88600 A POLITICAL ECONOMY ANALYSIS OF TURKEY'S HEALTH TRANSFORMATION PROGRAM DISCUSSION PAPER JULY 2014 Jesse B. Bump Susan Powers Sparkes A POLITICAL ECONOMY ANALYSIS OF TURKEY’S HEALTH TRANSFORMATION PROGRAM Jesse B. Bump and Susan Powers Sparkes July 2014 Health, Nutrition and Population (HNP) Discussion Paper This series is produced by the Health, Nutrition, and Population Global Practice. The papers in this series aim to provide a vehicle for publishing preliminary results on HNP topics to encourage discussion and debate. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent. Citation and the use of material presented in this series should take into account this provisional character. For information regarding the HNP Discussion Paper Series, please contact the Editor, Martin Lutalo at mlutalo@worldbank.org or Erika Yanick at eyanick@worldbank.org. © 2014 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 All rights reserved. ii Health, Nutrition and Population (HNP) Discussion Paper A POLITICAL ECONOMY ANALYSIS OF TURKEY’S HEALTH TRANSFORMATION PROGRAM: Jesse B. Bump a and Susan Powers Sparkes b a Department of International Health, Georgetown University b Department of Global Health and Population, Harvard School of Public Health Abstract: Beginning in 2003, Turkey initiated a series of reforms under the Health Transformation Program (HTP) that over the past decade have reshaped the health system. Understanding the political economy of this process is important for the future of Universal Health Coverage (UHC) in Turkey, and also for many other countries and the development agencies that assist them. This report analyzes the historical context and complex political economy challenges of the reform. Our findings are based on stakeholder interviews and a review of literature. First, we identified five contextual factors that were important in bringing health reform to the policy agenda in Turkey, and were helpful in sustaining the reform during adoption and implementation: (1) a long history of reform plans and attempts; (2) fiscal pressure to reform the social sectors; (3) public support for health reform; (4) strong economic growth; and (5) favorable demographic conditions. Second, we assessed four political economy challenges central to the reform and the strategies used by the Ministry of Health (MoH) to overcome them. First, the MoH built public support for reform among the broad base of beneficiaries by focusing on highly visible and fast changes. Second, the MoH overcame well-organized interest group opposition to the reforms by splintering their support or delegitimizing their views. Third, Turkey asserted its own domestic priorities over those of the IMF and World Bank in cases of direct conflict. Fourth, the MoH circumvented potential political and institutional opposition to the large expansion of benefits and coverage through a carefully sequenced adoption and implementation plan that could be executed mostly without requiring the support of other ministries. This analysis also highlights important trade-offs made by the MoH with respect to the redistribution of resources, quality of care, financial sustainability, and physician satisfaction, which will all have to be considered as Turkey enters its next phase of health system development. Keywords: Universal Health Coverage, health systems, Turkey, political economy, health reform Disclaimer: The findings, interpretations and conclusions expressed in the paper are entirely those of the authors, and do not represent the views of the World Bank, its Executive Directors, or the countries they represent. Correspondence Details: Jesse B. Bump, Department of International Health, Georgetown University, 37th and O Streets, Washington DC 20057. jbb77@georgetown.edu Susan Powers Sparkes, Department of Global Health and Population, Harvard School of Public Health, 665 Huntington Ave, Boston MA 02115. sps187@mail.harvard.edu iii Table of Contents EXECUTIVE SUMMARY ................................................................................................................. 5 ACKNOWLEDGMENTS ................................................................................................................. 7 PART I – INTRODUCTION ............................................................................................................. 8 PART II – CONTEXTUAL OVERVIEW ......................................................................................... 11 A History of Health Reform .................................................................................................... 11 Mounting Financial Pressure to Reform the Social Sectors .................................................. 11 High Public Support for Reform ............................................................................................. 12 Increased Fiscal Space for Health ......................................................................................... 12 Favorable Demographics and Low Expectations .................................................................. 13 PART III – POLITICAL ECONOMY CHALLENGES IN THE TURKISH HTP (2003-2012) ......... 14 Challenge #1: Engaging Beneficiaries to Build Support for Reforms .................................... 14 Challenge #2: Managing the Influence of Opposing Interest Groups ................................... 16 Challenge #3: The IMF, the World Bank, and Domestic Politics ........................................... 19 Challenge #4: Covering the Poor and Unifying Benefit Systems .......................................... 20 PART IV – DISCUSSION .............................................................................................................. 24 PART V – CONCLUSIONS ........................................................................................................... 26 PART VI – IMPLICATIONS ........................................................................................................... 27 REFERENCES .............................................................................................................................. 29 ANNEX I: STAKEHOLDER MAPS ............................................................................................... 32 ANNEX II: INTERVIEW GUIDE .................................................................................................... 43 ANNEX III: STATISTICAL ANNEX ............................................................................................... 44 4 EXECUTIVE SUMMARY Beginning in 2003, Turkey initiated a series of reforms under the Health Transformation Program (HTP) that over the past decade have reshaped the health system. Understanding the political economy of this process is important for the future of Universal Health Coverage (UHC) in Turkey, and also for many other countries and the development agencies that assist them. This report analyzes how the senior leadership of the Ministry of Health (MoH) was able to navigate complex political economy challenges that were central to the HTP’s adoption and implementation, as opposed to previous work that has emphasized what happened in the reform. Our findings are based on a review of published and unpublished literature and stakeholder interviews. We identified five contextual factors that were important in bringing health reform to the policy agenda in Turkey, and were helpful in sustaining the reform during adoption and implementation. First, in 2003 the MoH was able to begin reforms quickly because some components had already been developed, dating back at least to World Bank-supported planning in the early 1990s. Second, very high government spending on the social sectors was an important factor underlying the economic crises of 1999 and 2001, which created economic and political pressure to reform the health and pension systems. Third, the 2002 parliamentary elections delivered a legislative majority for the AK Party (Justice and Development Party), ending decades of coalition governance in Turkey. This majority was important because it limited the ability of other parties and special interest groups to block the reform process. Fourth, strong economic growth during adoption and implementation of the HTP increased the fiscal space available for health without imposing cuts elsewhere. Fifth, Turkey had a relatively young population with low expectations of the health sector. This meant Turkish citizens demanded fewer and lower cost interventions than would be needed for an older population, and their positive impressions of the reform were easier to create than would have been the case in a population with higher expectations. We then assessed four political economy challenges central to the reform and the strategies used by the Turkish MoH to overcome them while adopting and implementing the HTP between 2003 and 2012. First, the basic political economy of reform favors opponents because the costs tend to be concentrated on elite, well-organized groups and potential beneficiaries, although more numerous, tend to have few economic or political resources. The MoH overcame this challenge by quickly building support among the HTP’s broad base of beneficiaries, whose greater numbers translated into electoral power once engaged. To do so, the MoH concentrated on highly visible, fast reforms, such as refurbishing waiting rooms, ending unpopular existing policies, and greatly expanding the ambulance network. These initial moves built popular support for more difficult reforms to come, including changing provider payment systems, closing underperforming facilities, and merging social security systems. Second, interest groups tend to exercise strong influence on reforms because they are well-organized and seek to protect their interests. Early in the HTP, the Minister of Health and his senior leadership team developed specific strategies to persuade or overcome opposition groups. In some cases, the reform plans were adapted to accommodate concerns. In other cases, the MoH and AK Party leadership worked to neutralize the influence of opponents by splintering their support or delegitimizing their views. Third, by choice or necessity, often countries engage the IMF and World Bank in policy reforms, which can produce tensions between desiring resources from the institutions and wanting to remain in full control of the domestic policy agenda. Despite deep ties to the two institutions, Turkey asserted its own priorities, including in at least two cases of direct conflict by ensuring agreement among all ministries and by asserting national sovereignty in negotiations. Fourth, expanding coverage and integrating benefit packages represents an enormous challenge of political economy because the redistribution required to cover low-income groups implies a potential reduction in formal sector workers’ benefits and/or the use of some of their contributions for others. To overcome this problem, the MoH worked around obstacles to postpone or avoid potential opposition. For instance, the MoH used the existing 5 Green Card Program 1 as its primary vehicle for scaling up coverage for low-income households in part because modifying a program did not require parliamentary approval. The MoH consolidated control over the process by moving the Green Card program under their auspices and stimulated demand by expanding the benefits package, increasing the number of Green Cards in circulation, and making concurrent supply side improvements. We conclude with a brief discussion of the implications our analysis holds for Turkey’s current and future health agenda, and for other countries seeking to learn from this example. 1 The Green Card Program for the Poor was launched in 1992 to expand access to health services for the poor and underserved. The Yesil Kart, Green Card, was a non-contributory health insurance scheme for the poor. Under the HTP this program was one of the four consolidated under one social insurance program which signified a substantial increase in financial protection for the poor. 6 ACKNOWLEDGMENTS This report was produced at the request of the Turkish Ministry of Health under a consulting agreement with the World Bank. The authors are grateful to the World Bank for publishing this report as an HNP Discussion Paper. We gratefully acknowledge the support of the Japan Trust Fund, Task Team Leaders Rekha Menon and Claudia Rokx, and the World Bank-Japan Partnership Program on Universal Health Coverage. Also at the World Bank in Washington we thank Norosoa Andrianaivo and Sreypov Tep. At the World Bank office in Ankara we thank Ahmet Levent Yener and Elif Yonca Yükseker. At the Ministry of Health we thank Banu Ayar, Volkan Çetinkaya, and Uğur Dilmen. We are particularly grateful to Drs. Ayar and Çetinkaya for arranging and attending our interviews, facilitating our data requests, and reviewing our report. At the Harvard School of Public Health we thank Michael Reich and Marc Roberts. We extend special thanks to our interviewees. 7 PART I – INTRODUCTION 1. Beginning in 2003, Turkey initiated a series of reforms under the Health Transformation Program (HTP) that has few—if any—parallels in scope and speed. Understanding the political economy of this process is important for the future of Universal Health Coverage (UHC) in Turkey, and in light of the great interest in UHC schemes globally, is important for many other countries and the development agencies that assist them.2 2. Before the reforms, Turkey’s aggregate health indicators lagged behind those of OECD member states and other middle-income countries. Less than 70 percent of the population was insured and even those with insurance did not have adequate access to timely health services (Akdağ 2011). The health financing system was fragmented, with four separate insurance schemes and a “Green Card” program for the poor, eac h with distinct benefit packages and access rules. Both the Ministry of Labour and Social Security (MoLSS) and Ministry of Health (MoH) were providers and financiers of the health system, and four different ministries were directly involved in the public health care delivery system. 3. Turkey’s reform efforts were designed to rectify these problems and virtually all aspects of the country’s health system were affected, although the outcomes attributable to these changes are under debate. A recent high-profile article by Atun et al (2013) reports many positive effects of the HTP, roughly in line with previous work by many of the same authors. They find, for example, (i) insurance coverage increased from 64 to 98 percent between 2002 and 2012; (ii) the share of pregnant women having four antenatal care visits increased from 54 to 82 percent between 2003 and 2010; and citizen satisfaction with health services increased from 39.5 to 75.9 percent between 2003 and 2011 (see Annex III for additional statistics) ( Akdağ 2011; Atun, Aydın et al. 2013). However, many points made by Atun et al. attracted critical responses. 3 4. In part, these different views reflect the complicated political economy of health reform and wider disagreements about the future of Turkey. In this report we focus on four political economy problems central to health reform and analyze the approaches used by the Minister of Health and his senior leadership team to overcome them. Our report differs from the recent contribution of Atun et al (2013) on the same reform. Atun et al. examine what happened in the reform; they place it in historical context, concentrate on an analysis of its effect on many indicators, and extract lessons concerning what facilitating factors were most important. By contrast, we do not analyze the effect of the reform on any health indicators or outcomes. Instead, we analyze how the senior leadership of the MoH was able to navigate complex political economy challenges that were central to the reform’s adoption and implementation. The difficult and contentious elements of this process underlie some of the differences that remain between the account of Atun and colleagues and other observers. Our focus is on assessing these challenges in political economy terms to explicitly recognize the importance of both factors in shaping the distribution of resources for health. We discuss the basic dynamics of why these political economy challenges arise in health reform and then present our analyses of how they were overcome in the Turkish reform. Atun et al (2013) employ political concepts in their analysis, particularly political stability and political commitment. We view these as important, but they are not by themselves sufficient to overcome obstacles to 2 In recent years, policymakers in many low and middle-income countries (LMICs) have taken steps toward universal health coverage (UHC), including Brazil, China, Mexico, Rwanda, Thailand, and Vietnam. Pannarunothai, Patmasiriwat, et al. 2004. Ekman, Liem, et al. 2008. Hu 2008. Logie, Rowson, et al. 2008. Damrongplasit, and Melnick 2009. Frenk, Gómez-Dantés, et al. 2009. Paim, Travassos, et al. 2011. Tangcharoensathien, Patcharanarumol, et al. 2011. 3 Fourteen authors in ten letters questioned the reliability and interpretation of the data, arguing that the quality of Turkish data has declined (Aksakoglu 2014), that inequality has increased instead of decreased (Hamzaoglu 2014), that other data sources show a less favorable view of the HTP (Pala 2014), that physician satisfaction has deteriorated under the HTP (Tanik 2014), that the HTP has had the effect of privatizing health care (Civaner 2014, Kilic 2014), that the Atun et al. report systematically overlooked shortcomings of the HTP (Aktan, Pala et al. 2014), among other issues. 8 reform, as seen in many unsuccessful attempts in many countries, including the Clinton health reform attempt in the United States. 5. Following this introduction we discuss our analytic methods and their limitations. We begin our analysis with an assessment of five contextual factors that were favorable to the reform, including Turkey’s growing economy and young population. These factors provided a window of opportunity, of which the MoH and the AK Party took full advantage to adopt and implement reforms under the HTP. We then discuss four political-economy challenges that were pivotal to the trajectory of the reform. These four challenges were identified though a review of the literature on health reform and an analysis of our interviews with stakeholders involved with Turkey’s reform. We present them in roughly chronological order, although the sequences we describe overlap substantially. These challenges are: (i) engaging beneficiaries to gain support for reform; (ii) managing the influence of opposing groups; (iii) managing the influence of the IMF and World Bank on domestic politics; and covering the poor and unifying benefit packages. We provide an analysis of the strategies employed by the Turkish MoH to overcome each of these challenges in the adoption and implementation of the HTP between 2003 and 2012. We then discuss some of the implications of the chosen strategies for the future of the Turkish health system and for countries studying this example. 6. Neither the evidence we gathered nor our method of analysis is intended to support a normative or ethical analysis of the policy design. All design choices have consequences and where we clarify the effects of those choices it is to explain what happened and why; this report does not evaluate whether policy choices were “good” or “bad” b y any external standard. Health reform is a fundamentally political process because it affects the distribution of resources, rights, and responsibilities. It is also an economic phenomenon because health spending is typically several percent of GDP and reform influences when and to whom this money flows. Stakeholder Analysis 7. For this investigation we employ the political economy technique of stakeholder analysis, which is a structured method for assessing the “behavior, intentions, interrelations, agendas , interests, and the influence or resources” of relevant actors concerning a particular policy or issue (Brugha and Varvasovszky 2000). An assessment of these factors can be used to map supporters and opponents and inform strategies for increasing the likelihood of success, for instance by identifying possible coalitions of supporters or opportunities to diminish the commitment of opponents. Stakeholder analysis typically involves specifying the players relevant to an issue — those engaged and those potentially engaged—evaluating their stance on the issue, and forming judgments about their relative power to reach a calculus of political feasibility and develop strategies toward the desired aim (Reich 1995; Varvasovszky and Brugha 2000; Roberts, Hsiao et al. 2008). 8. Stakeholder analysis has been applied by many global health scholars to understand the politics of policy design and implementation, particularly in recent years. It provides a cross- sectional picture of actors, positions, and power at particular point in time. It can be used as a forward-looking exercise to guide planned or prospective activities, including health reforms for universal coverage (Gilson, Erasmus et al. 2012), health insurance premium change in Ghana, immunization campaigns in the same country, alcohol control policy in a Russian region, or Ugandan maternal and child health program, for instance (Gil, Polikina et al. 2010; Sarr 2010; Gilson, Erasmus et al. 2012; Abiiro and McIntyre 2013; Namazzi, Kiwanuka et al. 2013). Also, stakeholder analysis can be used retrospectively as a tool of historical analysis to understand and explain the processes that govern success and failure in policymaking, as Akinci et al (2012) have done with health reform in Turkey, and Bump and colleagues have done with diarrheal disease and the global health agenda (Akinci, Mollahaliloglu et al. 2012; Bump, Reich et al. 2012). In this study, we employ stakeholder analysis to assess the position, role, objectives and strategies of relevant actors concerning the design and adoption of the HTP between 2002 and 2004 (stakeholder tables included in Annex I). We also conducted a separate stakeholder analysis on the expansion of the Green Card Program between 2003 and 2012, as it was identified as the key vehicle used to achieve UHC in Turkey (Green Card Program expansion stakeholder tables included in Annex I). 9 Data Collection and Analysis 9. We chose stakeholder analysis as our primary method for assessing the political economy of Turkey’s HTP because of its strengths in str ucturing and clarifying the complex politics of health reform and the welter of important contextual factors that shape the interaction of payers. We hypothesized that important stakeholders in the Turkish health reform would be roughly similar to the stakeholder groups important in other health reforms. Based on a literature review and experiences in other countries, we constructed a preliminary list of these stakeholders, which we then refined according to published articles on Turkey and the views of our counterparts at the General Directorate of Health Research of the MoH in Turkey, and at the World Bank’s office in Ankara and at its headquarters in Washington, DC. The individuals and institutions identified through this process were then approached by our counterparts to the MoH and asked to participate in key informant interviews. To help ensure consistency and completeness, we developed a semi- structured interview guide (attached as Annex II). We used the guide along with our stakeholder table in our interviews with informants. As we conducted our interviews and improved our understanding of the relevant actors, we adjusted the stakeholder table. In about half the interviews—particularly these conducted later in the process—we shared a blank copy of table with interviewees. Senior officials and academics agreed that the group of important actors represented on our table is accurate and reasonably complete. 10. We conducted our key informant interviews in Ankara and Istanbul, Turkey in late March and early April of 2013. Interviews were conducted in English, in a mix of Turkish and English, and in Turkish with professional interpretation, as dictated by circumstances. Each interviewee was informed of the purpose of the study, our intention to take detailed notes of each interview, and our process for handling interview data. Permission was requested to take notes and to report quotes attributed to a general affiliation. We shared a draft report for comment with all interviewees, the World Bank, and the Government of Turkey. We revised the report to reflect comments and sensitivities. Detailed notes were taken during and immediately after each interview. 11. Those interviewed included current and former government officials in the MoH, the Undersecretariat of Treasury, the MoLSS, the Ministry of Development Planning (formerly State Planning Organization), the Ministry of Family and Social Policy, the Turkish Statistical Agency, and members of Parliament. We also interviewed representatives of non-governmental interest groups, including provider groups such as the Turkish Medical Association, the Turkish Midwifery Association, the Turkish Nurses Association, hospital administrators, university faculty, and university administrators. In May of 2013 in Washington, DC we interviewed current and former World Bank officials who had been involved with the reforms. 12. Our use of stakeholder analysis to investigate the political economy of health reform in Turkey allows us to identify which actors were influential in the reform, to analyze their positions, and to explore their motivations and strategies. This approach is valuable for explicating the broad political economy challenges behind the observed outcomes. The use of interviews gives us access to stories, personalities, and details unlikely to be captured in written sources. These advantages let us construct a realistic picture of how the reform was designed and provides some insight into implementation strategies. Limitations 13. Stakeholder analysis based on published articles and reports allows for the synthesis and evaluation of a great volume of qualitative information, but it achieves this scope at the expense of the very fine detail, nuance, and the contextual richness typical of historical accounts based on archival materials such as correspondence, confidential memos, verbatim transcripts, or other internal documents. Among the most important limitations of our methods is the choice of interviewees. In a process as large as health reform there are tens of millions of affected parties and our method of selecting interviewees could not possibly capture all perspectives. Undoubtedly, we did not interview all of the most important players, nor did we cover all details in the interviews we conducted. Our account is also limited by recall bias because it concerns events well in the past. Furthermore, our interviews were facilitated by the MoH and included only central-level stakeholders and interest groups, both of which would influence the perspectives we obtained. 10 PART II – CONTEXTUAL OVERVIEW 14. Health reform is a complex process that is influenced by contextual factors, including history, politics, economics, and the characteristics of the population to be served by the health system. These factors were important in bringing health reform to the policy agenda in Turkey, and were helpful in sustaining the reform during adoption and implementation. From the accounts of our interviewees and the written evidence we reviewed, we identified five contextual factors that were favorable to the political and economic viability of reform. First, in 2003 the MoH was able to begin reforms quickly because many components of the reform could be based on plans developed as part of earlier efforts dating to the early 1990s. Second, very high government spending on the social sectors was an important factor underlying the economic crises of 1999 and 2001, which created economic and political pressure to reform the health and pension systems for the Ministerial agencies in charge of financial and economic affairs. Third, the 2002 parliamentary elections delivered a legislative majority for the AK Party, ending decades of coalition governance. This majority was important because it limited the ability of other parties and special interest groups to block the process. Fourth, strong economic growth during adoption and implementation of the HTP increased the fiscal space for health without imposing cuts elsewhere. Fifth, Turkey had a relatively young population with low expectations of the health sector. This meant the primary demand was for fewer and lower cost interventions than would be needed for an older population, and that positive impressions of the reform were easier to create than would have been the case in a population with high expectations. These five factors were all favorable between 2003 and 2012, when the reforms took place. A History of Health Reform 15. The AK Party acted quickly on health reform to capitalize on the popular support it enjoyed following the 2002 elections. The MoH reduced the time needed to develop its policies by drawing largely from health reform plans that had been carefully devised and analyzed throughout the 1990s. Previously, various coalition and military governments, working under World Bank loan agreements, had developed reform plans, but had been unable to gain the broad support required to adopt and implement them (Tatar, Mollahaliloglu et al. 2011). Beginning with the 1990 Health Sector Plan, the MoH and the State Planning Organization had proposed a health system model based on; (i) purchaser-provider split, (ii) universal health insurance, (iii) a rational policy for human resources and payment on the basis of performance, and (iv) the establishment of a family practitioner model (Tatar, Mollahaliloglu et al. 2011; Yasar 2011). These reform proposals had not been fully adopted under the coalition governments. Upon taking power in 2003, Minister Akdağ and his reform team adopted them for their own use. This reform team was in place from 2003 to 2013, when Minister Akdağ stepped down as Minister of Health (Atun, Aydın et al. 2013). This team of trusted colleagues worked closely with the Minister on all aspects of the design, adoption and implementation of the reform. According to interviews with current and former senior officials closely involved with the process, the reform team drew on the technical expertise of those who had devised these reforms and in parallel developed a politically viable communications strategy to present and promote the HTP. Mounting Financial Pressure to Reform the Social Sectors 16. The financial crises of 1994 and 2000–1 were caused in part by high government spending on the social sectors, but before the AK Party came into power none of the coalition governments had been able to effect reform. The crises punctuated a decade of economic volatility, high and increasing public sector borrowing requirements, high interest rates, and increasing public sector deficits (Ertugrul and Selcuk 2001; International Monetary Fund 2002; Tatar, Mollahaliloglu et al. 2011). The weak economy undermined government efforts to provide resources for the health and the social security systems just as political fractiousness undermined attempts at reform. IMF standby agreements highlighted inefficiencies and spending on the social security and health systems (International Monetary Fund July 2002). Prior to the 2002 parliamentary elections, ministerial leadership spear-headed by the Minister in charge of Economic Affairs and Treasury Kemal Derviş, developed plans to address deficit issues, but the weak coalition governments could 11 muster neither the leadership nor the political support to fully implement reforms (Boulton and Wolf 2002; Akyüz and Boratav 2003). The IMF was concerned that the social security deficit was growing unsustainably, which in retrospect seems to have been accurate. The social security deficit grew from 1.9 percent of GDP in 2000 to more than 4 percent of GDP by 2005, the last year before reforms took effect in 2006 (World Bank 2006). Upon taking power, the AK Party inherited urgent problems in social security and health spending, and also was bound by commitments made by previous governments with the IMF to reduce government spending and stabilize the economy. High Public Support for Reform 17. A wave of anti-government sentiment and public support for reform were important factors in the AK Party’s victory in the 2002 elections. As political outsiders, the AK Party drew support from public dissatisfaction with the inaction of coalition governments, a series of high profile corruption scandals, and economic instability, including the crisis of 2000 –1 (Heper 2003). Crucially, the legislative majority won by the AK Party afforded it the power to pass reforms unilaterally and obviated the need to broker agreement between coalition partners. 18. After coming to power the AK Party was under intense public pressure to act on two fronts. One was to reform the economy. The other was to improve health service delivery, particularly in rural and poorer regions of the country where the party’s political base was centered. Inaction on either of these two fronts carried the risk of backlash in future elections (Özbudun 2006; Baris, Mollahaliloglu et al. 2011). Over the ten years of AK Party control covered by this report, health became an increasingly crucial element of the party’s political success. According to interviews with current and former senior party strategists, they initially viewed health as a way to address equity issues and send resources to their political base. As the magnitude of problems in the health sector became apparent, they began to appreciate how many citizens were in need. They also became increasingly aware of the potential electoral advantage that could be garnered by improving peoples’ access to quality health services. According to preliminary calculations, as many as 20 million people lacked basic services. Once the AK Party understood this need, its leadership put its full weight behind the HTP’s prompt adoption and implementation. As one inte rviewee formerly working as a senior strategist at the MoH characterized the sequence, “In the beginning the politicians chose health, but by the end health chose the politicians.” Increased Fiscal Space for Health 19. Favorable economic conditions increased the availability of resources for health sector, which the AK Party used to fund and implement the HTP. Enabling much of this expansion of fiscal space for health was strong overall economic growth that increased GDP by two-thirds in real terms, and real GDP per capita by 70 percent, from US$ 5952 to US$ 8493 between 2002 and 2011 (Figure 1) (World Bank 2012).However, the share of GDP spent on health remained at 5.3 percent between 2003 and 2011 (OECD 2013). Calculated over the same period on a per capita basis in constant 2005 dollars (PPP) health spending grew by 109 percent (OECD 2013). Figure 1: GDP per capita in Turkey, 1990-2012 (constant 2005 US$) 10000 8000 6000 4000 2000 0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 Source: World Development Indicators 12 Favorable Demographics and Low Expectations 20. The young Turkish population required relatively basic, primary care services as compared to expensive, hospital-based treatments for older populations. In 2002, 30 percent of Turkey’s population was under 15 years of age and only 6 percent was over 65 years of age (World Bank 2012). In addition to these favorable demographics, after years of failed attempts to reform the public health system, Turkish citizens had low expectations for the delivery of even these basic services. In 2003, only 39.5 percent of the population indicated that they were satisfied with the quality of care (OECD, WHO et al. 2008; Bleich, Özaltin et al. 2009). This scenario gave the MoH an opportunity to make quick gains in patient satisfaction with relatively small improvements to the system. 13 PART III – POLITICAL ECONOMY CHALLENGES IN THE TURKISH HTP (2003-2012) 21. Health reform is a politically charged process because it involves the reallocation of resources and responsibilities. Common goals of health reform are to increase equity, improve access for the poor, and to establish some minimum basket of services for all citizens. These goals all require the redistribution of resources, which raises the prospect that the process will generate winners and losers. Typically, groups that are well-off before the reform perceive discussions of policy change as potential threats to their benefits. This leads to a collective action problem because although a reformed system could become more efficient and provide more benefits for everyone, it is typically contested by small, well-organized groups whose interests would be impacted by a reform, including physicians, other providers, and groups with generous benefits, such as civil servants. Those who have the most to gain from reform are those who receive few or no services, but these groups tend to be unorganized and unengaged in the reform process, even if they are large in number. As Mancur Olson (1971) wrote, a collective action problem arises because “rational, self - interested individuals will not act to achieve their common or group interests,” unless the number of individuals in a group is small, or there is some incentive that makes individuals act in their common interest. Small groups will act towards a collective goal because they face relatively low costs to organize due to more uniform, individual interests and will reap high potential benefits per capita. Whereas large groups will face relatively high costs when seeking to organize for collective action and the potential benefits per capita may be small because they are distributed across many members. 22. For these reasons, health reform is typically a contentious, difficult process. In the subsections below, we discuss four political economy challenges, including engaging beneficiaries to build support for reforms, managing the influence of opposing interests, managing the influence of the IMF and World Bank on domestic politics, and covering the poor and unifying benefit systems. Challenge #1: Engaging Beneficiaries to Build Support for Reforms 23. The basic political economy of reform favors opponents rather than potential beneficiaries because the costs tend to be concentrated on well-organized groups with high access to political and economic resources, which they can use to impede, dilute, or otherwise influence the process in their favor. By contrast, although potential beneficiaries tend to be far more numerous, they have few economic or political resources and are not likely to be engaged or organized in support of reforms. Successfully adopting and implementing reform requires strategies for overcoming this challenge. In most settings, opponents include the urban elite: physicians and other health workers whose rights, responsibilities, and pay are often directly affected by reform, and formal-sector workers, whose taxes can be used to finance services for the poor and whose benefits might be reduced accordingly. Potential beneficiaries are usually poor, reside in rural areas, and are underserved by the health system. The lopsided distribution of political and economic resources favors reform opponents in almost every way. A central challenge of reform is to somehow change or overcome this imbalance to generate sufficient popular support to overcome any remaining opposition. 24. In the Turkish Case: An important aspect of MoH’s overall strategy was to quickly build support among the intended beneficiaries of reform, whose large numbers represented a potentially enormously powerful political force if organized behind the HTP. Those with no or limited access were the bulk of Turkey’s citizens, living in the East and South in smaller cities and rural areas. Their support was important to the AK Party in general and would be required for more difficult aspects of the reform, which urban elites and organized interest groups would oppose, according to political analysis commissioned by the MoH (Rossetti 2004). The same analysis also showed that in Turkey’s recent past, political parties attempting reform had faced a backlash in subsequent 14 elections; to avoid this fate the AK Party would have to demonstrate results before the next general election, in 2007 at the latest.4 25. The MoH built public support reform very quickly by focusing its early efforts on highly visible changes to the existing health system, acting first in the areas with the least services. Many interviewees with experience in the reform related this emphasis on immediate and noticeable improvements with the dual purpose of improving service delivery and patient satisfaction, while bolstering the political viability of the HTP. A recent World Health Organization report discusses the Minister of Health’s approach as similar to how a team of medical doctors treats a trauma patient, first treating the most life-threatening problems before moving onto systemic and long-term issues (Johansen and Guisset 2012). For instance, among the first changes ordered by the senior leadership team was the abolishment of the unpopular practice of holding patients in facilities until their bills were paid. In our analysis, the choice to focus on this issue shows a nuanced assessment of the priority of reforms. Holding patients as pawns until families could settle bills no doubt deterred some care-seeking, but this practice was employed after care was delivered, meaning that its primary consequences were probably in creating extremely negative feelings toward the health system and imposing financial distress on families. Ending the practice showed sensitivity to the moral and political dimensions of health care and not only to the medical issues, as implied by the trauma team analogy. We emphasize this point because generating support for reforms inside the short election cycle required changes that could improve popular impressions immediately. 26. In fact, regardless of their public rhetoric on trauma teams, the senior leadership of the MoH realized that citizen impressions of the health system were formed mainly by primary-care experiences, rather than in secondary or tertiary care, according to current and former officials who were engaged in formulating the HTP strategy. Initial reform efforts sought to improve these impressions, for instance by refurbishing waiting rooms, conducting outreach activities to make patients feel welcome, and converting break rooms to exam rooms to increase the capacity to deliver services. As one senior MoH official related as a generalized example, prior to the reform a health facility might have had a staff of 30 health workers but care would be provided in only three rooms. Other rooms were dedicated to other purposes, including separate break rooms for each type of employee—physicians, nurses, technicians, and others. Some were used as private offices, but MoH calculations estimated organizational and infrastructural inefficiencies limited productivity of these facilities to about 30 percent of their potential capacity. These calculations indicated that far more of the demand for health services could be met in public facilities if existing employees were motivated to work there and existing infrastructure were reoriented around the primary purpose of care delivery. The MoH therefore concentrated initially on making more efficient use of its existing resources, which it could do quickly, and did not prioritize the slower, more costly task of building new facilities. To overcome the resistance of some health workers, the Minister of Health and his team personally visited facilities and directly ordered the consolidation of break rooms and other space-saving measures to increase care delivery capacity. 27. The senior leadership also rapidly expanded emergency transport services, increasing the number of ambulances and extending the system’s reach with specially equipped boats and fixed - and rotary-wing aircraft to serve remote areas and afford fast transport for critical cases. A senior MoH official involved in the design and implementation of the HTP estimated that these enhancements led to an approximately 5-fold increase in the number of emergency transport vehicles in the country over the first 10 years of reform. Again, this was a choice made on several grounds, only one of which was the technical consideration of bringing more people to care in times of need. Senior officials stressed to us that the ambulances were powerful, eminently visible symbols of a health system that cared for its citizens. Particularly for rural citizens covered for the first time, the emergency transport system was a compelling demonstration of the responsiveness of the MoH and the AK Party. An interviewee involved in these policy decisions described the initial 4 General elections are held at least every five years, but they are often held earlier at the request of parliament or if parliament is dissolved by the president. 15 phase as capacity building. He said the leadership in the early stages did not want to build new infrastructure, and focused instead on increasing capacity of the existing system. 28. These relatively simple changes improved public support for the reform and helped create the political momentum for more difficult, large-scale changes to the system planned for future years. Rapid and publicly-visible changes to improve the accessibility of the system appear to be reflected in the percentage of people reporting problems making an appointment for an examination or analysis, which dropped from 59.59 percent in 2003 to 29.30 in 2005 (Turkish Statistical Institute 2003-2012). Over the same period—the first two years of the HTP—citizen satisfaction with health services overall rose from 46.17 percent to 55.27 percent (Turkish Statistical Institute 2003-2012). These satisfaction rates were reflected by public support for the AK Party in both the 2007 and 2011 elections. In these the AK Party continued to build on its electoral majority by placing its health reform achievements as a centerpiece of the party platform (Bryant 2010). Challenge #2: Managing the Influence of Opposing Interest Groups 29. Interest groups, including physicians and beneficiary groups, tend to exercise strong influence on reforms because they are well organized, have high access to political and economic resources, and are usually closely engaged in the process because it directly affects their interests. Physicians tend to view government insurance programs as a threat to their autonomy and are therefore likely to oppose such reforms using their political influence and their authority within the health system (Immergut 1990). Trade unions and formal sector workers typically have health benefits under pre- reform systems. Often, health reform threatens their interests because extending coverage to low- income groups means redistribution, which might be expressed as a reduction in their benefits or a diversion of their funds. For instance, formal sector workers in Ghana were initially opposed to health reform in 2003 because increased access for the poor was partly funded through a 2.5 percentage point diversion of their pension contributions (Rajkotia 2007). Organized beneficiary groups also tend to oppose the integration of other health insurance schemes into their own because theirs is typically the best funded and unification therefore means a dilution of resources available for their uses. Unifying benefit systems can be desirable from an efficiency perspective because it has the potential to simplify administration and reduce overhead, which would free more resources for the provision of services. But the political economy of unification is extremely difficult because its redistributive elements impose immediate costs on small, organized, and powerful interest groups. On the other side, the benefits are abstract, would occur in the future, and would accrue to the poor, who, in most cases, are not organized to provide support because they have few economic and political resources, and are not likely to be engaged in the process. Organized interest groups often pose significant opposition to reforms, as they have in the United States and Mexico, for example (Hacker 1998; Lakin 2010). 30. In the Turkish Case: In the initial stages of the HTP, the Minister of Health and his senior leadership team identified groups important to the reform and developed strategies to persuade or overcome those expected to oppose it. Plans to manage this opposition were then incorporated into the reform strategy. After delineating stakeholder groups, the Minister of Health and his senior leadership team began to engage opposition groups to gauge the possibilities to win their support. In some cases, the reform plans were adapted to accommodate concerns. In other cases, the MoH and AK Party leadership worked to neutralize the influence of opponents by splintering their support or delegitimizing their views. 31. Political analysis played an important role in the management of opposition groups. Several interviewees directly involved with the reform mentioned the influence of a stakeholder analysis report and a then-recently published book, Getting Health Reform Right, which stresses the importance of politics in determining both the trajectory and outcome of reform (Roberts, Hsiao et al. 2003; Rossetti 2004). The authors note that “astute policy developers begin political analysis early in the policy cycle.” The MoH also had the book translated into Turkish so it would be more accessible to a greater proportion if its staff. The stakeholder analysis was commissioned in 2003 to provide a roadmap to the politics of the reform in Turkey and a guide to dealing with opposition. The report analyzed the positions and influence of stakeholders involved, assessed future electoral ramifications, and proposed strategies to manage interest groups during the implementation phase 16 of the reform (Rossetti 2004). The report identified public providers, members of the social security institutions, and the central government bureaucracy and civil servants as key actors opposed to the HTP. By understanding the influence opponents were expected to have and the reasons for their positions, the government could plan how to manage the politics of policy adoption and implementation. 32. Trade unions were one of the most influential beneficiary groups opposed to reforms. Senior Government staff, including representatives from the MoLSS held a long series of meetings with union representatives to discuss how the reforms would affect the benefits of their membership. The MoH prepared numerous analyses to forecast benefits under various assumptions to reassure representatives that in no case would benefits decrease under the HTP, and that in most cases benefits would increase. Several informants involved in designing the reform reported that the inter- ministerial working group had initially planned a basic benefit package with options for supplementary care. However, as the group continued its discussions, equity emerged as an increasingly important consideration. In its final form, the reform’s long -term goal objective was to provide all citizens with the same benefits as retired civil servants, who had the most generous of all pre-reform packages. This strategy dramatically increased the resources required for reform, but it helped ensure that most organized beneficiary groups would not oppose the reform. To address financial sustainability, the reform included cost control in the form of a family physician gatekeeper and capitation system. These mechanisms were expected to reduce the potential for physician induced demand and limit overuse of secondary and tertiary care. However, the MoH was not able to implement the capitation system. The gatekeeper function was not implemented due to public opposition and also because there were not enough family physicians to serve in that role, anyway. Once beneficiary groups understood the benefits they would gain, they began supporting the reform and influenced its design to enhance their benefits. 33. A second influential opposition group was white collar civil servants, who opposed the reform for two general reasons. First, as the beneficiaries of the most generous entitlement package, they feared that reform would diminish their benefits. Second, most of these elite civil servants were secularists and tended to oppose the AK Party politically. They were also concerned about the implications of the concurrent social security reform, particularly for the retirement age, which stood at 48 years of age for women and 52 years of age for men. The MoH and MoLSS leadership first deployed a persuasion strategy in an attempt to convince the white collar civil servants that their benefits would not decrease. But the attempts were not successful, and elite civil servants appealed to President Sezer to block the reform. To overcome their opposition, the MoH and MoLSS decided to exempt all existing civil servants from the reform, agreeing to apply new rules only to those hired in the future. 34. Health workers are typically amongst the most influential groups in health reform because they are called upon to deliver health services, are well organized, have an influential position in society, and usually have access to political resources (Immergut 1992). The role of health workers in the reform was identified in the political analysis report, and likely was intuitive to the Minister and senior officials because almost all of them had long experience in the health system. The Minister and his team engaged in dialogue with health workers in the early stages of the HTP and devised strategies to manage their interests. The participation of health workers in the planned reforms was essential to improving service delivery, since health workers are the ones who actually deliver the services. But one of the biggest problems in increasing delivery was a shortage of trained professionals; for instance, Turkey’s ratio of physicians per population was only about one - third the EU average when the HTP began (Tatar, Mollahaliloglu et al. 2011). For the reform to succeed, the workforce would have to operate at higher capacity. The MoH provided incentives for health workers to deliver more services by linking pay with the quantity of services provided and patient satisfaction. This system dramatically increased the salaries of physicians providing services in the public sector. The pay-for-performance scheme was used as a way to allocate additional pay to physicians and nurses, and had the additional advantage of avoiding the cumbersome bureaucracy and legal obstacles, associated with adjusting pay under the formal civil service regulations. The MoH also brought more delivery capacity into the public sector facilities by ending so-called dual practice arrangements, under which physicians would spend some of their 17 time in public facilities, but also see patients in private practices. The increased pay available under the pay-for-performance scheme was intended to compensate physicians for the remuneration they could no longer earn in private practice, draw them into public service full time, and incentivize them to provide more care. In the view of senior officials we interviewed, the pay-for-performance scheme was an important motivating factor for health workers, who have faced heavy workloads under the increased demand created by the HTP. Some interviewees also referenced this policy as a critical factor in decreasing dual practice, even before the MoH was able to pass the Law on Full-Time Practice, which prohibited MoH physicians from also working in the private sector. 35. The pay-for-performance system pleased physicians primarily engaged in service delivery, but it did not change the opposition of all physicians or other health workers. With groups that remained opposed the MoH worked to marginalize their influence. These opponents included the Turkish Medical Association (TMA) and the Turkish Nurses Association. The government accomplished the reform without the support of these powerful groups by fracturing their membership and swaying popular opinion against their leadership. Remaining opposed were the medical elite—far fewer in number and limited to specialized facilities in major cities. Although the membership of the TMA was unified in opposition at the beginning of the reform, much of its membership was primarily engaged in service delivery and became supportive because of the incentives available under pay for performance. The organization’s leadership, those in specialized roles, and members of the academic elite remained opposed, but were relatively few in number. A senior TMA official observed that the many specialists in staunch opposition had had little influence on the reforms, citing vast public support for the Ministry’s plans and publicity campaigns against the medical elite. Our interviews with senior officials at these organizations and in academic medicine revealed intense dissatisfaction with the reform, and a deep distrust of the MoH leadership. As the pay-for-performance system does not include allowances for teaching or research, those interested in these activities felt personally punished, in the assessment of those we interviewed. These people expressed dissatisfaction with personnel allocation policies, as well. After graduating from medical school, all new physicians perform two years of public service. Before the reforms, the best students with academic interests were assigned to leading medical schools as assistant professors. But under the reforms, nearly all graduates are now assigned to public facility roles, to MoH hospitals, or to new medical schools with close relationships to the MoH. As a result, there are few younger faculty trained in what were formerly the most prestigious places. Specialist physicians we interviewed explained their discontent with a long list of problems. Because of cost controls, they are not always able to procure the supplies they need to serve patients, they said. Because the scheduling system makes appointments that are 15 minutes in length, there is too little time to adequately diagnose problems or provide lifestyle guidance, we were told. Because of these and other problems, many faculty have left —some for private practice, some for other countries, and some to retirement. The capacity to perform complex procedures at leading medical centers has been severely reduced, we were told. Our interviewees also expressed concerns for the quality of care under the HTP because quality assurance rests on patient satisfaction, but in many cases patients are not well informed about what care is appropriate. As complaints from patients to the MoH constitute a serious issue for providers, the physicians we interviewed expressed concerns for the loss of autonomy and a compromised doctor-patient relationship. 36. To neutralize the threat to the reforms posed by opposition groups, a new union of health workers was established to draw supporters away from existing professional associations and undermine their support base. By creating factions within health workers, they reduced the power of TMA and Turkish Nurses Association to act as a united voice for all providers. A similar strategy was employed to counter resistance from YÖK – the organization responsible for supervising all Turkish universities, and which at the beginning of the HTP was still controlled by appointees of previous governments and distrusted the AK Party. Interviewees from elite universities voiced concern that the reforms had channeled resources away from their institutions, as a political tool by the AK Party leadership to undermine their influence. The government worked around this opposition until 2008, when the newly-elected President Gül was able to appoint a new head of YÖK friendly to the AK Party. These tactics allowed the MoH and AK Party to overcome the resistance from previously strong interest groups that remained in staunch opposition to the reform. 18 37. The MoH ensured that the central components of the HTP were adopted and implemented by engaging those groups likely to oppose the reforms. Some they persuaded to support the reforms and others they were able to neutralize or overpower. Challenge #3: The IMF, the World Bank, and Domestic Politics 38. By choice or necessity, many countries engage the IMF and World Bank in policy reforms, but the relationship between these institutions and domestic political economy processes can be complicated. The power dynamics can be especially complex at times of crisis when countries are at the most need of assistance. The IMF’s role in economic stabilization and fiscal reform as a lender of last resort gives it very high political and economic power. The World Bank’s engagement is particularly notable in social sector reform because of its strengths in project design and implementation, as well as its role in discussions between various ministries, including both those that provide finances and those that deliver services. Both international institutions become involved in the policy reform process through their lending activities and technical expertise. Middle- income countries can strategically use financial and technical resources available from both the IMF and World Bank. However, there is the potential for tension between the policies and advice of the international institutions and the direction of the domestic policy agenda. 39. In the Turkish Case: The Turkish Government sought support from the World Bank and IMF over the course of the HTP (2003–2012). Both institutions were closely engaged in the Turkish economy and health sector when the AK Party took office in late 2002, including through IMF standby agreements relating to the economic crises of 2000/2001 (International Monetary Fund July 2002). As part of these agreements, Turkey had pledged to reduce inflation and bring its public sector debt under control (Alper and Alper 2003). The MoH had a long-standing relationship with the World Bank dating to the first health policy loan agreement in 1990. These ties strengthened under the AK Party government; in 2004 the MoH signed a US$ 75M loan agreement for hiring technical expertise to support the design and implementation of the HTP (Republic of Turkey and International Bank for Reconstruction and Development 2004). The MoH was able to benefit from the financial and technical resources as well as crucial political support for the Washington institutions, while maintaining its own leadership over the full course of the reform’s design, adoption, and implementation. The MoH drew on the Bank’s support to improve the reforms’ legitimacy in domestic politics, and the AK Party government re lied on the IMF’s approval of domestic budgets and financial affairs to help calm markets at home and abroad. 40. AK Party leadership and several ministries relied on IMF standby agreements to stabilize the economy and public support by IMF representatives to reassure financial markets. As part of these agreements, IMF representatives were intimately involved in policy discussions that surrounded expenditures and revenues. One of Turkey’s most important pledges under the IMF standby agreements was to control social sector spending. A major part of the underlying fiscal problem was uncertainty that had resulted from several years of incorrectly low forecasts and subsequent actual expenses that were far higher, which led to budget overruns and large deficits. The Minister of Health and his team worked to convince IMF officials of the fiscal soundness of the health reform proposal by presenting detailed forecasting models of the potential for long-term cost saving under a UHC system that promoted primary care, even if some short-run costs might be higher. These discussions could be highly contentious because of divergent interpretations of the forecasts and because of the potential conflict between providing universal coverage and reducing the country’s high public sector debt. In these discussions Minister Akdağ relied on the strong support of the Prime Minister for the HTP to advance the reform proposal despite reservations by some parties. 41. The MoH had a long-standing relationship with the World Bank and continued to work closely with it over the course of the reform. The well-established working relationship with the World Bank on health system improvements allowed the Minister of Health and the inter-ministerial working group to quickly access flexible, external resources as needed. Working group members interviewed for this study recalled that the Government agreed to a relatively small World Bank loan (US$ 75M) specifically to gain access to the expertise of the Bank’s staff and its international network of consultants. The capacities gained this way were augmented by hiring into the Ministry many of the consultants who had worked for the World Bank in Turkey in support of health projects 19 in the 1990s. Many of our interviewees recalled the importance of advice and strategies contributed from these sources, including about a dozen background papers funded by Japan via a World Bank trust fund. One of these papers analyzed the politics of the proposed reform, the stakeholders involved, its future electoral ramifications, and proposed strategies to manage interest groups during the implementation phase of the reform (Rossetti 2004). The MoH used the loan proceeds to commission reports and implement programs to support the design and implementation of the HTP with much greater flexibility than would have been possible with government revenues. 42. Interviewee accounts referenced the largely, positive working relationship between the MoH and World Bank throughout the reform process. However, at times MoH plans did not fit within typical World Bank timelines, or processes, or procedures. In these cases, staff members of the MoH and the World Bank worked together to find mutually acceptable solutions. For instance, Minister Akdağ proposed hiring teams of Field Coordinators to directly manage the implementation of the HTP in each of Turkey’s 81 provinces. The purpose of these teams would be to communicate directly with the Minister, providing him with detailed and timely information from the front lines of the reform. As recounted by several interviewees directly familiar with the establishment of the Field Coordinating Teams, the hiring process was politically complex. Each team was to be led by an advisor personally known and trusted by the Minister, a requirement designed to limit local influence on the information, which might have downplayed problems or overly emphasize progress. The Minister was particularly concerned about obtaining accurate information to ensure that local committees were performing according to MoH policy. This was potentially contentious because it asserted MoH authority over some matters that local committees had sometimes handled themselves. Green card distribution had in some cases been a local patronage opportunity, for instance. 43. But there were significant challenges to hiring the Field Coordinators as envisioned by the Minister. First, he needed to entice trusted staff to take leave from their current jobs in Ankara to work in outlying provinces. Second, he needed a way to ensure that Field Coordinators would still have jobs available to them when their tenure was over. Third, he needed to work within World Bank procurement guidelines to place both qualified and loyal staff in these positions. The basic tension arose between the Turkish government regulation that civil service slots could be held open indefinitely only for employees leaving for international agencies, and the World Bank’s standard guidance against sole-source, or non-competitive contracting, which was preferred by the Ministry because of its concern for hiring trusted and loyal Field Coordinators. Ultimately to introduce the Field Coordinators into the health system and overcome these obstacles, the Minister and his leadership team succeeded in negotiating these bureaucratic constraints with a creative mixture of funding and hiring through different multilateral agencies. Putting loyal staff in a front-line monitoring role extended the Ministry’s centralized authority deep into the field with accurate information from the local level and a much greater ability to intervene when complexities or problem threatened the reform’s progress. 44. The Minister of Health and his team learned from the history of interactions between the IMF, World Bank, and the Turkish Government to use the institutions’ valuable resources to their utmost advantage. Their leadership skills are demonstrated by their resolute commitment to the central objectives of the HTP and figuring out mechanisms to effectively achieve them, as well as their willingness to accept and benefit from the technical expertise of the World Bank. Challenge #4: Covering the Poor and Unifying Benefit Systems 45. Many countries have attempted to create UHC systems by expanding entitlement programs for low-income groups, but in recent years only Turkey has succeeded in then integrating the entitlement programs into a unified system covering all citizens. Ghana, Mexico, and Thailand, for instance, have all undertaken reforms to move toward UHC. These countries have all been successful in expanding coverage, particularly for the poor. But merging the subsidized programs for low-income households with existing schemes for formal sector workers has remained elusive (Hughes and Leethongdee 2007; Agyepong and Adjei 2008; Knaul, González-Pier et al. 2012). Combining coverage plans for the poor with those for formal sector employees is very difficult because it represents an enormous challenge of political economy. Formal sector workers tend to 20 be well organized and influential politically, and usually enjoy the most comprehensive benefit package, at least in part because they usually contribute the most resources to the system. Formal sector workers therefore have high economic and political power. Reform usually threatens their interests because the redistribution required to cover low-income groups implies a potential reduction in their own benefits and/or the use of some of their contributions for others. Formal sector beneficiaries have resisted integration efforts for these reasons. Low-income groups whose members stand to benefit from reform tend to have relatively little economic power and are usually not engaged in the electoral process, where their large numbers could constitute substantial political power. This generic political economy is thus very unfavorable for reform because its likely opponents have direct interests at stake and have access to political and economic resources, while potential proponents are likely to be unorganized, disengaged, and have limited access to either political or economic resources. 46. In the Turkish Case: To expand coverage and move towards a unified UHC system, the MoH worked around obstacles to postpone or avoid potential opposition. We highlight three important steps: First, the MoH decided to use the existing Green Card Program as its primary vehicle for scaling up coverage for low-income households in part because modifying a program did not require parliamentary approval, whereas starting a new program would have. The MoH simply adapted the Green Card Program to fit the policy objectives of the HTP under its own authority. Second, the Minister of Health and his senior leadership team brought the Green Card Program under the Ministry’s auspices so they could control it completely. Third, they stimulated demand for the Green Card Program by expanding the benefits package, increasing the number of Green Cards in circulation, and making concurrent supply side improvements. Throughout this process the MoH carefully monitored the program’s progress in implementation. This progress t owards expanding coverage was embraced by low-income households in rural areas of Turkey. And as the benefit package expanded and the cost of the program grew the MoH reduced the potential for opposition to an integrated system by social security beneficiaries—because the discrepancies between benefits was shrinking—and financing authorities because the marginal cost of doing so was also shrinking. 47. Once UHC was identified as the primary objective of the HTP, senior leaders at the MoH debated several policy options before settling on the Green Card, according to several interviewees who participated in the process. They chose to work with an existing program, rather than creating an entirely new scheme for low-income households, which would have required legislative approval. The Green Card Program was chosen because it was the only part of earlier attempts to create a universal, general health insurance scheme that had been passed by the parliament and implemented. But before the HTP, the Green Card program had significant limitations. It provided coverage for low-income households only for inpatient expenses incurred in public facilities, and it was widely regarded as unsuccessful because of corrupt enrollment procedures, a limited benefit package, and poor public service quality (Karadeniz 2012; Menon, Mollahaliloglu et al. 2013). Estimates based on the 2003 Household Budget Survey show that there were only 2.5 million beneficiaries and of those households enrolled only 31 percent were in the poorest decile (Aran and Hentschel 2012). But these problems aside, the Green Card program did exist in law and did operate, even if imperfectly, both of which gave the Minister of Health and his leadership team an avenue for delivering services, expanding entitlements, advancing their policy goals, and generating public support without having to enter the parliamentary process and sustain the attendant delays. 48. As a second step, the MoH drew on the strong support of the Prime Minister to bring the budget and administration of the Green Card Program under its own auspices. Prior to the reforms, Social Solidarity Foundations under the Prime Minister’s Office ran the Green Card Program. In 2004, the MoH requested direct control of the program so that it could then oversee its expansion efforts and work to address bottlenecks in implementation without needing approval from other ministerial entities. Because the Social Solidarity Foundations were under the Prime Minister’s control, he was able to easily transfer the Green Card P rogram to the MoH based on the Minister’s request. Interviewees involved in this process reported that a first step in gaining administrative 21 control of the Green Card Program was to replace the “green cards” with new “green booklets” as a mechanism to make all enrollees report to local authorities, where they could be counted. 49. The MoH gained administrative and budgetary control of the Green Card Program in 2004; however, avoided creating local opposition by initially refraining from changing any enrollment procedures, which would have affected patronage relationships at the local level between local committees and enrollees. Local committees continued to be responsible for processing applications; however, the MoH had direct monitoring and reporting mechanisms to ensure their policy objectives were followed. In doing so, they avoided opposition from local committees, which were accustomed to making eligibility determinations. Instead, the government waited and incorporated eligibility determination into the Ministry of Family and Social Policies’ IT system in 2012 only after enrollment abuses arose in the public discourse (Hurriyet Daily News 2010). 50. Third, once the administrative arrangements for Green Card reform were in place, the MoH worked to increase demand for the program by expanding the benefits package and improving the public sector delivery system. The MoH expanded the benefit package, adding coverage for outpatient services in 2004 and coverage for outpatient medicines in 2005 (Menon, Mollahaliloglu et al. 2013). The MOH needed to increase demand for the Green Card Program to power the enrollment required to raise coverage rates and ultimately to secure the requisite electoral support for the reforms from low-income households residing in rural areas. Many key informants described an explicit MoH policy to distribute as many Green Cards as possible to bring more people into the health system. This was both technically and politically expedient because it provided more benefits to citizens, empowered local committees to facilitate the process, and increased the electoral support for the AK Party amongst its base. Their efforts to stimulate demand proved successful as the number of Green Card holders increased from 2.5 million in 2003 to 9.1 million in 2011 (Figure 2) (OECD and World Bank 2008, Turkey MoH 2012). Figure 2: Number of Green Card Beneficiaries, 2003 2011 12 9.4 9.3 9.6 9.4 9.1 10 8.3 6.9 7.3 8 Millions 6 4 2.5 2 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: Menon, Mollahaliloglu et al. 2013 51. These expansion efforts required additional funding. A member of the senior leadership team reported that by gradually expanding both the benefits and beneficiaries of the Green Card Program, the MoH was able to desensitize those within the government responsible for financing the reform. But viewed over several years, expenditure on the Green Card Program increased dramatically—from 3.8 percent of public health expenditures in 2003 to 10.8 percent in 2007 and 8.4 percent in 2009. Green Card expenditures as a percent of total public expenditures increased from 0.4 percent in 2003 to 1 percent in 2009, and from 0.2 percent of GDP in 2003 to 0.4 percent of GDP in 2009 (Menon, Mollahaliloglu et al. 2013). 52. Concurrent with efforts to maximize Green Card enrollment, the MoH also invested in improvements to the public health delivery system. It focused its efforts on rural and poorer areas of the country, where Green Card eligible individuals resided. In an interview, Minister Akdağ explained that providing financial protection for poor households was not enough, and that the MoH also had to ensure they had access to health services (Baris, Mollahaliloglu et al. 2011; Johansen and Guisset 2012). The family medicine program, conditional cash transfers for maternal health services, pay for performance scheme, and merging the SSK hospitals into the MoH system, were 22 all reforms that directly benefitted Green Card enrollees. These supply-side efforts, combined with the expanded benefit package, were used as enticements to increase demand for the Green Card Program. 53. By the time the MoLSS and MoH required approval by the Turkish Parliament to merge the social security institutions and create a unified social security scheme in 2006, the MoH had already greatly expanded the benefits and increased the number of Green Card beneficiaries. Despite this progress, unification and integration efforts were still delayed. First, the MoLSS and MoH had to overcome a Constitutional Court challenge to the unified system from President Sezer, who was generally opposed to AK Party initiatives, on the grounds that it would disadvantage civil servants. Once provisions were made to ensure that current civil servants could keep their existing benefits, delays in integrating the Green Card Program were caused by difficulties in creating a rigorous means-tested eligibility determination system. Several interviewees reported abuse in the enrollment system and the lack of capacity within the MoH and MoLSS to effectively implement an income-determination system. There had also been popular backlash against the Green Card Program based on news reports of abuse and over-enrollment. The MoH responded with promises to reform the system and conduct more rigorous income tests (Hurriyet Daily News 2010). Interviewees reported that in response to this public disagreement the MoH began a formal review of Green Card enrollees to reduce abuse in the system and worked to ensure premium support was provided only to eligible households. 54. By the time actual integration of Green Card holders into the General Health Insurance System commenced in January 2012, the hard work of increasing benefits, expanding coverage, and standardizing enrollment systems had already been done. On January 1, 2012 the newly formed Ministry of Family and Social Policies (MoFSP) took over responsibility for determining eligibility for premium support from the MoH (Menon, Mollahaliloglu et al. 2013). Green Card beneficiaries had 12 months to reapply to receive premium support, which led to the official abolishment of the Green Card Program. In the first year of the new system, 7.5 million people were eligible for full premium support, and an additional 4.5 million were eligible for reduced premium payments (Ministry of Family and Social Policies May 2013). Under this new system, the MoFSP determines eligibility, the Ministry of Finance pays the premiums for beneficiaries directly to the Social Security Institution, the Social Security Institution pays for the health services, and the MoH delivers the health services to the beneficiaries. 55. This incremental approach to expanding coverage and unifying all health coverage schemes allowed the MoH to provide benefits and get the support of low-income households for the health reform without stirring up opposition from financing agents over the program’s fiscal implications. As we discuss under contextual factors, it certainly helped that the Turkish economy grew during the reforms. Exempting the highest entitlement group from reform neutralized their opposition, and as rising benefits for the poor narrowed and eventually closed the gap in benefits eliminated resistance from other high-benefits groups. 23 PART IV – DISCUSSION 56. In this section we discuss some of the most significant strategies used by the MoH to overcome the four political economy challenges as identified through our interviews, and in our subjective judgment. We organize our discussion according to the political economy challenges we identified above. 57. Building support among beneficiaries: To build support among beneficiaries of the reform, MoH leadership prioritized service delivery improvements for underserved populations over investments in other areas of importance to the health system. For instance, the quality of care, research, teaching, and specialized medicine did not receive the attention or resources dedicated to primary care. In the short run, the redistribution of resources was designed to address inequities in the system and build political support for the reforms, although the relative neglect of specialists and academic medical centers poses grave risks to the future of the system, as we mention in the next section. Building support among beneficiaries required improving popular attitudes toward the health system and attracting more people to public facilities. Recast, the problem was to encourage citizens to consume more health services. The success of the HTP in increasing patient visits and citizen satisfaction testifies to the efficacy of these efforts. However, no country has been able to provide as many health services as are demanded by citizens. The imbalance of unlimited demand for health services and finite resources is addressed through rationing systems. As the demand for health services continues to rise, fiscal concerns will require Turkey to consider ways to limit the care-seeking or ration care provided—problems that arise because of the HTP’s success and whose solution is essential to the long-term financial sustainability of the system. 58. Managing the influence of opposition of groups: The MoH succeeded in winning the support of some of the groups that initially opposed the HTP, but even ten years into the reforms some groups remain vehemently opposed. Many specialist physicians, the elite ranks of nurses and midwives, and some other health workers have been marginalized and our interviews reveal highly negative feelings among some of these professions toward the MoH and the policies that guide the health system. Satisfaction surveys of urban physicians conducted by the TMA, for instance, show very negative views of current job satisfaction and low expectations for future improvements (Tanik, Bilaloğlu et al. 2013). These sentiments may undermine the profession by discouraging physicians from staying in practice and deterring students from entering medical school. Similarly, among nurses, midwives, and other health professionals low satisfaction threatens the performance of current workers and may diminish the number of people choosing to train in these areas. These dynamics threaten the supply and quality of services the system can provide. 59. Managing the influence of the World Bank and IMF on domestic politics: The MoH’s strategic management of its relationships with the IMF and World Bank carries risks and reflects the importance of negotiating a united front between ministries. The MoH is not typically a strong ministry in domestic politics and therefore tends not to have as much leverage as other in its negotiations with International Financial Institutions, particularly not with the IMF, which focuses on macroeconomic issues. The Ministry of Finance and the Undersecretariat of Treasury are usually the ministerial bodies with the most bargaining power in these interactions due to their role in the fiscal affairs of their country. The creation of an inter-ministerial working group to devise the HTP comprised of members from the financially-oriented ministries, as well as those most concerned with service provision, gave the MoH a far stronger position compared with planning on its own. This ministerial coordination, along with the complete support of the Prime Minister, allowed the Minister of Health and his team to take an aggressive position with both organizations to fully leverage their resources in support of the HTP. 60. Covering the poor and unifying benefit systems: The sequenced approach taken to expanding the Green Card Program shows one way to incorporate a targeted entitlement program for low-income groups into a unified UHC social security system covering the whole population with a harmonized benefit package. The MoH’s decision to use the exist ing Green Card Program as the key vehicle to achieve UHC allowed it to avoid legislative delays or roadblocks associated with 24 creating an entirely new social security scheme for low-income households. The gradual expansion of benefits and coverage built public support and momentum behind the concept of UHC so that the unified system was politically palatable by the time Green Card beneficiaries were to be integrated with the social security institution. By the time the Green Card Program was replaced by the premium assistance scheme and fully integrated with the social security institution in 2012, political opposition and the prospect of increased expenditure were no longer important issues. These obstacles had already been overcome by the incremental approach to expanding coverage. 25 PART V – CONCLUSIONS 61. We have attempted to shed light on how the Minister of Health and his leadership team addressed political economy conflicts at the heart of Turkey’s comprehensive reforms. Our analysis complements many previous studies of the technical aspects of Turkey’s reform and its results by showing how, political economy challenges are inherent in different aspects of health reform and explicating the approaches used in this case to address them. We believe that although the specific presentation of these challenges does reflect factors particular to Turkey at the time of the reforms, the basic conflicts are likely to arise in many settings because they reflect groups and interests common in societies around the world. Similarly, we believe that the core elements of the solutions employed by the Minister and his leadership team could be adapted to other settings because they reflect forces common to politics in most places. Building support quickly; persuading, neutralizing, or marginalizing opposition groups; asserting domestic priorities in engagements with the IFIs, and carefully sequencing reforms to avoid or limit potential problems are sufficiently general propositions as to be relevant in most democracies. These conclusions, however, do not guarantee success in all cases. For instance, the speed and scope of the Turkish reforms reflects many capacities developed over decades and required substantial monetary resources, even by the standards of the world’s 15th largest economy. Turkey’s reforms were based largely on redistribution and took place in a time of economic expansion, which presents different challenges than does resource creation or reforms undertaken in worse economic conditions. 26 PART VI – IMPLICATIONS 62. Turkey’s current and future health agenda reflects both benefits and consequences of the HTP. All design choices have implications for the future; we discuss them because of their importance for Turkey, and for other countries hoping to glean lessons from the Turkish experience. Based on our stakeholder interviews, we identified four issues with important implications for the future of the Turkish health system. These include the redistribution of resources, financial sustainability, quality of care, and dissatisfaction among medical specialists are paramount to the maintenance and improvement of the health system. 63. Redistribution of resources: The political importance of acting quickly and the technical decision to greatly increase benefits for most citizens led to the practical imperative to redistribute existing resources and also increase total allocations for health. In general, resources were diverted away from elite urban institutions and devoted instead to the provision of health services in secondary cities and rural areas. In the short run, this choice increased the availability of inexpensive, basic services, which have a high return on investment and tend to benefit younger populations, as compared to the high-cost interventions typically required later in the life cycle. The reform also targeted resources towards Green Card beneficiaries, which helped enroll low-income households that had had little or no coverage in the past. 64. Although this redistribution addressed inequities, it appears to have weakened some parts of the health system. For example, the rapid establishment of the family medicine model and a robust primary care system was achieved in part by rechanneling resources. For example, tertiary care centers did not receive their customary allocation of assistant professors because those physicians were assigned to public hospitals in service delivery roles. Unless remedied, the development of fewer highly trained specialists will endanger the quality and availability of tertiary care. Additionally, the pay for performance system encourages the reallocation of time toward service delivery, but does not reward teaching, mentoring, or research—all crucial activities for academic medical centers that provide the most advanced services, develop new techniques, conduct research, and train specialists. As the HTP continues to unfold, resources will be required for research, specialized care, and specialist training. Reinvigorating these areas is essential for the long-term performance of the health system and is also important for Turkey’s goal of becoming a medical tourist destination. 65. Quality of care: The HTP focused in large part on increasing the use of the public healthcare system and introducing private initiatives at the same time. The next phase of health reform will need to focus on quality and safety in healthcare to both maintain demand and continue to improve the health of the Turkish population (Ministry of Health Turkey 2012; Atun, Aydın et al. 2013). The pay for performance system is designed to encourage greater service delivery, but it is not well suited to evaluate the quality of those services. The system should be reevaluated to include more mechanisms to ensure high quality care. Patient feedback systems partially address this concern, but patients are not necessarily able to accurately judge the quality of care they receive. Over- reliance on patient perception as a quality indicator can undermine the quality of services by reducing provider authority and compromising the clinician-patient relationship. 66. Financial sustainability: The financial sustainability of the health system will be an ongoing concern as Turkey’s population ages, the prevalence of chronic diseases increases, and economic growth forecasts weaken. The initial design of the HTP included checks on future health spending, but these have yet to be implemented. HTP planners included way to limit the growth of government spending on health by using a family physician gate keeping system, and offering a basic guaranteed benefit package that could be supplemented with additional insurance, but both of these measures were abandoned as politically unviable. Without these mechanisms, the decision to offer extensive benefits through the General Health Insurance Scheme carries with it long-run cost pressures. Thus far, policymakers have prioritized coverage expansion, improved service delivery, and equity over concerns about cost increases under the HTP. As the demand for higher- cost services rises, the politically difficult task of introducing some form of rationing will be required to permit long-term sustainability. 27 67. The pay for performance system carries additional financial risks due to its potential to promote induced demand. The financial incentives for clinicians encourage the excess provision of medical services, and patients are typically willing recipients because they usually want more care. The HTP set out to increase demand for health services among citizens who previously had not met their needs through the public system. As the health system transitions from focusing on the quantity of services to also emphasize the quality of care, the MoH will need to manage utilization patterns to ensure effective and efficient care-seeking behavior. 68. Dissatisfaction among medical specialists: The expansion of service delivery and increases in utilization both translated to greatly increased workloads for healthcare providers. For physicians satisfied to work in service delivery roles, the pay-for-performance system has yielded generously commensurate pay increases. However, these pay increases have not fully assuaged all provider groups. Many providers do not want the majority of their salaries to be derived from performance bonuses (Tanik 2014). Dissatisfaction remains extremely high among elite specialists and academicians, and tensions between these groups and the MoH are very high regarding professional autonomy, institutional independence, dual practice laws, compensation policies, and other areas. Also, nurses, midwives, pharmacists, and other provider groups have not received pay increases as large as those for physicians, and wage-related dissatisfaction is acute for some of their members. The dissatisfaction among providers —particularly specialists—is an extremely important threat to the integrity and capacities of the health system because it drives top performers into private practice, curtails research, and weakens teaching capacity, all of which can have negative influences on the future because they have the potential to push new college graduates to choose other professions. The MoH will need to work with provider groups to improve relations, and restore cooperation in all areas of the health system. It should carefully monitor the numbers and quality of students choosing health and medical education and where they choose to practice during their careers. 28 REFERENCES Abiiro, G. A. and D. McIntyre. 2013. "Universal Financial Protection through National Health Insurance: A stakeholder Analysis of the Proposed one-time Premium Payment Policy in Ghana." Health Policy Plan 28(3): 263-278. Agyepong, I. and S. Adjei. 2008. "Public Social Policy Development and Implementation: A Case Study of the Ghana National Health Insurance Scheme." Health Policy and Planning 23: 150 - 160. Akdağ, R. 2011. Turkey Health Tranformation Program Evaluation Report 2003-2010. Ministry of Health. Akinci, F., S. Mollahaliloglu, et al. 2012. "Assessment of the Turkish Health Care System Reforms: A Stakeholder Analysis." Health Policy 107(1): 21-30. Akyüz, Y. and K. Boratav. 2003. "The Making of the Turkish Financial Crisis." World Development 31(9): 1549-1566. Alper, C. E. and Z. Ö. Alper. 2003. "Emerging Market Crises and the IMF: Rethinking the Role of the IMF in Light of Turkey's 2000–2001 Financial Crisis." Canadian Journal of Development Studies/Revue canadienne d'études du développement 24(2): 267-284. Aran, M. A. and J. S. Hentschel. 2012. Protection in Good and Bad Times? The Turkish Green Card Health Program. Washington, DC, World Bank. Policy Research Working Paper 6178. Atun, R., S. Aydın, et al. 2013. "Universal Health Coverage in Turkey: Enhancement of Equity." The Lancet 382(9886): 65-99. Baris, E., S. Mollahaliloglu, et al. 2011. "Healthcare in Turkey: from Laggard to Leader." BMJ 342. Bleich, S. N., E. Özaltin, et al. 2009. "How does Satisfaction with the Health-care System Relate to Patient Experience?" Bulletin of the World Health Organization 87(4): 271-278. Boulton, L. and M. Wolf. 2002. Turkey's Ailment: The Country's Chances of Controlling Public Debt and Securing Economic Reecovery could be Jeopardised by the Ill-health of the Prime Minister. Financial Times. London. Brugha, R. and Z. Varvasovszky. 2000. "Stakeholder Analysis: A Review." Health Policy Plan 15(3): 239-246. Bryant, S. 2010. Erdogan Seeks Re-elections as Turkish Healthcare Plan Trumps Headscarves. Bloomberg News. Bump, J. B., M. R. Reich, et al. 2012. "Diarrhoeal Diseases and the Global Health Agenda: Measuring and Changing Priority." Health Policy Plan. Damrongplasit, K. and G. A. Melnick 2009. "Early Results From Thailand's 30 Baht Health Reform: Something To Smile About." Health Aff 28(3): w457-466. Ekman, B., N. T. Liem, et al. (2008). "Health Insurance Reform in Vietnam: A Review of Recent Developments and Future Challenges." Health Policy Plan. 23(4): 252-263. Ertugrul, A. and F. Selcuk. 2001. "A Brief Account of the Turkish Economy, 1980--2000." Russian andEast European Finance and Trade 37(6): 6. Frenk, J., O. Gómez-Dantés, et al. 2009. "The Democratization of Health in Mexico: Financial Innovations for Universal Coverage." Bull World Health Organ 87(7): 542-548. Gil, A., O. Polikina, et al. 2010. "Alcohol Policy in a Russian Region: A Stakeholder Analysis." Eur J Public Health 20(5): 588-594. Gilson, L., E. Erasmus, et al. 2012. "Using Stakeholder Analysis to Support Moves Towards Universal Coverage: Lessons from the SHIELD Project." Health Policy Plan 27 Suppl 1: i64-76. Hacker, J. S. 1998. "The Historical Logic of National Health Insurance: Structure and Sequence in the Development of British, Canadian, and U.S. Medical Policy." Studies in American Political Development 12(01): 57-130. Heper, M. 2003. "The Victory of the Justice and Development Party in Turkey." Mediterranean Politics 8(1): 127-134. Hu, S. 2008. "Universal Coverage and Health Financing from China's Perspective." Bulletin of the World Health Organization 86(11): 819-819. 29 Hughes, D. and S. Leethongdee. 2007. "Universal Coverage In The Land Of Smiles: Lessons From Thailand’s 30 Baht Health Reforms." Health Affairs 26(4): 999-1008. Hurriyet Daily News (2010). Turkey to Halve Number of Green Card Holders by 2012. Hurriyet Daily News. Istanbul. Immergut, E. M. 1990. "Institutions, Veto Points, and Policy Results: A Comparative Analysis of Health Care." Journal of Public Policy 10(4): 391-416. _____. 1992. Health Politics: Interests and Institutions in Western Europe. Cambridge;New York, Cambridge University Press. International Monetary Fund. 2002. Turkey Article IV Consultation. Washington, DC. _____. July 2002. Turkey: Request for Standby Agreement. Washington, DC. IMF Country Report No. 02/136. Johansen, A. and A.-L. Guisset. 2012. Successful Health Systems Reforms: The Case of Turkey. Copenhagen, World Health Organization Europe. Karadeniz, O. 2012. "Extension of Health Services Coverage for Needy in Turkey: From Social Assistance to General Health Insurance." Journal of Social Security 2: 103-123. Knaul, F. M., E. González-Pier, et al. 2012. "The Quest for Universal Health Coverage: Achieving Social Protection for all in Mexico." The Lancet 380(9849): 1259-1279. Lakin, J. 2010. "The End of Insurance: Mexico’s Seguro Popular, 2001 – 2007." Journal of Health Politics Policy and Law 35(3). Logie, D. E., M. Rowson, et al. 2008. "Innovations in Rwanda's Health System: Looking to the Future." The Lancet 372(9634): 256-261. Menon, R., S. Mollahaliloglu, et al. (2013). Towards Univeral Coverage: Turkey's Green Card Program for the Poor. Washington, DC, World Bank. Ministry of Family and Social Policy. 2013. Premium Support Eligibility 2012 and 2013. Government of Turkey. Ankara. Ministry of Health Turkey. 2012. Strategic Plan 2013 - 2017. Ankara. Namazzi, G., N. Kiwanuka, et al. 2013. "Stakeholder Analysis for a Maternal and Newborn Health Project in Eastern Uganda." BMC Pregnancy and Childbirth 13(1): 58. OECD, WHO, et al. 2008. OECD Reviews of Health Systems: Turkey. Paris. Olson, M. 1971. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA, Harvard University Press. Özbudun, E. 2006. "From Political Islam to Conservative Democracy: The Case of the Justice and Development Party in Turkey." South European Society and Politics 11(3-4): 543- 557. Paim, J., C. Travassos, et al. 2011. "The Brazilian Health System: History, Advances, and Challenges." Lancet 377(9779): 1778-1797. Pannarunothai, S., D. Patmasiriwat, et al. 2004. "Universal Health Coverage in Thailand: Ideas for Reform and Policy Struggling." Health Policy 68(1): 17-30. Reich, M. R. 1995. "The Politics of Health Sector Reform in Developing Countries: Three Cases of Pharmaceutical Policy." Health Policy 32(1-3): 47-77. Republic of Turkey and International Bank for Reconstruction and Development. 2004. Loan Agreement: Health Transition Project. Loan number 4737 TU. Roberts, M., W. Hsiao, et al. 2003. Getting Health Reform Right: A Guide to Improving Performance and Equity. New York, Oxford University Press. _____. 2008. Getting Health Reform Right: A Guide to Improving Performance and Equity. New York, Oxford University Press. Rossetti, A. G. 2004. Republic of Turkey Health Transformation Program: Strategic Options for the Implementation of the Health Sector Transformation in Turkey Ankara, Ministry of Health and the World Bank. _____. 2004. Republic of Turkey Health Transformation Program: Strategic Options for the Implementation of the Health Sector Transformation in Turkey. Ankara, Ministry of Health and the World Bank. Sarr, F. 2010. "Efficiency of Immunization Service in the Gambia: Results of a Stakeholder Analysis." East Afr J Public Health 7(1): 68-73. 30 Tangcharoensathien, V., W. Patcharanarumol, et al. 2011. "Health-financing Reforms in Southeast Asia: Challenges in Achieving Universal Coverage." Lancet 377(9768): 863- 873. Tanik, F. A., E. Bilaloğlu, et al. 2013. "The Greatest Motivation: Assurance of Practicing the Profession with Dignity. Motivational State of Physicians in Turkey." World Medical Journal 59(3). Tatar, M., S. Mollahaliloglu, et al. 2011. Turkey Health System Review. Health Systems in Transition. E. O. o. H. S. a. Policy. Turkish Statistical Institute. 2003-2012. Life Satisfaction Survey. Ankara. G. o. Turkey. Varvasovszky, Z. and R. Brugha 2000. "A Stakeholder Analysis." Health Policy Plan 15(3): 338- 345. World Bank. 2006. Turkey Public Expenditure Review. Washington, DC. Report No. 36764-TR. _____. 2012. World Development Indicators. Washington, DC. World Health Organization. 2013. Turkey National Health Accounts - Health Expenditure Series. Yasar, G. Y. 2011. "‘Health Transformation Programme’ in Turkey: An Assessment." The International Journal of Health Planning and Management 26(2): 110-133. 31 ANNEX I: STAKEHOLDER MAPS TABLE 1: GOVERNMENTAL STAKEHOLDERS IN HTP DESIGN AND ADOPTION (2002–4) STAKEHOLDER ROLES OBJECTIVES STRATEGIES GOVERNMENTAL Ministry of  Designed HTP  Improve the delivery of  Make simple, highly visible Health  Leader of inter- medical services with a improvements first to generate ministerial working focus on rural and poor support for more difficult group populations. reforms.  Unify fractured financing  Use political support of Prime and delivery systems. Minister to win budgetary  Convince government to increases over objections of spend more on health, other ministries. reorient system around  Use inter-ministerial working primary care and service group to coordinate objectives, delivery, improve equity, strategies and policies with increase citizen Treasury, Finance, SSK, IMF satisfaction, and build and World Bank. more infrastructure.  Sought to increase electoral support by improving service delivery, reducing opposition by fragmenting physician/health worker groups, and advocating for increased training of health workers. Ministry of Labor  Part of inter-ministerial  Increase retirement age  Use increased health access and Social working group. from 50 (average) to 65 to and benefits as compensation Security (SSI)  Leader of social security reduce burden on to workers for extending component of overall government expenditure. retirement age. reform effort.  Protect existing benefits  Work to exempt existing  Concerned that HTP for social security retirees from new regulations. could increase health beneficiaries.  Apply political pressure to keep expenditure.  Require referral system own hospital system. for secondary or specialist care. Ministry of  Part of inter-ministerial  Manage long-term cost of  Model economics of reform Development working group. reform. and cooperate with (formerly State  Concerned about long-  Accurately forecast international agencies and Planning run expansion of costs growth in expenditure and consultants to incorporate Organization) in a reformed health potential savings from interests into policy. system because of need efficiency gains.  Help guide the inter-ministerial to build more  Increase revenues from working group. infrastructure and more comprehensive deliver more services. enrollment. 32 Ministry of  Part of the inter-  Reform health sector so  Use domestic political power Finance ministerial working expenditures did not and pressure from IMF to limit group. exceed revenues. MoH budget.  Allocated financial  Improve efficiency of resources for the entire government health Turkish government. spending.  Worked closely with  Limit expenditure growth IMF in implementing by requiring referrals for standby agreement. specialist care.  Worked with MoH and  Increase revenues by Ministry of Labor and offering supplementary Social Security to insurance. concurrently reform health and social security systems to improve long-term financial sustainability. Under Secretariat  Part of inter-ministerial  Promote efficiency and  Worked through inter- of Treasury working group. cost control measures to ministerial working group to  Involved in reforming limit long-run spending. forecast costs and plan reform. both health and  Comply with IMF standby  Used IMF concerns as a way to pension system to agreement. voice cost control agenda. control spending.  Special projects unit working on health reform for over a decade. Prime Minister:  Controlled  Increase equity and access  Support MoH over other Abdullah Gül governmental agenda to health services to make ministries in negotiations with (2002-2003)  Introduced social sector good on campaign IMF, World Bank and inter- Prime Minister: reforms aligned with AK pledges. ministerial working group. Recep Tayyip Party agenda.  Solidify electoral success Erdoğan (2003 of AK Party by increasing March – Present) voter satisfaction. Parliament  Controlled by AK Party  Support Prime Minister  Use normal legislative (Health and worked to and Minister of Health to processes to support reform Commission, implement its agenda deliver more health when parliamentary approval Budget and and policy initiatives. services and as a result was needed (primarily as of Planning increase electoral 2006). Commission) satisfaction and gain  Submitted legislation for reelection. approval to implement HTP. President Ahmet  Concerned that reform  Maintain benefits for civil  Delay reform by sending it to Necdet Sezer would diminish benefits servants. Constitutional Court (2006). for civil servants. Constitutional  Did not support AK  Preserve traditional  Use institutional authority of Court Party initiatives. secular elite power. the court to block parts of reform and delay implementation. Strongly Supportive Neutral Opposed Strongly supportive opposed 33 TABLE 2: HEALTH CARE PROVIDER STAKEHOLDERS IN HTP DESIGN AND ADOPTION (2002–4) STAKEHOLDER ROLES OBJECTIVES STRATEGIES PROFESSIONAL ASSOCIATIONS AND INTEREST GROUPS Turkish  Represented Turkish  Supported objectives of  Leadership attempted to Medical physicians as a private expanding access to services. influence reform in dialogue Association sector licensing  Leadership concerned with with MoH. requirement. preserving autonomy due to mistrust of AK Party agenda.  Did not want to see a negative impact on income, quality of services, autonomy of profession, or decline in research and teaching capacity in elite universities.  Leadership wanted to preserve the status quo. Dentists  Represented all dentists  Sought to increase income and  Cooperated with the MoH in in the country. availability of dental care in designing the dental public system. component of the HTP.  Wanted to increase insurance coverage for dental services and increase business for dentists. Pharmacists  Represented all  Wanted to preserve autonomy  Resisted MoH efforts. pharmacists, who were and income level.  Sought to increase benefits all independent for medicines and practitioners. reimbursement rates from MoH. Turkish  Represented all nurses  Wanted to improve stature and  Worked with Turkish Nurses in the country. income. Medical Association to Association  Did not want to lose autonomy. influence MoH plans.  Feared that privatization would take place under the HTP insurance model and instead wanted a National Health Service model. Midwives  Represented all  Promoted midwifery as a central  Attempted to influence MoH midwives in the element in home-based care and through dialogue. country. community-level outreach.  Maintained primary role  Wanted to have a central role in in childbirth, particularly family physician care model. in rural areas. ACADEMIC Higher  In charge of allocating  Did not want to cooperate with  Attempt to outlast AK Party Education funds for university AK Party or join its initiatives. influence by delaying action Council hospitals and approving  Preserve traditional, secular elite in hopes of future electoral (YOK) new training power. results. opportunities for health  Resist perceived Islamist workers. influence.  Aligned with opposition groups to the AK Party. Medical  Primary providers of  Preserve existing system.  Attempt to influence the faculties of high-end, specialty care.  Maintain autonomy and high- reform and mitigate its elite  Trained health workers. end specialty care provision and impact. universities  Aligned with opposition training.  Hope to outlast AK Party groups to AK Party. influence by delaying action and diminishing AK Party influence in future elections. 34 Medical  Stood to benefit from  Gain financial, infrastructure and  Cooperated and worked faculties of reform, but unsure of training resources. with MoH to support other consequences for  Increase prominence of reforms and allocate university themselves. institutions and staff. resources to their universities. Strongly Supportive Neutral Opposed Strongly supportive opposed 35 TABLE 3: ORGANIZED BENEFICIARY GROUPS IN HTP DESIGN AND ADOPTION (2002–4) STAKEHOLDER ROLES OBJECTIVES STRATEGIES Active civil servants  Important  Preserve existing benefits  Negotiated that new social representation in policy and retirement age. security age limit would making entities. only apply to new hires as  Historically aligned with of 2008. opposition. Blue and white  Large share of pop.  Wanted to increase  Did not organize collar workers and represented access, benefits and independently of SSI. (received benefits majority of formal improve health services. through SSK) sector workers. Retired civil  Strong voting block  Maintain existing benefits.  Lobbied for exemption from servants (received HTP provisions. benefits through  Ensure support of President Emekli Sandıği ) to make sure they would not have benefits reduced. Self-employed and  Lack of enrollment in  Increase access to services  Did not organize artisans (eligible for Bağ-Kur meant the and increase benefit independently. benefits through MoH needed increased package. Bağ-Kur) enrollment amongst  Ensure financial this population. protection. Labor Union for  Large voting bloc.  Wanted to preserve and  Cooperated with the MoH. Transportation and augment benefits. Construction Workers Labor union for  Focus on public sector  Preserve the status quo.  Cooperated with the MoH private health delivery meant they and promoted beneficial workers (Saglik-Is were not the targeted contracting arrangements. under Turk-Is) group of the HTP.  Contracted with MoH to augment services. Labor union for  Frontline implementers  Wanted increased pay and  Formed to represent public health of reform. improved terms of work. workers who supported the workers (Saglik-Sen  Support AK Party. reform, thereby reducing under Memur-Sen) the membership of opposing unions and professional associations. General Public  Voters overwhelming  Urban elites didn’t  Express electoral support by supported support AK Party voting for AK Party.  AK Party in 2002 initiatives and wanted parliamentary them voted out of office in elections. future elections.  Ability to change  Majority of population leadership in future supported reform to the elections. social sectors and improvements in health service delivery.  Wanted to obtain more benefits and improve quality of health system. Strongly supportive Supportive Neutral Opposed Strongly opposed 36 TABLE 4: INTERNATIONAL ACTOR STAKEHOLDERS IN HTP DESIGN AND ADOPTION (2002–4) STAKEHOLDER ROLES OBJECTIVES STRATEGIES World Bank  Provided funding and  Support Turkish Government  Support MoH and inter- technical assistance for efforts to design and ministerial working group the HTP. implement a technically sound with funding and reform. technical assistance.  Promote UHC as part of the reform.  Gain access to the reform process. IMF  Provided Turkish  Opposed to any increase in  Provided standby Government with expenditures. agreements and funding. standby agreement and  Reduce government debt and  Negotiate with ministries bailout funds after recent stabilize economy. and exert pressure on financial crises.  Improve efficiency of government actors to  Oversaw expenditure government spending and constrain spending. reform plans of the constrain growth in health Turkish Government. spending. World Health  Assisted government  Improve the health of Turkish  Respond to requests for Organization with health programs. population and extend basic technical assistance and services to underserved financing of health populations. services. EU  Negotiating with Turkey  Wanted Turkey to improve its  Used EU membership to on accession into EU. health system to continue on require changes in its accession process. Turkey’s social sectors.  Focus on expansion of primary care system. Strongly Supportive Neutral Opposed Strongly opposed supportive 37 TABLE 5: GOVERNMENTAL STAKEHOLDERS IN GREEN CARD EXPANSION (2003–2012) STAKEHOLDER ROLES OBJECTIVES STRATEGIES Ministry of Health  Leader of inter-  Increase benefits and number  The Minister was careful ministerial group. of enrollees. not to think of Green Card  Leader and designer  Merge Green Card Program enrollees as a burden on of reform effort. with social security institutions. the system and rather, stressed their right to basic health services.  As citizen satisfaction grew for the reforms, it was increasingly difficult for those concerned about the cost of the program to openly oppose it.  Ministry pushed to provide as many Green Cards as possible to citizenry. Ministry of Labor  Part of inter-  Become primary purchaser of  Difficulties in merging and Social Security ministerial group. health services for all Turkish Green Card with other (SSI)  Social Security population. social security schemes Institution to be  Concerned about cost of and in determining housed under its integrating Green Card eligibility. auspices and to act beneficiaries at higher coverage  Needed to figure out as purchaser in the levels and rates. mechanism for general tax system. revenue to pay for Green  Green Card Card premiums. beneficiaries to be  Delayed integration of merged into Social Green Card Program with Security Institution. SSI. Ministry of  Newly formed in  Implement an effective and  Quickly built IT capacity to Family and Social 2011 and previously rigorous income testing tool. determine eligibility and Policy was Social Solidarity regain control over this Foundations. process.  Had been in charge of Green Card Program prior to 2004 and took over eligibility determination in 2012. Local Committees  Responsible for  Maintain political power in  Kept ability to give out ultimately giving out Green Cards. Green Cards, but given determining  Give out as many Green Cards scrutiny had to be more eligibility for Green as possible. transparent about it. Card Program until  Huge push to give out 2012. Green Cards allowed them to provide maximize benefits given out to constituencies. 38 Ministry of  Part of inter-  Manage long-term cost of  Model economics of Development ministerial working reform. reform and cooperate with (formerly State group.  Accurately forecast growth in international agencies and Planning  Concerned about expenditure and potential consultants to incorporate Organization) long-run expansion savings from efficiency gains. interests into policy. of costs in a  Increase revenues from more  Help guide the inter- reformed health comprehensive enrollment. ministerial working group. system because of need to build more infrastructure and deliver more services. SSI  Under pressure to  In favor of integrating Green  Delayed integration of integrate Green Card Card scheme and having a Green Card Program with holders under unified system. SSI. General Health  However, ability to determine  Eventually allowed Insurance Scheme. eligibility and merge Ministry of Family and participants was a challenge. Social Policy to take over eligibility responsibility and then received premiums from Ministry of Finance derived from general tax revenues. Ministry of Finance  Part of the inter-  Understood political  Pushed for concurrent ministerial working importance of expanding the efficiency improving group. Green Card Program, but measures that would help  Had to allocate concerned about fiscal impact. offset the costs of funds for health increased benefits for sector. Green Card holders. These included a referral chain and supplementary insurance with a basic benefit package that did not go through. Under Secretariat  Part of inter-  Understood political  Stressed importance of of Treasury ministerial working importance of expanding the efficiency measures along group. Green Card Program, but with coverage expansion.  Had to allocate funds concerned about fiscal impact. for health sector. AK Party  Had to approve all  Saw the number of votes  Supported increased benefits parliamentarians health reform they could garner and numbers of Green Cards, legislation. through increased Green particularly in advance of Card enrollment and elections. benefits. Opposition  Ideologically against  Against overall reform  Did not want Green Card votes parliamentarians the AK Party but difficult to go against to go to AK Party. government. general message of the reforms. Strongly Supportive Neutral Opposed Strongly opposed supportive 39 TABLE 6: NON-GOVERNMENTAL STAKEHOLDERS IN GREEN CARD EXPANSION (2003–2012) STAKEHOLDER ROLES OBJECTIVES STRATEGIES BENEFICIARY GROUPS Green Card eligible  Highly dissatisfied  Wanted expanded  Enrolled in Green Card Program in with current benefits, benefits and more greater numbers and voted for quality of health transparent system. the AK Party. services and level of  Greatly expanded financial protection in benefits and improved health system. health service delivery  Needed to voluntarily system were important enroll in system to selling points. achieve UHC.  Supported greater  Important voting bloc scrutiny, so there was for AK Party. less ability for the local committees to give card based on political favor. Green Card  Highly dissatisfied  Wanted improved  Voted for the AK Party. Those that enrollees with current benefits, services and expanded were enrolled that were not quality of health benefits. technically eligible could continue services and level of  Greatly expanded to game the system until merging financial protection in benefits and improved under General Health Insurance health system. health service delivery Scheme.  Needed this system were important population to have selling points. improved access to  Those who were enrolled quality services to but not technically achieve better health eligible based on income outcomes. criteria would eventually  Important voting bloc lose out on benefits. for AK Party. Non-eligible Green  Benefitted from lack  Do not want to lose  Continued to use Green Card Card enrollees of control of Green benefits of Green Card benefit, which contributed to Card Program and may have to pay delays in integrating it with enrollment. premiums for social General Health Insurance Scheme. security institution. Uninsured  Cannot take  Do not want to pay  Some took advantage of population advantage of illegal health insurance expanded Green Card program use of Green Cards. premium but do want and became beneficiaries.  Were now pressured improved access to  Others were induced to join SSI. to join social security quality health services.  A small percentage remains institution and pay uninsured and has chosen to premiums. remain without coverage. Social security  Large share of  Did not want to pay for  They were appeased as long as beneficiaries population that could expanded they did not have to sacrifice any block the reform if benefits/enrollees or see benefits and instead were given they were in their benefits diminished more benefits and better access organized and vocal in any way. to health services. opposition.  They negotiated for improvements in their own benefits instead of working against the Green Card Program expansion. 40 High-income  Small share of the  Do not want to have to  Remained in opposition parties without public population that was dedicate additional tax and continued to pay for private insurance primarily in revenue to pay for healthcare services. opposition to AK increased benefits for Party. Green Card enrollees. HEALTH PROFESSIONALS MOH hospital  Had workload  Want increased pay and  Increasing numbers worked for physicians increased due to ability to have dual MoH. expanded coverage. practice.  Took advantage of increased pay  Hoped that more  Wanted better access in family medicine practice and Green Card patients and quality of services other P4P oriented profession. would go to family for Green Card holders.  Lobbied against parts of reforms physician clinics with  Did not want increased that increased their workload and outpatient coverage workload. did not allow for dual practice. inclusion.  Supported referral requirement. MOH general  Central role in  Wanted increased pay.  Many became family practitioners practitioners implementing  Supported efforts to and benefitted from P4P scheme. primary care focus of promote preventive and  Lobbied against increased the reform. primary care. workload has placed a large  Had to meet the burden. increased demand for outpatient services due to expanded benefits coverage. SSI hospital  Became part of the  Against not being able  Increasing numbers worked for physicians MOH and in favor of to have dual practice. MoH. increased pay from  Now had to treat both  Took advantage of increased pay in P4P. social security and family medicine practice and other Green Card patients and P4P oriented profession. wanted increased pay.  Lobbied against reforms. Private physicians  Still able to see  Did not want to see a  Continue charging patients OOP in patients who are able decline in their paying the private sector. to pay. patients because more  Start contracting with Social coverage for Green Card Security Institution and charging holders means that extra payments for services patients are channeled provided to social security into public sector. beneficiaries.  In favor of not allowing dual practice because then there is less competition. 41 Nurses and  Greater focus on  Approved of increased  Continued to play important role in midwives family medicine benefits and improved healthcare and entered into physicians meant a access to quality health dialogue to make sure community diminished role for services for lower care and home based care were nurses. income households. part of primary healthcare system.  Nurses also not part  Wanted an active and of P4P scheme central role in the new directly and are part healthcare system, of pooled funds. including P4P.  Increased focus on facility births and role of midwives diminished.  In family medicine clinic – now choose a nurse, midwife or health assistant to work with physician. INTERNATIONAL ACTORS World Bank  Had worked with  Recognized need to  Provided technical assistance to Government since expand coverage and the MoH, in addition to loans to early 1990s to benefits of Green Card design and implement the HTP. propose UHC. Program.  Conducted in depth modeling  Supported using it as the exercises to forecast various vehicle to achieve UHC. benefit package and coverage expansion scenarios. IMF  As a creditor to  Wanted to control  Continual negotiations and Turkish Government, government spending pressure on MoH and other was concerned about and decrease government actors to reduce fiscal costs of government debt. health spending. expanded coverage. EU  Working with Turkey  As part of accession  Used public support for EU for it to become a negotiations, EU was accession and its negotiating member of the EU. pushing Turkey to power to push for primary improve its health care healthcare agenda and universal system, expand coverage health coverage. and focus on primary health care benefits. Strongly supportive Supportive Neutral Opposed Strongly opposed 42 ANNEX II: INTERVIEW GUIDE 1. Explain project a. Asked by MoH and WB to do P-E analysis of health reform i. Fast explanation of P-E b. Our focus on some general policy design and then more on policy adoption, Green Card implementation c. Why? Important for Turkey, LMICs, facilitators 2. Permissions a. Can we take notes? b. How we will use information c. How we handle quotes 3. Personal history a. Training, positions, background b. What brought interviewee in contact with the reform? When? What capacity? 4. What conditions led to reform? a. Why 2003? b. What was the impetus? c. What factors coalesced? 5. What was the policy design process? a. How were the components chosen? b. How were they sequenced? c. What groups were involved and/or affected? i. What were expected positions? ii. Strategies were designed to win support? d. Why was the green card program chosen as the primary vehicle for delivering more services? 6. Policy adoption a. Was the decision to adopt executive or legislative? i. What parts went through the legislature? ii. What parts were executed by MoH authority? b. What was the legislative process? c. Who were the stakeholders? d. What strategies were used to build support and overcome opposition? 7. Green Card Program a. How did the MoH increase enrollment in to the GC so quickly? b. How did enrollment procedures change? i. How were these rules communicated? ii. How were they enforced? c. How did the central MoH work with local committees to expand enrollment? i. What were reporting and enforcement processes? d. Was there any variation in enrollment rates by province? e. How were citizens persuaded to enroll in the voluntary system? f. Who were the stakeholders and what were their positions? i. What were the strategies used to overcome opposition g. How did the MoH plan for the additional services expanded enrollment would require? 43 ANNEX III: STATISTICAL ANNEX Table 1: Health Spending in Turkey, 2002–2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Total health spending as share 5.4 5.3 5.4 5.4 5.8 6.0 6.1 6.1 5.6 5.3 5.2 of GDP (%) Public health spending as share 3.8 3.8 3.8 3.7 4.0 4.1 4.4 4.9 4.4 4.2 4.1 of GDP (%) Private health spending as 1.6 1.5 1.5 1.8 1.8 1.9 1.6 1.1 1.2 1.1 1.1 share of GDP (%) Public health spending as share 70.7 71.9 71.2 67.8 68.3 67.8 73.0 81.0 78.6 79.6 79.2 of total health spending (%) Public health spending as share 9.1 9.7 10.8 11.2 11.9 12.1 12.8 12.3 11.5 11.4 10.7 of government expenditure (%) Out-of-pocket spending as 19.8 18.5 19.2 22.8 22.0 21.8 17.4 14.1 16.3 15.4 15.8 share of total health spending (%) Total health spending per 466 471 547 622 738 813 913 874 889 901 937 capita, PPP (constant international $ for the year) Source: World Bank, World Development Indicators 2014, TURKSTAT 2014 44 Table 2: Use of Health Services Analyzed by Region, Socioeconomic Groups, and Health Insurance (1998-2003) Antenatal visit (%) Proportion of births in a health Proportion of births attended by Immunisation uptake (all) (%) facility (%) trained staff (%) 1993 1998 2003 2008 1993 1998 2003 2008 1993 1998 2003 2008 1993 1998 2003 2008 Location Urban 73·2 74·4 83·6 96·4 73 76·4 84·2 91·8 80·8 81·9 85·1 94·1 74·8 49·5 62·6 78·4 Rural 48·8 51·2 56·8 87·7 42·2 61·0 60·9 74·1 52·8 67·6 62·2 76·2 52·3 38·3 34·0 63·8 Region West 85·4 83·7 89·7 97·2 80·4 86·7 92·0 93·4 88·0 89·2 90·4 95·8 76·5 50·3 64·3 79·3 South 75·5 73·6 83·4 95·9 64·0 69·3 79·0 90·1 75·9 84·2 83·4 90·5 81·1 56·8 63·2 75·5 Central 59·4 71·8 82·1 95·6 64·2 82·6 90·1 94·2 72·2 85·6 91·0 96·1 64·9 52·5 59·7 82·1 North 63·1 58·5 82·5 93·3 66·8 77·9 87·3 92·2 71·6 83·1 81·6 91·1 62·5 60·0 60·5 77·3 East 35·1 43·3 54·1 84·0 30·3 47·8 54·1 61·8 41·4 53·4 57·1 65·8 41·6 25·1 35·3 51·9 Mother’s education No 36·9 36·0 45·0 81·7 32·2 41·3 42·7 61·1 42·7 49·8 47·1 65·8 47·8 31·6 22·5 58·0 education Primary 69·7 70·1 78·4 94·1 68·1 75·9 83·6 89·3 76·6 82·3 85·0 91·2 69·9 45·1 59·9 75·5 education Secondar 89·9 91·1 96·2 99·0 86·5 92·5 94·6 93·5 93·5 94·0 91·6 95·1 81·9 61·2 65·3 78·1 y education or higher Mother tongue Non- 39·0 43·9 53·2 86·7 27·7 42·6 50·0 66·8 39·3 51·9 53·9 71·2 37·3 25·7 33·1 59·6 Turkish Turkish 70·0 75·9 86·0 96·4 69·6 83·4 90·4 94·0 77·7 87·9 90·0 95·4 72·2 53·6 62·9 79·3 Asset quintile 1 36·1 36·8 44·9 84·1 33·0 45·7 50·4 67·1 43·1 52·2 53·0 70·4 48·0 30·3 29·0 54·9 5 90·9 94·8 97·0 99·0 90·0 95·1 96·9 92·4 94·0 93·2 93·8 94·6 84·4 70·4 71·1 79·0 Mother’s health insurance SIO 77·0 83·1 91·4 97·0 76·8 85·5 92·6 92·3 83·7 89·5 91·1 93·8 76·4 55·1 67·5 78·7 Green 29·7 55·6 56·6 87·4 37·8 67·3 63·0 72·2 56·8 71·7 63·0 77·1 20·0 32·7 38·8 60·0 Card None 52·8 55·5 64·3 90·0 47·7 61·1 64·7 83·2 57·9 68·4 68·8 83·5 56·2 41·1 42·5 72·8 Total 63·0 66·1 74·6 93·4 60·1 70·9 76·3 85·7 69·0 76·8 77·4 88·0 65·0 45·3 53·2 73·7 Atun et al 2013 analysis of data from the Turkish Demographic and Health Survey 1993, 1998, 2003, and 2008 (references 12–15 in appendix and appendix pp 2–13). Sample consists of children younger than 5 years for the first three analyses, and children between 12 months old and 24 months of age for immunisation. A fully immunised child has received the following vaccines: BCG; diphtheria– tetanus–pertussis 1, 2, and 3; polio 1, 2, and 3; and measles. SIO=Social Insurance Organisation. Source: Atun et al 2013, p. 20 45 Table 3: Turkey Mortality Indicators, 2002–2012 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Maternal mortality ratio 28 20 (modeled estimate, per 100,000 live births) Mortality rate, infant (per 1,000 live births) 26.1 24.1 22.1 20.3 18.5 17 15.6 14.5 13.6 12.9 12.2 Mortality rate, under-5 (per 1,000 live births) 31.5 28.8 26.2 23.9 21.8 19.8 18.2 16.9 15.8 15 14.2 Source: World Bank, World Development Indicators 2014 Figure 1: Satisfaction Rate with Public Services in Turkey, 2003–2012 Source: Turkish Life Satisfaction Survey, 2003–2012 46 Beginning in 2003, Turkey initiated a series of reforms under the Health Transformation Program (HTP) that over the past decade have reshaped the health system. Understanding the political economy of this process is important for the future of Universal Health Coverage (UHC) in Turkey, and also for many other countries and the development agencies that assist them. This report analyzes the historical context and complex political economy challenges of the reform. Our findings are based on stakeholder interviews and a review of literature. First, we identified five contextual factors that were important in bringing health reform to the policy agenda in Turkey, and were helpful in sustaining the reform during adoption and implementation: (1) a long history of reform plans and attempts; (2) fiscal pressure to reform the social sectors; (3) public support for health reform; (4) strong economic growth; and (5) favorable demographic conditions. Second, we assessed four political economy challenges central to the reform and the strategies used by the Ministry of Health (MoH) to overcome them. First, the MoH built public support for reform among the broad base of beneficiaries by focusing on highly visible and fast changes. Second, the MoH overcame well-organized interest group opposition to the reforms by splintering their support or delegitimizing their views. Third, Turkey asserted its own domestic priorities over those of the IMF and World Bank in cases of direct conflict. Fourth, the MoH circumvented potential political and institutional opposition to the large expansion of benefits and coverage through a carefully sequenced adoption and implementation plan that could be executed mostly without requiring the support of other ministries. This analysis also highlights important trade-offs made by the MoH with respect to the redistribution of resources, quality of care, financial sustainability, and physician satisfaction, which will all have to be considered as Turkey enters its next phase of health system development. ABOUT THIS SERIES: This series is produced by the Health, Nutrition, and Population Global Practice of the World Bank. The papers in this series aim to provide a vehicle for publishing preliminary results on HNP topics to encourage discussion and debate. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent. Citation and the use of material presented in this series should take into account this provisional character. For free copies of papers in this series please contact the individual author/s whose name appears on the paper. Enquiries about the series and submissions should be made directly to the Editor Martin Lutalo (mlutalo@ worldbank.org) or HNP Advisory Service (healthpop@worldbank.org, tel 202 473-2256). For more information, see also www.worldbank.org/hnppublications. 1818 H Street, NW Washington, DC USA 20433 Telephone: 202 473 1000 Facsimile: 202 477 6391 Internet: www.worldbank.org E-mail: feedback@worldbank.org