Appendix B. 42 7 References Aghion, Philippe and Jean Tirole. 1997. "Formal and Real Authority in Organizations". Journal of Political Economy. 105:1, 1-29. Alexander, J. The Changing Character of Hospital Govemance. Chicago: Hospital Research and Educational Trust. Baker, G. and K. Wruck. 1990. "Organizational Changes and Value Creation in Leveraged Buyouts: The case of O.M. Scott & Sons Company". Journal of Financial Economics. XXV. Barnum, Howard and Joseph Kutzin. 1993. Public Hospitals in Developing Countries. Baltimore, MD: Johns Hopkins University Press. Barzelay, Michael. 1991. "Managing Local Development: Lessons from Spain". Policy Sciences. XXIV:3, 271-290. Barzelay, Michael. 1992. Breaking Through Bureaucracy: A New Vision for Managing in Government. Berkeley, CA.: University of California Press. Barzelay, Michael. 1993. "The Single Case Study as Intellectually Ambitious Inquiry". Journal of Public Administration Research and Theory. 111:3. Bernheim, B. Douglas and Michael Whinston. 1986. "Common Agency." Econometrica 54, No. 4:911-930. Bird, Richard and Christine Wallich. 1993. "Fiscal Decentralization and Intergovernmental Relations in Transition Economies". Policy Research Department Working Papers, Public Economics Division. Washington, D.C.: World Bank. Bogdan, R. 1972. Participant Observation in Organizational Settings. New York: Syracuse University Press. Bowles, Samuel, and Herbert Gintis. 1993. "The Revenge of Homo Economicus: Contested Exchange and the Revival of Political Economy." Journal of Economic Perspectives. VII: 1. Brealy, Richard A. and Stewart C. Meyers. 1991. Principles of Corporate Finance, 4th Edition. NewYork: McGraw-Hill. Cremer, Jacques, Antonio Estache and Paul Seabright. 1995. "The Decentralization of Public Services: Lessons from the Theory of the Firm", in Antonio Estache (ed.), 43 Decentralizing Infrastructure: Advantages and Limitations. World Bank Discussion Paper #290. Washington, D.C.: World Bank. Delbecq A. and S. Gill. 1988. "Developing Strategic Direction for Governing Boards." Hospital and Health Services Administration. XXXIII: 1. Dillinger, William. 1994. "Decentralization and its Implications for Urban Service Delivery". Washington, D.C.: UNDP/UNHCS/World Bank Urban Management Program. Dixit, Avinash. 1996. The Making of Economic Policy: A transaction-cost politics perspective. Cambridge, MA: MIT Press. Duca, Diane. 1996. Non-Profit Boards. New York: John Wiley & Sons Inc. Eid, Florence. 1998. Unpublished Policy Note: "Assessing the MOH's Project to Corporatize Public Hospitals." Beirut: The American University of Beirut. Espinoza, Jose and Mario Marcel. 1994. "Descentralizaci6n Fiscal: El caso de Chile". Serie Politica Fiscal, No. 57. Santiago: Comisi6n Economica Para America Latina y el Caribe. Fenno, Richard. 199?. Watching Politicians: Essays on Participant Observation. Gertner, R. and S. Kaplan. 1996. "The Value-Maximizing Board". Working Paper, University of Chicago. Gibbons, Robert. 1997. "Game Theory and Garbage Cans: An introduction to the economics of internal organization". In Nonrational Elements of Organization Decision Making, edited by R. Stern and J. Halpern. Cornell: Cornell University Press. Government of Lebanon. 1996. Law #544 Mandating the Establishment of Public Enterprises for the Management of Ministry of Public Health Hospitals, and its Implementation Decrees. Government of Lebanon. 1972. Decree # 4517 Defining the Establishment and Operations of Public Enterprises and Autonomous Agencies. Govindaraj, Ramesh and Mukesh Chawla. 1996. "Recent Experiences with Hospital Autonomy in Developing Countries -- What can we learn?". Boston, MA: Data for Decision Making Project, HSPH. Harding, April and Alex Preker, eds. 1999. "Conceptual Framework for Reforms in Health Care Delivery," forthcoming in Innovations in Health Care Delivery: 44 Organizational Reforms Within the Public Sector. Washington, DC: The World Bank. Hart, Oliver. 1995. Firms, Contracts and Financial Structure. Oxford: Clarendon Press. Hart, Oliver, Andrei Schleifer and Robert Vishny. 1997. "The Proper Scope of Governnent: Theory and an Application to Prisons". The Quarterly Journal of Economics. November issue. Holmstr6m, Bengt. 1982. "Moral Hazard in Teams." Bell Journal of Economics. XIII. Holmstrom, Bengt. 1989. "Agency Costs and Innovation". Journal of Economic Behavior and Organization. XII:3. Holmstr6m, Bengt and Paul Milgrom. 1990. "Regulating Trade Among Agents." Journal of Institutional and Theoretical Economics 146, No. 1:85-105. Holmstrom, Bengt and Paul Milgrom. 1991. "Multitask Principal Agent Analysis: Incentive Contracts, Asset Ownership, and Job Design." Journal of Law, Economics and Organization 7, Special Issue: 24-51. Holmstr6m, Bengt. MIT 15.946 - Spring 1996: Lectures of the Special Seminar in the Economics of Organizations. Jensen, Michael. 1988. "Takeovers: Their Causes and Consequences". Journal of Economic Perspectives. 2: 21-48. Jorgensen, Danny. 1989. Participant Observation : A Methodology for Human Studies. Newbury Park, CA.: Sage Publications. Kaul, Mohan. 1997. "The New Public Administration: Management Innovations in Government". Public Administration and Development. XVII: 1, 13-26. Kelsey, Jane. 1995. The New Zealand Experiment: A world model for structural adjustment?" Auckland: Auckland University Press. Kikeri, Sunita, John Nellis & Mary Shirley. 1992. Privatization: Lessons of Experience. Washington, D.C.: World Bank. Kovner, A. 1985. "Improving the Effectiveness of Hospital Governing Boards." Frontiers of Health Services Management. II: 1. Kovner, A. 1990. "Improving Hospital Board Effectiveness: An Update." Frontiers of Health Services Management. 111:6. 45 Kreps, David. 1992. "Transaction Cost Economics", teaching note, GSB, Stanford University. Kreps, D. 1993. "Corporate Culture and Economic Theory," in Perspectives on Positive Political Economy, James Alt and Kenneth Shepsle, eds. Cambridge: Cambridge University Press, pp.90-143. Ma, Albert. 1994. "Health Care Payment Systems: Cost and Quality Incentives". Journal of Economics and Management Strategy. 3(1):93-112. Marx, Karl, and Fredrich Engels. 1967 [1888]. The Communist Manifesto, edited iy J.P. Taylor, London: Penguin. Miles, Mathew and Michael Huberman. 1994. Qualitative Data Analysis. Thousuand Oaks, CA.: Sage. Milgrom, Paul and J. Roberts. 1992. Economics, Organization and Manageme it. Englewood Cliffs, NJ: Prentice Hall, Inc. Moore, Mark H. 1996. Creating Public Value: Strategic Management in Government. Caambridge, MA: Harvard University Press. Mubarak, Issam. 1999. "A Study on the Public Enterprises Managing MOH Public Hospitals in Lebanon, Presented Under the Auspices of the Ministries of Health of Lebanon and Italy at the Riviera Hotel, 17.11.1999." (Title translated from Arabic). Unpublished Manuscript. North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. Oates, Wallace. 1990. "Decentralization of the Public Sector: An Overview", in Robert Bennet (ed.), Decentralization, Local Government and Markets. Oxford: Oxford University Press. Oates, Wallace. 1972. Fiscal Federalism. New York: Harcourt Brace Jovanovich. Parsons, Talcott. 1963. "Power and the Social System." In Proceedings of the American Philosophical Society (June). Polenske, Karen R. 1999. "A Property-Rights Framework for Economic-Development Strategies." Unpublished Manuscript. Poulantzas, Nicos. 1973. "Class Power." Political Power and Social Classes, translation edited by Timothy O'Hagan. London: New Left Books and Sheed and Ward. 46 Robinson JC. "Consolidation of Medical Groups into National Physician Practice Management Organizations." Journal of the American Medical Association 1998, in press. Robinson JC. "Decline of Hospital Utilization and Cost Inflation under Managed Care in California." Journal of the American Medical Association, 276(13):1060-1064, 1996. Robinson JC. "Health Care Purchasing and Market Change in California," Health Affairs, 1995 14(4):117-130. Savage, Grant, et. al,. 1997. "Governance of Integrated Delivery Systems/Networks: A Stakeholder Approach". Health Care Management Review. 22(1):7-20. Schleifer, Andrei and Robert Vishny. 1997. "A Survey of Corporate Governance". The Journal of Finance. 52(2):737-783. Schleifer, Andrei. 1998. The Economics and Politics of Transition to an Open Market Economy: Russia. Paris, OECD: Development Centre Studies. Seabright, Paul. 1994. "Accountability and Decentralization in Government: An Incomplete Contracts Model". CEPR Discussion Paper #889. London: Centre for Economic Policy Research. Shah, Anwar. 1991. "Perspectives on the Design of Intergovernmental Relations". Working Paper #726. Washington, D.C.: World Bank. Shirley, Mary and Lixin Xu. 1998. "Information, Incentives and Commitment: An Empirical Analysis of Contracts between Government and State Enterprises". Journal of Law, Economics, and Organization, Vol. 14, No. 2. Shonick, William and Ruth Roemer. 1983. Public Hospitals Under Private Management. Berkeley, CA: Institute of Governmental Studies. Shortell, S. 1989. "New Directions in Hospital Governance." Hospital & Health Services Administration. XXXIV: 1. Silverman, Jerry. 1992. Public Sector Decentralization. Washington, D.C.: World Bank. Silverstone, Brian, et. al., eds. 1996. A Study of Economic Reform: The case of New Zealand. Amsterdam: Elsevier Science B.V. Taylor, Barbara, Richard Chait and Thomas Holland. 1996. "The New Work of the Non-Profit Board," Harvard Business Review. September-October. 47 Tirole, Jean. 1994. "The Internal Organization of Government." Oxford Economic Papers 46, No. 1: 1-29. Weber, Max. 1978. Economy and Society (Vol. I). Berkeley, CA: University of California Press. Weiner, B. and J. Alexander. 1993. "Corporate and Philanthropic Models of Hospital Governance: A Taxonomic Evaluation." Health Services Research. XXVIII:3. Williamson, Oliver. 1997. "Transaction Cost Economics and Public Administration" in Public Priority Setting: Rules and Costs, edited by Peter Boorsma, Kees Aarts and Albert E. Steenge. Dordrecht, The Netherlands: Kluwer Academic Publishers. Williamson, Oliver. 1991. "The Economics of Governance: Framework and Implications," in The New Institutional Economics, a collection of articles from the Journal of Institutional and Theoretical Economics, Texas A&M University Press, pp.54-81. Wilson, James Q. 1989. Bureaucracy: What Government Agencies Do and Why They Do It. New York: Basic Books. World Bank. 1995. Bureaucrats in Business: The Economics and Politics of Government Ownership. Washington, D.C.: Oxford University Press. Yin, Robert. 1994. Case Study Research: Design and Methods. Thousand Oaks, CA: Sage Publications. 48 8. Appendix A POLICY NOTE ASSESSING THE MOH'S PROJECT TO CORPORATIZE PUBLIC HOSPITALS Florence Eid AUB/Massachusetts Institute of Technology feid@mit.edu (forwarded to AUB while in Lebanon). Beirut: 03/686824 VIII.24.98 1. Overview and policy issues The health sector is among the most complex and pressing aspects of administrative reform in Lebanon today, not only because of the urgency of curbing the excessively high costs of the sector, but also because the reform proposes new legislation which is to form the basis of an institutional structure changing the operation of all public hospitals in the country. The objective of the new institutional structure is to improve quality of service in public hospitals (and cut costs in the sector) by granting hospitals a degree of managerial and financial autonomy from the Ministry of Health (MOH). Of great currency worldwide but faced with mixed results, this type of reform is being referred to internationally as "corporatization" and is understood as a middle ground between public sector ownership and management, and privatization. This note summarizes preliminary findings based on work carried out on to date, and makes recommendations on how to proceed. The work entailed (1) analysis of the decrees of application under Law 544 governing public hospital autonomy; (2) interviews with ministry officials and hospital staff and users, and; (3) site visits. The questions I asked through this work have to do with: * The appropriateness of the new structure given macro considerations in the health sector and given lessons learned from the experiments with autonomy that preceded the new Law #544 governing public hospital autonomy; * Progress and obstacles in implementation given the experiences to date of two hospitals, Nabatiyye and Tannourine, and of Dahr el-Bashek, a hospital that attempted autonomy through a public-private association, before Law #544 was passed. 2. Overview offindings to date What follows is a summary assessment of the reform: 1. There exist various gaps in the decrees of application, such as lack of clarity on lines of accountability between public hospital boards and the Ministry of Health, and between hospital directors and hospital boards. 49 2. There are implicit contradictions in the spirit of the law. For example, while some provisions in the decree on finance are clear in the establishment of numerous MOH controls over revenue and expenditure decisions of hospitals and hospital boards, other provisions appear to grant virtually free reign over the transfer of use and ownership of the physical assets of public hospitals. 3. The new autonomous structure is vulnerable to political influence, which appears to have affected implementation in the case of two hospitals, Tannourine and Nabatiyye. Politicization derives partly from the manner in which hospital boards are appointed, the size of hospital boards and their terms, the mechanisms of coordination between the MOH and hospital boards (through the person of the ministry representative), and various insufficiently defined oversight functions of the MOH. 4. Lack of clarity on the objectives and implementation details of autonomy. Among the important actors that remain unclear about the reforms are Ministry of Health middle managers and those below them, public hospital managers and those below them who are not yet fully aware of the content of the decrees of application of the new law governing autonomous hospitals. 5. Insufficient exploitation of lessons learned from the "informal" experience in public hospital autonomy prior to Law No. 544. For instance, one important conclusion from the analysis of the experience of the Support Committee of the Hospital of Dahr el-Bachek is that the five-member board, meeting twice a month, had hardly enough time to address all important policy matters facing the hospital. Given this statement by various members of the retired Support Committee, it is not clear how a board of three members for hospitals of under 100 beds (the majority of hospitals in the country) is expected to be sufficient. 6. Furthermnore, and related to point 5., preliminary findings point to the fact that in most countries where public institutions are vulnerable to "political capture", boards are constituted such that a number of spots are reserved for "political" appointments, while a number of other spots is reserved for "technocratic" appointments, guaranteeing a balanced mix between important political interests and rational policy decisions. Implementing such an idea in not beyond reach for a country like Lebanon. 7. Finally, the above, in addition to readings of the decrees lead to the conclusion that the conceptual underpinning, or model for the proposed reforms is unclear. Information I have been provided through interviews indicates that the French and Tunisian models might have been drawn upon. It is not clear why these in particular would have been selected, nor is it clear that any other lessons learned from international experience have been exploited in conceptualizing the Lebanon reforms. 3. Why not just do away with public hospitals? The reason why total privatization of health delivery should not be an option in Lebanon goes beyond the standard public good/equity considerations. The linchpin of an effective system of private delivery of public services is strong regulatory capacity, which we lack in Lebanon. Instead, the Lebanese public sector has proven to be vulnerable and fertile ground for the politicization and corruption of individual 50 transactions, especially when they are relatively small and numerous, which is the reason why expenditures on the cost reimbursement system in the MOH increased exponentially in the past 7 years. The granting and oversight of contracts under a privatized system requires a regulatory system that is accountable, and that benefits from reliable quality and performance measures. Given the existing regulatory weakness of the MOH, it is not clear that such a system will be instituted and can be effective. On the other hand, bad public health provision will cast further doubt on the capacity of Lebanese public hospitals to deliver such services, and will strengthen the rationale for privatization. Seen from this perspective, a strong and carefully crafted set of decrees of application governing autonomous hospitals is crucial. To achieve this, an interim revision and restructuring of the current decrees is of priority today. Continuing with the decrees we have recently passed will not only create weak and difficult-to- regulate public hospitals, it will also put in place and entrench local interests that will be difficult to remove once we have even clearer evidence of the structure's weakness, probably two years into implementation. 4. What remains to be done - A fresh reading of the decrees of application in view of the uncertainties, complications and obstacles on the ground to date. As I suggest this, I acknowledge the great deal of work that has clearly been put into the current versions of the decrees of application. What is unfortunate is that the few weaknesses they contain happen to be key determinants of success under the new regime. - A review of most relevant international experiences in this area to date, and incorporation of appropriate lessons of this experience into the refining and implementation of institutions of autonomy in the public health sector in Lebanon. Many of the questions currently being posed in Lebanon have already been posed and resolved elsewhere. While it is important to pay attention to the particularities of the Lebanese case, there is no need to "reinvent the wheel" for all aspects of the reform. * The above review would need to be carried out in tandem with discussions/revisits of current macro-considerations in the sector, knowing that such considerations might have evolved since the initial discussions of Law #544. To illustrate, it makes no sense trying to adjust details of the governance and operation of a public hospital without ensuring that such adjustmnents are in line with a clear and well-articulated strategy that accounts for the public hospital implementation constraints we are encountering on the ground. 51 9 Appendix B PRELIMINARY IDEAS FOR THE REFORM OF THE PUBLIC HOSPITAL CORPORATIZATION IMPLEMENTATION DECREES UNDER LAW #544 IN LEBANON. 1 To correct for problems of principal collusion and graft, revise the structure of boards and define prerogatives better. Perhaps rotate the presidency of boards to minimize concentration of power and collusion between board and manager. Also re- think length of terms and conditions for reappointment. 2 To improve probability of principal coordination, establish a system of Rules of Order for board meetings such that important policy matters are guaranteed due process in discussions. Incorporate a quorum requirement into Decree on Boards of Directors. 3 To resolve some collective action problems, ensure that certain decision rights are allocated to agents outside/above the board, especially when unanimity is important, but cannot be guaranteed through the board. 4 Institute training and continuing professional education for board members. This would contribute to the coordination of principals through the development of a common corporate culture. Policy Research Working Paper Series Contact - Title Author Date for paper WPS2706 Household Income Dynamics Jyotsna Jalan November 2001 C. Cunanan in Rural China Martin Ravallion 32301 WPS2707 Financial Intermediary Development Thorsten Beck November 2001 A. Yaptenco and Growth Volatility: Do Mattias Lundberg 38526 Intermediaries Dampen or Magnify Giovanni Majnoni Shocks? WPS2708 Accountability and Corruption: Daniel Lederman November 2001 P. Soto Political Institutions Matter Norman Loayza 37892 Rodrigo Reis Soares WPS2709 Explaining Leakage of Public Funds Ritva Reinikka November 2001 H. Sladovich Jakob Svensson 37698 WPS271 0 Breaking up the Collective Farm: Martin Ravallion November 2001 C. Cunanan Welfare Outcomes of Vietnam's Dominique van de Walle 32301 Massive Land Privatization WPS2711 Identifying Class Size Effects in Miguel Urquiola November 2001 H. Sladovich Developing Countries: Evidence from 37698 Rural Schools in Bolivia WPS2712 Contract Risks and Credit Spread Mansoor Dailami November 2001 W. Nedrow Determinants in the International Robert Hauswald 31585 Project Bond Market WPS2713 Causes of Inequality in Health: Adam Wagstaff November 2001 H. Sladovich Who You Are? Where You Live? Pierella Paci 37698 Or Who Your Parents Were? Heather Joshi WPS2714 On Decomposing the Causes of Adam Wagstaff November 2001 H. Sladovich Health Sector Inequalities with an Eddy van Doorslaer 37698 Application to Malnutrition Inequalities Naoko Watanabe in Vietnam WPS2715 Paying for Health Care: Quantifying Adam Wagstaff November 2001 H. Sladovich Fairness, Catastrophe, and Eddy van Doorslaer 37698 Impoverishment, with Applications to Vietnam, 1993-98 WPS2716 Does Foreign Bank Penetration George R. G. Clarke November 2001 P. Sintim-Aboagye Reduce Access to Credit in Robert Cull 37644 Developing Countries? Evidence Maria Soledad Martinez Peria from Asking Borrowers Policy Research Working Paper Series Contact Title Author Date for paper WPS2717 Bridging the Economic Divide within Raja Shankar November 2001 A. Santos Nations: A Scorecard on the Anwar Shah 31675 Performance of Regional Development Policies in Reducing Regional Income Disparities VVPS2718 Liberalizing Basic Carsten Fink November 2001 R. Simms Telecommunications: The Asian Aaditya Mattoo 37156 Experience Randeep Rathindran WPS2719 Is There a Positive Incentive Effect Truman G. Packard November 2001 T. Packard from Privatizing Social Security? 89078 Evidence from Latin America WPS2720 International Migration and the Global Andres Solimano November 2001 A. Bonfield Economic Order: An Overview 31248 WPS2721 Implications for South Asian Countries Sanjay Kathuria November 2001 M. Kasilag of Abolishing the Multifibre Will Martin 39081 Arrangement Anjali Bhardwaj WPS2722 Japan's Official Development Masahiro Kawai November 2001 J. Mendrofa Assistance: Recent Issues and Shinji Takagi 81885 Future Directions VVPS2723 Using Development-Orineted Equity Helo Meigas November 2001 S. Torres Investment as a Tool for Restructuring 39012 Transition Banking Sectors WPS2724 Tropical Bubbles: Asset prices in Santiago Herrera November 2001 R. lzquierdo Latin America, 1980-2001 Guillermo Perry 84161 WPS27' 25 Bank Regulation and Supervision: James R. Barth November 2001 A. Yaptenco What Works Best? Gerard Caprio Jr. 38526 Ross Levine WPS2726 Applying the Decision Rights Florence Eid November 2001 A. Santos Approach to a Case of Hospital 31675 Institutional Design