From an iron triangle to an iron duet?

zyxwvu zyxwvu zyxwv zyxwv European Journal of Political Research 21: 91-108,1992. @ 1992 Kluwer Academic Publishers. Printed in the Netherlands. zyxwv zyxwv From an iron triangle to an iron duet? Health policy making in Israel YAEL YISHAI Department of political science, University of Haifa, Haifa. Israel Abstract. The paper addresses five constructs of policy making: policy curtain, iron triangle, issue network, policy community and an iron duet. The five constructs are distinguished on the basis of two variables: exclusion, denoting the degree of openness to new participants, and interdependence, pertaining to relationship between participants. A policy curtain prevails in the pre-agenda stage denoting both exclusion and lack of dependence. Iron triangle is characterized by exclusion and interdependence; issue network portrays inclusion and lack of interdependence; a policy community features inclusion and interdependence. A fifth construct - an iron duet combines characteristics of an iron triangle, an issue network and a policy community. The paper suggests that the five constructs of policy making are not only products of political environment but are affected by distribution of resources and primacy of values within a specific issue-area. It further argues that a sequential development may take place following alterations in the division of power between state agencies and organized groups. The involvement of powerful groups of professionals in a policy community, produced by mutual dependence of state and association, may lead to formation of a policy (iron) duet. The historical evolution of specialization entitlement in Israeli health policy serves as a case study to illustrate the major arguments of the paper. The connection between administrative structures and processes of intermediating interests has aroused considerable scholarly attention in the past decade. Questions have been presented regarding both the participants in the policy process and the nature of their relationship. Within the broad spectrum of literature three major concepts may be identified: (a) ‘iron triangle’; (b) ‘issue network’, and (c) ‘policy community’. These concepts present different constructs of the policy process, describing the type of participants and the nature of their interrelationship. They were generally applied to country-wide political environments. The iron triangle and the issue network were found to characterize US policy making; policy communities were mostly identified in the UK (Jordan, 1981). Each of these concepts connotes a junction where decision makers and representatives of organized groups meet; each portrays a different set of relationship between intra-government and non-government actors. This paper analyzes the distinctions between the three patterns. It demonstrates that the three concepts fail to deal with the pre-policy stage where built-in constraints bloc the policy process. The discussion further claims that (a) the prevalence of one form or another of policy making is not necessarily a product of the political environment, but is rather influenced by an ever- 92 changing distribution of resources and values among actors participating in the formulation of a specific policy issue. (b) Policy constructs may follow a pattern of sequential development. An alteration of power relations or value priorities in society may generate a movement from one policy construct to the other. (c) The involvement of powerful groups, such as professional associations, in the policy process affects the type of the policy construct. It is thus not only the number of participants which counts, but their specific role in the policy process. Professionals monopolizing expertise are bound to affect the type of a policy construct. The verification of these assumptions will rest on the analysis of specific sub-sector in health policy-making in Israel between 1950 and 1990: the entitlement of medical specialization. zyxwv zyxwvu zyxwvu Constructs of policy making Iron triangle The concept of iron triangle was developed by the critics of pluralism, who realized that ‘the flaw in the pluralist heaven is that the heavenly chorus sings with a strong upper class accent’ (Schattschneider, 1960: 35). Policy making was seen as taking place in ‘triangles’ composed of the interest groups having a stake in the issue, the relevant administrative agency, and the relevant legislative committee (Wolfe, 1977; Lowi, 1979). The concept of the iron triangle is founded on two assumptions: first, that small circles of participants, including ‘sub-governments’ ,have succeeded in becoming largely autonomous. A stable set of participants coalesced to control fairly narrow public programmes that are in the direct economic interest of each party to the alliance. The iron triangle is thus widely insulated from outside forces in the environment. The second assumption was that the members of the iron triangle are mutually dependent by virtue of their social attributes and shared interests. Policies are made by a limited number of elite members linked together by their desire to promote their common goals. The iron triangle exposes the ‘irony of democracy’ (Dye & Ziegler, 1970) where policy decisions are restricted to the few, although formal rules of the game may be strictly adhered to. A pattern of stable, predictable and relatively closed private relationships between a limited number of interest groups and decision makers is the ground rule of an iron triangle. 93 Issue network The concept ‘issue network’, introduced in the US in the late 1970s, presented a different view of the policy process. In its original version (Heclo, 1978) the issue network protrays characteristics diametrically opposite to those of the iron triangle. The distinctions between the iron triangle and the issue network are not just a matter of degree, but of principle. To begin with, issue networks do not exclude potential participants. In fact, the boundaries between networks and their environment are not clear-cut and their membership is fluid. Actors change, moving in and out constantly. According to Heclo (1978: 102) issue networks ‘comprise a large number of participants with quite variable degrees of mutual commitment or of dependence on others in their environment; in fact it is almost impossible to say where a network leaves off and its environment begins’. Furthermore, members of an issue network are not mutually dependent. The major activists of the policy triangles are members of the establishment possessing political, organizational and economic power. In contrast, those that take the lead in an issue network are the experts and the knowledgeable. Issue networks allow the participation of technocratic specialization in the policy process. Admittedly the network, like any other organization, is imbued with politics and the power game is clearly visible, but skill and expertise dominate the scene. As Heclo has put it (1978: 103): ‘More than mere technical experts, network people are policy activists who knows each other through the issues.’ A new breed of ‘policy politicians’ was described as emerging ‘experts in using experts, victualers of knowledge in a world hungry for right decisions’. Heclo’s answer to the question ‘Who governs?’ thus tilts toward the experts (see also Walker, 1989). Their increased power in the policy process does not imply a mere change in the triangle’s head. In fact, Heclo (1978: 102) stated explicitly that ‘questions of power are still important. But for a host of policy initiatives undertaken in the last twenty years it is all but impossible to identify clearly who the dominant actors are.’ Issue networks are thus open to those demonstrating a capability to contribute to the solution of a governance problem. The ‘Technopols’ are those able to move among the various networks, recognized as knowledgeable about the substance of issues concerning these networks. The melting of the ‘iron’ is produced by means of expertise. The lax definition of technical knowledge leads to an inevitable widening of the policy process. The boundaries of the issue network are widely penetrated by those able to demonstrate relevant expertise. Their autonomy is protected by the ethos of professionalism based on individual responsibility and personal accountability. zy zy 94 zyxwvuts zyxwvut zyxwv zyxwvut Policy community The concept of policy community was introduced by Richardson and Jordan (1979) in their discussion of decision making in contemporary Britain. Policy community has been defined as ‘a comparatively small circle of participants that a civil servant might define as being of relevance for any particular policy’ (Jordan, 1981: 105). A policy community shares some characteristics with both the iron triangle and the issue network. Both the policy community and the issue network are not immune from outside penetration. Both repeatedly incorporate new elements from the environment to take part in the policy process. A shift in the policy focus triggers alteration in the community’s membership and allows the inclusion of new actors. As Jordan (1981: 102) had noted: an issue network is characterized by ‘a dramatic increase in the number of participants’. The policy community, however, is also similar to an iron triangle at least in one respect: it is characterized by a strong interdependence of its members. It consists of a ‘myriad of interconnecting, interpenetrating organizations’ (Richardson and Jordan, 1979:74). A community, moreover, is reserved mainly to ‘legitimized clientele’ (Jordan, 1981: 105; see also Rhodes, 1985). Policy curtain The three patterns of policy making refer to a process of policy making in which the issue is not monopolized by state agencies but expanded to nongovernment actors. In the iron triangle the issue is expanded, albeit partially, to members of strong interest groups and legislative representatives; in the issue network the target of expansion are the experts and professionals; in the policy community expansion is aimed at ‘legitimized clientele’ whose participation is pertinent to the issue under deliberation. In all three constructs, intermediation between organized groups and political authorities takes place. The case may be, however, that no such intermediation is underway; nor is expansion evident. Not having achieved a placement on the political agenda, the issue is monopolized by power holders. Its deliberate removal from the agenda obstructs the policy process (Cobb, Ross and Ross, 1976). Demands put forward by non-governmental actors are not even rejected. They are simply not considered. Barring the entrance of the issue to the agenda may be viewed as a pre-policy stage where a ‘policy curtain’ inhibiting the entrance of an issue to the policy-making arena. The ‘curtain’ demonstrates the monopolization of the policy process by the elite, that is, the total exclusion of outside participation. It denotes a policy not to have a policy although a problem is evident. The presence of a curtain demonstrates that the elite enjoys a great zy zyxw zyxwv 95 measure of autonomy, that is, is independent of external actors and is able to ignore their demands. A policy curtain is thus the ‘negative’ of a policy community: the former is highly exclusive and lacks any form of dependence; the latter is highly inclusive and portrays a great measure of interdependence. Under circumstances of a policy curtain state agencies, or their extensions, refuse to grant rights of participation to outsiders despite their essential social role. In a democratic society decision-making within such a structure is bound to be ephemeral. It is unlikely for the state, despite its putative supremacy, to continuously ignore organizations on whose functional role it depends. The distinction between the four concepts - iron triangle, issue network, policy community and policy curtain - thus boils down to two major analytical concepts: exclusion and dependence. The operational definition of the variables under review is as follows. Exclusion refers to the refusal of the state to grant rights of participation to outside contenders. The question of access to institutions of policy-making is central in political analysis. The state is portrayed as normally preventing some organized interests from participating fully (Olsen, 1981). Consequently, organizations not participating in governmental policy making are absent because they are excluded by those having the power to deny them access and influence (on the distinction between participation and influence see Alford and Friedland, 1975). Hence the empirical evidence for exclusion is the absence of those likely to have an interest in the issue from its deliberation and shaping into public policy. Dependence is identified by an exchange taking place between participants in the policy process. The contents of this exchange cannot be measured accurately enough to permit precise calculations of net returns. Nor is it realistic to compare one form of dependence to another owing to the different scales employed by each participant. It is a common theme, however, that the state depends on outside groups for knowledge and expertise (Brenner, 1969), whereas interest organizations depend on the government for recognition and status (Offe, 1981). To its full extent dependence encompasses both state and intermediating organizations. Owing to the obvious distribution of power within society between government and the governed the major variable under concern is the state’s dependence on outside resources possessed by organized groups. Linking together the two variables distinguishing between the various forms of policy-making - exclusion and dependence -produces a four total classification. The four combinations of exclusion and dependence bring about the following situations: (a) A policy curtain prevails under circumstances of elite monopolization of the policy process, that is, when exclusion is high and interdependence is low. Parallel to a state of ‘no decision’ (Bachrach and Baratz, 1963) a 96 zyxwvuts zyxwvuts zyxwvu policy curtain denotes a state of ‘no policy’. This is the case when the issue is not placed on the agenda, when alternatives for implementation and action are not considered. (b) An iron triangle emerges when the curtain is partly lifted and exclusion is less. A policy process is under way. The issue has gained an agenda status and is deliberated and weighed. However, the state allows access only to its close allies. The policy making arena is characterized by a high degree of interdependence. (c) The combination of low exclusion and high dependence characterizes a policy community whose participants, both state and group representatives, are mutually linked by common bonds and interests. In a policy community organized interests are not only allowed access to forums of decision-making but are actually integrated into the policy process by virtue of control over a functionally important resource. (d) The fourth case presents the issue network portraying a low degree of both dependence and exclusion. The arena is open to a host of advocators whose autonomy based on individualistic orientations, precludes any measure of interdependence. (e) A fifth category of policy construct may evolve termed an iron duet. An iron duet is characterized by a combination of properties derived from three policy constructs. The formation of a ‘duet’ is triggered by problems requiring technical solutions based on knowledge and expertise (issue network); it is open to outside participants and based on mutual understanding and positive tradeoff (policy community). At the same time the duet tends to feature characteristics of an ‘iron’ structure: professional associations tend to forge intimate relationship with bureaucrats leading to the exclusion of other interest parties. The unchallenged alliance between the state and the experts produces the ‘iron duet’ typifying technocratic politics. The five configurations of policy constructs are apparently not linked to each other in any sequential manner. The policy under concern - entitlement of medical specialization ( a sub-section of health policy-making) - demonstrates, however, a certain movement from one construct to the other. It reveals that changes in the distribution of power between the state and outside contenders, caused by changing historical trends more than by deliberate efforts on the part of the actors, had an impact on the structure of the policy process. The remainder of the paper will describe the different phases of policy making. It will demonstrate how an issue, hidden by a policy curtain, was finally elaborated within an iron duet. The discussion will conclude with the purported reasons that enabled the transfer from one policy construct to the other. zyx zy zyxw 97 Entitlement of medical specialization in Israel The problem Medical specialization is an aspect of professional quality control. Although the process of acquiring a medical degree is long and arduous, the acquisition of expertise requires additional study and clinical practice. The wide diversification and expansion of medical knowledge has given an impetus to the expansion of specialization. The practice of entitling specialization varies across countries. In the US the medical profession itself is responsible for granting titles testifying to a physician’s professional qualifications (Starr, 1983). In the UK specialization is accredited by the General Medical Council, a quasi-governmental organization staffed by professionals. In Israel the state in cooperation with the medical association is responsible for the entitlement of specialization. Control of the specialization process indicates the degree of the state’s involvement in medical affairs. It reveals in a nutshell the state’s propensity to grant access to public organizations. The process serves as an adequate case study for the examination of mutual relationships among members of a policy making construct. zyxwv Policy curtain: high exclusion, low dependence Israel was founded in 1948, after 26 years of a British mandata over the area. In its inception the state faced two major problems: protecting its borders against the invasion of neighboring Arab states and absorbing mass immigration from Europe and Arab-speaking countries. Within three years of independence Israel doubled its population from some 600,000 to over one million. Immigration was warmly welcomed. In fact, Israel was described as a ‘country of immigrants, established for and run by immigrants’ (Ben Porath, 1986: 47). Israel’s raison d’Ctre is to ingather, retain, and forge into one nation Jewish immigrants from diverse countries and backgrounds. Newcomers have been welcomed not only because of ideological reasons but also because they were essential to Israel’s economic growth. The country’s rapid economic development within the first decade of its existence (an average annual 10 percent rise in national income) was attributed to its population increase, the main source of which was immigration (Halevi and Klinov-Malul, 1968). Immigration policy was highly pertinent to the question of medical manpower. Since its inception Israel has enjoyed one of the highest physician-topopulation ratios in the world. An ‘economic’attitude would have suggested pressures to limit the number of physicians. Friedman and Friedman (1980) have argued that medical associations should be numbered among the strongest trade unions, owing to their ability to restrict and control entrance to the profession. The Israel Medical Association (IMA) has done nothing to the sort. The ‘open door’ policy welcoming virtually all Jewish immigrants was applied also to physicians. Regardless of the number of immigrant physicians who entered the country and regardless of their background or qualifications, virtually all of them have been employed. Concomitantly the licensing procedure for immigrant physicians was very permissive. If completion of medical studies at one of the medical schools listed by the World Health Organization was demonstrated, a one-year license was granted. A t the end of the year a permanent license was awarded on the basis of positive evaluation of professional performance. Out of the thousands of physicians who immigrated to Israel over the years not a single one was denied the right to general practice (Shuval, 1985). Under these circumstances it is understandable that the entitlement of specialization could have put constraints on the policy of absorption. In the first decade of the state’s existence the issue of medical specialization was subject to ‘non-decision’. Mobilization of bias that blocked the road to the agenda was caused both by the distribution of resources and by the prevailing set of values. It should be remembered that many physicians who escaped the holocaust came to Israel without formal documents. Many others have resumed their studies only after the war was over. Their immigration may have interrupted their course of specialization. Eager to absorb immigrants and to encourage their arrival to Israel, the state minimized the hurdles placed before their professional careers and enabled each physician to practice medicine according to his or her preferences. The frame of a policy curtain was also influenced by the structure of health policy making in the early years of statehood. From its inception the Israeli state was a leviathan responsible for almost all domains of human needs - with one exception: medical care. The state provided housing, education, and even entertainment. It controlled the economy by a dence network of regulations and orders, incentives and prohibitions. Only health was left outside its responsibilities. The reason for this strange preclusion was grounded in the division of power in the health sector. The chief provider of health services was The General Sick Fund (Kupat Holim) affiliated with the Labor Federation (Histadrut). Like the British trade union movement, the Histadrut is an organization integrated with the Labor Party, which controlled Israel’s government for nearly 30 years. The government relieved itself from one of its major responsibilities, because the Sick Fund, serving as its political extension, monopolized the provision of health care. The Provision of health services became a major instrument for commanding resources and furnishing political support for the party in government. The politicization of health services is not a unique Israeli phenomenon (Altenstet- zyxw zy 99 ter and Bjorkman, 1981) but in Israel it has reached an unprecedented level. Since 1937 medical insurance has been an inseparable part of the regular union dues. A great part of the public who became members of the Labor Federation did so only in order to enjoy the health services it provided (Arian, 1981; Yishai, 1982). Health services were thus used as a means for political mobilization. Furthermore, the Sick Fund was also a reservoir for jobs. Its large work force could be counted on for political support. Close control of the Sick Fund’s staff has been maintained by political appointments to the Labor Party’s boards and governing institutions. The predominance of the Sick Fund in the health field and the vested interests of the leading coalition party - the Labor Party - in enhancing its power, made superfluous the transfer of health control to the state. In fact, in the first three years of statehood no ministry of health was established. When such a ministry was formed it was granted to a junior coalition partner lacking any viable political power and unable to intrude into the activities of the Sick Fund, which operated as a state within a state. Health issues were governed by a non-governmental institution (the Sick Fund) in whose interests the party controlling the state had a considerable political stake. Powerful as it was, the Sick Fund could not have issued regulations regarding medical specialization. It lacked the professional authority and expertise qualifying it to do so. The stage was thus set for a ‘policy curtain’ where policy is not forged and decisions are not reached. Although physicians did approach health authorities demanding regulation of specialization entitlement, the state failed to respond. Ideological imperatives coupled with political interests sustained the ‘curtain’ which prevented even the formal deliberation of the issue. zy zyxwv Iron triangle: exclusion and interdependence In the second phase policy making was carried mainly within the framework of an iron triangle consisting of the ministry of health, the representatives of the Sick Fund and the legislators. The Sick Fund was no longer a branch of state authorities but acquired some autonomy, evolving into a non-governmental actor, albeit intimately linked to the Health Ministry. The potential contribution of the professionals to the deliberation of the issue was totally ignored. Lifting up the curtain and moving toward an iron triangle was caused by two major developments: first, the growing powers of the state; second, the increased need to introduce legislation into the health policy domain. After the guns of the Independence War (1948-49) had fallen silent, and the mass immigration was absorbed, Israel established itself as having one of the strongest state machineries in the democratic world. The country was founded by people who were brought up in East European countries where a ‘statist’ 100 political culture prevailed and where statehood held pride of place. Their expectations were that a proper state must be a reified one, that is, one standing outside of, and above, its citizens and existing independently of them (Elazar, 1986: 186). Such a state was viewed as a major instrument for social change, and accordingly was expected to be comprehensive in its approach to its citizens, prepared to intervene in every aspect of life in order to bring about the necessary changes. The perennial state of belligerency contributed to the state’s powers, which were evident in all walks of life. Israel has had to devote a large and increasing share of its resources to defense. Its economic activities were conducted under the shadow of preparedness for overt or covert war. The government controlled the economy through subsidizing products and services and through transfer payments to various social sectors. Gradually the state’s authority was applied to the health scene too. The establishment of government-owned hospitals during the 1960s considerably increased the volume of state activity in the health sector. Demands for the regulation of specialization persisted. Self-regulation of the medical profession was ruled out because of the intrusive nature of the state. Authorities did consider specialization as an issue meriting regulation, but linked it to another problem, situated higher on the political agenda: health services in the country’s peripheral zones. The transfer of human and material resources from the center to the periphery has been one of the state’s promulgated (though much less applied) social policies. Israel’s coastline is densely populated, with some 70 percent of the population inhabiting the metropolitan areas and smaller cities. Its peripheral zones remained vulnerable from a defense perspective and have suffered continuous economic retardation. In the early 1950s Israel’s periphery was scattered with ‘development towns’ inhabited mainly by immigrants from Afro-Asian countries. Despite government assitance many of these sites remained miserable, portraying characteristic problems of spatial concentration of poverty. One of the paramount disadvantages of living on the periphery was the absence of advanced medical services, which were available largely in the major health centers located in metropolitan areas. The low standards of medical care were caused mainly by the physicians’ refusal to move to the periphery. Most of them preferred to practice in the big cities despite the ideological imperative of advancing the periphery. The incentives offered by health authorities to attract physicians to peripheral areas proved to be abortive. There was a dire need for legislative decrees to facilitate medical care in remote areas. Policy was formulated within an iron triangle from which the IMA was practically excluded. The Health Ministry played a dominant role. Representatives of the Sick Fund cooperated with their traditional partners in the ministry; members of Knesset provided the third side of the triangle. All 101 participants were loyal members of the Labor Party, which dominated the policy process. The medical profession was not partner to deliberations. Expertise did not grant physicians entrance to the policy arena. In May 1965 a Private Member’s Bill endorsed by the government, with the blessing of the Sick Fund, was submitted to the Knesset agenda, proposing that each newly licensed physician be compelled to serve in the pheriphery for a three year period. It was further proposed to regulate the entitlement of specialization but to deny it from any physician who had not served his or her term on the periphery. The state’s intentions have not materialized into legislation. The iron triangle failed to deliver policy owing to the adamant opposition of the physicians whose objection was demonstrated outside the policy arena. The iron triangle was sufficiently effective to initiate a policy change; it failed to turn this change into action. zyxwv Issue network: inclusion without dependence In the mid-1960s the IMA was granted access to forums of decision-making. Inclusion did not trigger dependence. The major cause of change was the organizational resources accumulated by the physicians’ association. At the disposal of the IMA were three major resources: members, monopoly of representation, and funds. The Israel Medical Association was founded in 1912 (then the Hebrew Medical Association). Although membership is not mandatory the association was joined by virtually all physicians in Israel. Incentives for joining were both material and non-material. The IMA provides selective services to its members such as insurance and legal advice. Professional identification is also a major reason for becoming a member (Yishai, 1990). The medical association monopolizes the representation of physicians, regardless of their nationality, specialization or occupational status. Organizational monopoly has been perceived as a major asset (Berlant, 1975; Wilsford, 1987) and a necessary condition for obtaining access to the decisionmaking process. Despite internal divisions between general practitioners (employed in community clinics) and hospital doctors, the IMA maintained its integrity and did not split into rival associations based on narrower professional criteria. The reason for the IMA’s comprehensive structure is mainly legal. A state decree recognizes the IMA as a ‘representative organization’ empowered to conduct wage negotiations. Defection from the medical association may thus lead to exclusion from the bargaining process. The IMA’s lavish funding by its members is the third source of organizational strength. Although the medical association is a voluntary body, dues are collected on a non-voluntary basis. The Israeli law provides for an ‘organizational tax’ paid by employees even if they are not full-fledged mem- 102 bers in their respective association. The automatic deduction was designed to solve problems of free riders taking advantage of the organization’s attainments without sharing in the costs. The effectiveness of this strategy is determined by the composition of the labor force. Self-employed workers are not affected by this legislation. The IMA has benefited from the fact that over 95 percent of the physicians in Israel are employed by public authorities. Deduction of organizational dues was thus widespread. The IMA was saved the trouble of prodding members and reminding them to fulfill their financial duties; the kitty was kept full owing to the state’s decree. The power accumulated by the IMA increased the chances of its inclusion in the policy process. In the period under review (the late 1960s) physicians had increasingly gained membership in forums set up by state authorities for deliberating alternatives regarding the entitlement of specialization. However, dependence between the organization of medical doctors and state agency failed to emerge, mainly because the state could still rely on its ‘own’ professionals, employed in the civil service. These were recruited on the basis of personal loyalties and political affiliations and were unlikely to challenge authorities. The fledgling discussions on the institutionalization of specialization practices did include physicians, but their impact on the formulation of policy was meager. Their expertise was sought, but not traded for influence. Professional bureaucrats, however, failed to deal effectively with the specialization issue. They may have enjoyed power within their ministry, but they lacked the authority necessary to impose regulations on the medical community. The inclusion of the IMA in the policy process as an equal partner was thus a product of a growing reliance on professional expertise and declining ability to suffice with internal sources of knowledge. zyxw Policy community: inclusion and dependence Since 1972 the Israeli Medical Association, through its Scientific Council, has been responsible for the administration and evaluation of formal written and oral qualifying examinations for medical specialization. The authority is acknowledged by the state and is inscribed in legislation. The IMA shares responsibility for accrediting physicians: it is a member of a policy community in charge of implementing regulations. The formation of this community was triggered by the exchange taking place between state organs and the IMA. Toward the late-1960s the issue of specialization could no longer be evaded, nor could it be handled by an iron triangle or a loose issue network. Technological advancement in medical care spurred demands for regulation of specialization entitlement. No measures could have been taken, however, without the active participation of the IMA. The contribution of individual physicians, 103 based on professional expertise alone, could not have grappled with the complexities of the problem. Rules had to be determined regarding the length, timing and location of training for specialization. Medical institutions and hospital departments competed for tutorial recognition, which carried prestige and material resources. Strong pressures were exerted by rank-and-file physicians, many of whom already had many years of practice and were not prepared to go through the arduous formal training period. Only powerful and authoritative physicians’ association could have withstood the pressure. The combination of prestige, power and professional expertise paved the way for the establishment of a policy community. The representatives of the medical association became not only legitimate members but essential partners. The IMA provided expertise for the introduction of specialization entitlement; by virtue of its comprehensive authority it was also responsible for implementing the rules. A process of exchange took place between health authorities and the medical association, in which both costs and rewards were traded. The IMA had to pay some organizational price for its membership in the policy community. In the first stages of institutionalizing specialization the IMA coped with difficulties and labored under distinct disadvantages. To begin with there was an excess demand for specialization entitlement which drained the IMA’s Scientific Council of its resources. The number of physicians trained for specialization grew from 1166 in 1975 to 3370 in 1987. This tremendous growth far exceeded the increase in the physicians population. There was also the problem of immigrant physicians who demanded acknowledgement of their specialty status in their country of origin. Many of them arrived at a middle or later stage of their career and shunned examinations which could serve as a major deterrent to advancement in the occupational system. The resulting pressures were inevitable. Physicians of Soviet origin resorted to court procedures when they were refused specialty status, charging the IMA with discriminatory practices. Despite these difficulties the IMA has been highly rewarded. The cooperation between state and association has enabled the latter to perform an essential function while enjoying the power and status conferred by the former. The IMA was granted recognition and resources. It had power over a domain to which its membership attached prime importance. The price it had to pay for these returns was meager: dissatisfaction of some members whose demands were not adequately met. Their recalcitrance did not pose a serious threat to the IMA’s cohesion. The government was also benefited by the exchange. To begin with, the state has been relieved of a function it did not want and could not have performed. It was unwilling to become entangled with the intricacies of the medical profession and was not ready to confront doctors who could not stand up to the Scientific Council’s stringent requirements. The IMA has acted as a genuine watch dog, carrying the banner of professional quality. Although, as 104 zyxwvuts noted above, some officials of the Health Ministry’s bureaucracy have had a medical training, the state’s administrative staff lacks the professional competence for composing and evaluating the examinations. Inclusion of the IMA in the policy process has been expedient also from a financial perspective. The physicians training for specializations have been required, by law, to pay an examination fee which covered only part of the expenses. The financial responsibility for conducting the specialization process was transferred to the IMA. Promotion of public interests may also be regarded as an indirect state benefit. The high standards set by the IMA’s Medical Council presumably have contributed to the quality of Israeli hospital medicine. The Council played a crucial role in regulating medical manpower. It is a fundamental national imperative to train family doctors in order to improve the quality of primary care, serving as the front line in health services. Owing to the principle of ‘linkage’ characterizing the wage system the state was unable to offer financial incentives to family doctors, fearing that an avalanche of wage demands would shake Israeli’s tight budget. State authorities had to rely on the IMA to attract medical manpower to the less glorious and less prestigious position in family health care. The Scientific Council addressed the problem of primary medical care by developing specialization programs for general practitioners. Although there is no clear evidence to indicate if its efforts in this regard have proven effective or alternatively if market forces have played their role, the proportion of physicians registered for training in family medicine has constantly been growing. The price paid by the state for these benefits was not high in comparison with its gains. All it did was to allocate authority to an external body and incorporate it, both legally and functionally, in the policy process. zyxwvuts From a policy community to an iron duet Within the community responsible for entitlement of specialization the ‘rules of the game’ governing behavior were similar to those practised in British local government, and for that matter, in the industrial sector. Wright described these rules in what he termed as a ‘policy network’ (1988: 609). One important rule of the game is mutuality. Members of networks accept and expect that mutual advantages and benefits will result from their participation in the network. Policy issues are handled within the policy network, on the basis of trust and a respect for confidence. Undertakings given by members of the network of commitments entered into informally are customarily honored. . .there is no only a willingness to consult informally, but an expectation of zyxwvuts 105 zy consultation. On the basis of that respect for confidence, members of a network are prepared to exchange ideas or discuss future policies or make available to each other information which would be regarded as sensitive or even damaging in the public domain. This is an accurate description of the health policy community focusing on specialization of physicians. Benefits are mutual since both state authorities and the medical profession stand to gain from the practice. Trust and respect are forged on the basis of shared interests of physicians employed in the Ministry of Health and those occupying positions of power within their own association. In the approximately two decades since the introduction of the first specialization legislation not once have the parties to the community been subject to internal controversy or rivalry over resources. The policy community proved to be a perfect instrument for both formulating and implementing policies favored by its members, linked together by bonds of mutual dependence. There was, however, a fly in the ointment of the policy community, which set the stage for an iron duet. Access was limited and inclusion restricted. Membership was granted mainly to the leaders of the professional community rather than to rank-and-file physicians or to representatives of other health-related professionals. Oligarchic overtones were clearly visible. Legislatures were also excluded from the process of policy-making, even though considerations of national importance did surface. Regulation of specialization could have determined priorities within the medical market and did have an impact on the quality of medicine in the country. The preference given to general practice did indeed comply with social needs, but many other problems (such as the chronic shortage of anesthetics and gerontologists) remained unattended. The clients, having direct stake in the quality and availability of health services were, obviously, also excluded from the policy ‘community’. The only voice heard was that of the professionals, who, together with their counterparts in the ministry acted within a framework of a ‘policy duet’. The policy community proved very effective in terms of interest group-state relationship, but ‘professional power’ (Freidson, 1986) prevented the development of an open, widely penetrated policy structure, amenable to outside inputs. The iron duet adopted many characteristics of an iron triangle: exchange of information is done behind closed doors, hidden from the public eye; relationship between members are intimate and stable. Professional jargon adds another layer to the atmosphere of confidence and removes the issue from public scrutiny. The iron duet is largely insulated from its environment by clear-cut boundaries; its members possess monopolistic power by virtue of their expertise and knowledge. The administrators taking part in the policy community are not ‘experts using experts’ but experts joining experts to zyxwvu zyxwvut shape public policy, which may or may not be congruent with the welfare of the public at large. The iron duet proves to be an instrument of efficiency, not of equity. Concluding remarks: Why do specific policy constructs emerge? The foregoing discussion has attempted to tackle, through the analysis of specialization entitlement in Israel, structures and processes of decision-making in which both the state and an intermediating organization have a high stake. The issue under review was important in a state committed to the provision of health services; it was no less important to the country’s physicians, whose professional interests were at stake. The following conclusions emerge: - The term policy curtain was discussed in the literature dealing with the hidden agenda that is, with (non) policy-making. Such a curtain exists when (a) a set of values prevents the rise of the issue on the political agenda, and (b) when power relations in society do not favor its elaboration. The specialization issue was not granted agenda status because it was incompatible with absorption policy, aiming at providing smooth entrance to immigrant physicians. Empowerment of a non-state agency (the Sick Fund) also precluded the open deliberation of specialization by stage authorities. - An iron triangle is likely to develop under two complementary situations: first, a problem gains critical importance and forces itself into the political agenda; second, the state commands sufficient political and organizational resources to enable uncontested domination over the selection of partners. An iron triangle emerged when the question of health services in the periphery gained salience. The three sides of the triangle branched from the same tree: a powerful political party in control of all centers of power. - The failure of an iron triangle to deal effectively with a policy issue may lead to the emergence of an issue network, where professionals play a leading role. Civil service professionals greatly contribute to melting the iron. The roles of professionals in state bureaucracies (Mosher, 1978), especially the medical profession (Haywood and Hunter, 1982), are of particular importance. But not less crucial to the development of an issue network is the accumulation of power by an organized professional association. - The shift from an issue network to a policy community is encouraged by two contrasting processes: (a) an increase in the scope of state policy making which leads to centralization, and (b) the need to rely on actors outside the realm of the state, which leads to decentralization of power. A positive balance of payoffs between state and associations is a necessary condition 107 for the emergence of a policy community. As the paper amply demonstrated these circumstances were evident in the case under concern. - Finally, the policy community is not necessarily the end of the ‘policy road’. The shift from the ‘community’ to the iron duet came about because professionalism was a dominant factor. The reliance on expertise generated exclusion and closure. It precluded from the process of policy making public representatives, including legislatures. The ‘professionalization’ of the issue thus threatens to turn a policy community, assumed to be founded on principles of public exposure, openness and equity, to an iron duet. The iron duet exposes contradictions embedded in the policy process. It emanates from a policy community allowing for inclusion of ‘outsiders’; at the same time it leads to exclusion which invalidates the essence of the community. It is the lack of alternative for both state and experts that casts their mutual relations in iron rather than in soft rubber. The case of specialization entitlement bring forth a conclusion regarding the policy process. First, as the case under consideration reveals, changes in the political environment (that is, social demands, actors’ resources and/or values) does bear influence on the pattern of policy making. Political culture may have a long-term bearing on a ‘policy style’ (Richardson et al., 1982) but short-term fluctuations are nevertheless considerable. Analytical concepts delineating the characteristics of policy formation should thus be used with prudence, since the movement from one policy construct to the other is more frequent than believed. In fact, as the case of specialization entitlement in Israel has demonstrated, it may occur within the narrow confines of a particular policy issue. Furthermore, the clear distinctions between the policy constructs may be useful for analytic purposes. In reality, however, as the case under consideration has demonstrated, one construct may breed another, even though the two show considerable dissimilarities. zy zyxw zyx zyxwvut References Alford, R.R. & R. Friedland (1975). ‘Political participation and public opinion’. Annual Review of Sociology 1: 429-79. Altenstetter, A.C. & J.W. Bjorkman (1981). ‘Planning and implementation: A comparative perspective on health policy’. International Political Science Review 2: 11-42. Arian, A . (1981). ‘Health care in Israel: Political and administrative aspects’. International Political Science Review 2: 43-56. Bachrach P. & M. Baratz (1963). ‘Decisionsand nondecisions: An analytical framework’.American Political Science Review 57: 632-42. Ben-Porath,Y.(1986). ‘Patternsand peculiarities of economic growth and structure’, TheJerusalem Quarterly 38: 43-63. 108 zyxwvuts zyxwvutsr zyxwvuts zyxwvu zyx Berlant, J.L. (1975). Professions and Monopoly. Berkeley: University of California Press. Brenner, M.J. (1969). ‘Functional representation and interest group theory’. Comparative Politics 2: 111-34. Cobb, R.W., J.K. Ross & M.H. Ross (1976). ‘Agenda building as comparative political process’. American Political Science Review 70: 126-138. Dye, T.R. & H. Ziegler (1970). The Irony of Democracy. A n Uncommon Introduction to American Politics. Belmont, Ca.: Wedsworth. Elazar, D . (1986). Israel. Building a New Society. Bloomington: Indiana University Press. Freidson, E. (1986). Professional Powers. A Study of the Institutionalization of Formal Knowledge. Chicago, Ill.: University of Chicago Press. Friedman, M. & R . (1980). Free to Choose. A Personal Statement. 2nd ed. New York: Harcourt, Brace, Jovanovich. Halevi, N. & R. Klinov-Malul (1968). The Economic Development of Israel as a Case Study. Jerusalem and New York: Praeger with the Bank of Israel. Hayward, S . & A. Hunter (1982). ‘Consultative processes in health policy in the United Kingdom: A view from the centre’. Public Administrution 69: 143-62. Heclo, H. (1978). ‘Issue networks and the executive establishment’, in A. King (ed.), The New American Political System. Washington, D.C.: American Enterprice Institute Jordan, A.G. (1981). ‘Iron triangles, woolly corporatism, or elastic nets: images of the policy process’. Journal of Public Policy, 1: 95-124. Lowi, T.R. (1979).The End of Liberalism. 2nd. edn. New York: Norton. Mosher, F.C. (1978). ‘Professions in Public Service’. Public Administration Review 38: 144-150. Offe, C.(1981). ‘The Attribution of public status to interest groups: Observations on the WestGerman case’, in: S. Berger (ed.), Organizing Interests in Western Europe. Pluralism, Corporatism and the Transformation of Politics. Cambridge: Cambridge University Press. Olsen, J.P. (1981). ‘Integrated organizational participation in government’. in: P.C. Nystrom and W.H. Starbuck (eds.), Handbook of Organizational Design. Oxford: Oxford University Press. Rhodes, R.A.W. (1985). ‘Power-dependence, policy communities and inter-governmental networks’. Public Administration Bulletin 49: 4-31. Richardson, J.J. & A.G. Jordan (1979). Governing Under Pressure: The Policy Process in a Post-Parliamentary Democracy. Oxford: Martin Robertson. Richardson, J.J., G. Gustafsson, G. Jordan (1982). ‘The concept of policy style’, in: J. Richardson (ed.), Policy Styles in Western Europe. London: Allen and Unwin. Schattschneider, E.E. (1960). The Semi-Sovereign People. New York: Holt, Rinehart and Winston. Shuval, J.T. (1985). ‘Social functions of medical licensing: A case study of Soviet immigrant physicians in Israel’. Social Science and Medicine 20: 901-909. Starr, P.(1983). The Social Transformation of American Medicine. New York: Basic Books. State Comptroller (1987). Annual Report 1987. Jerusalem: The Government Printer (Hebrew). Walker, J. (1989). ‘Policy communities as Global Phenomena: Governance 2(1): 1-4. Wilsford, D. (1987). ‘The cohesion and fragmentation of organized medicine in France and the United States’. Journal of Health Politics, Policy and Law. 12: 481-503. Wolfe, A. (1977). The Limits of legitimacy. New York: Free Press. Wright, M. (1988). ‘Policy community, policy network and comparative industrial policies’. Political Studies 36: 593-612. Yishai, Y. (1982). ‘Politics and medicine: The case of Israeli national health insurance’. Social Science and Medicine 16: 285-91. Yishai, Y. (1990). The Power of Expertise. Israel Medical Association. Jerusalem: Jerusalem Institute for the Study of Israel (Hebrew).