As in many low- and middle-income countries, out-of-pocket payments by households account for a large share of health care spending in Lao PDR. These payments can deter use of services and increase the risk of financial catastrophe and impoverishment. Consequently, the Government of Laos is attempting to expand health insurance and . risk-protection coverage through four different schemes. This thesis examined two of those schemes: community-based health insurance (CBHI) and social health insurance (SHI). Using a conceptual framework that was developed based on the theoretical and empirical literature, three sub-studies were designed and implemented to explore: the determinants of household enrolment in CBHI; the determinants of enrolment of firms in SHI; and the impacts of CBHI enrolment on utilisation and financial protection. Data for the CBHI studies were collected using household and village surveys with 3000 households (14,804 individuals) in 87 villages, and six focus group discussions with members and non-members. In the SHI study, a survey was administered to 130 private firms. The CBHI and SHI studies employed a cross-sectional, case-comparison design and used a variety of econometric and qualitative methods in the analysis, including propensity score matching. The findings from the two enrolment studies identified various factors that drive and hinder enrolment in health insurance. The impact evaluation showed that CBHI had a positive effect on utilisation and financial protection, but given the low coverage of the scheme and low utilisation, the impacts on a population level are negligible. Moreover, the poor are the least likely to enrol and the poor who are enrolled incur higher out-ofpocket expenditures than the uninsured. The policy implications for Laos are discussed in the context of the international debate regarding the potential contributions of CBHI and SHI in national health financing strategies as countries progress towards universal coverage.

Health insurance arrangements developed in various social settings as a means of pooling health risks and health resources in order to protect members' income against unpredictable health costs but also in order to guarantee their access to health care. Problems of unregulated health insurance markets, like adverse selection and risk selection, are frequently discussed in academic and political circles in the context of either inefficiency or inequity. Though interest in regulation as a health sector reform instrument is growing, empirical studies of unregulated health insurance markets are still rare, particularly, in low and middle-income country settings. This thesis contributes to the body of research and literature that attempts to identify empirical evidence for adverse selection and risk selection. It aims to examine the following research question: Are unregulated health insurance markets characterised by adverse selection and/ or risk selection and do they thereby create inefficiency or inequity. The objective is to demonstrate empirically whether or not these markets experience selection processes. First, this thesis derives a group method for empirical investigations into adverse selection and risk selection from which testable hypotheses can be derived. Second, this method is applied to case study data from a middle-income country. Longitudinal panel data is analysed, describing South Africa's health insurance market of medical schemes in the context of its post-deregulation experience over the four-year period 1995-1998, after premia risk-rating was legalised. The interpretation of the empirical results leads to three main findings. First, intense competition in the contested health insurance market causes favourable risk selection of low risks into and out of medical scheme plans. Second, unfavourable selection by medical scheme plans in the form of dumping high risks can be concluded. Third, there is no evidence for adverse selection and the typical adverse selection cycle cannot be observed. Exploring the policy relevance of the results, it is concluded that the effects of less health insurance regulation, in the context of middle-income country health sector reforms, conflicts with the common health policy objective of equity. More competition and efficiency comes at the price of less equity in health care access for the poor and sick, confirming the known efficiency-equity trade-off.

This thesis focuses on the policy-making process of the National Health Insurance scheme (NHI) in the Republic of Korea (Korea). The analysis of the policy process of the Korean NHI scheme also makes it possible to observe the development of social policy in Korea. Health care in Korea was basically provided through the market until the implementation of the NHI scheme in 1977. The health care programme was initially introduced for a restricted section of the workforce, but gradually the programme was expanded to cover the entire population. The study addresses the questions of why and how has the NHI scheme developed. The policy-making process of the NHI scheme in Korea can be explained better by socio-political elements than by economic factors. Policy-making in the 1960s and 1970s was carried out by a limited number of policy-makers within a confined policy-making institution. At the beginning of the 1980s, however, the government pursued a more explicit strategy of reform. Since then, the range of the participants embedded in the policy-making arena has gradually become diverse and complex. As democratic processes became stronger, the policy-making structure became dynamically transformed, and power in the process was distributed among various social actors in the society. The economic crisis at the end of the 1990s had a significant impact on the style and structure of policy-making. There was a greater involvement of civic and interest groups in the policy-making process, and the government was less able to take any unilateral policy decisions. The policy-making process of the NHI scheme over the past four decades led to the development of the reformist and anti-reformist groups, and these groups contributed to building ideological foundations not only for the NHI development but also for social policy development in Korea. Two distinctive features were identified as one of the many by-products created by the NHI policy process. First, the policy-making style in the health care policy developed from 'authoritarian leadership' to 'pluralist and corporatist styles'; second, citizenship has been developed in the society and has influenced the policy-making process.

The health of migrants has attracted increasing attention in the international policy dialogue in recent years. Thailand is one of many countries where migrant health is a major political issue. This is because the country is situated at the centre of the Indochinese Peninsula and its economy is fast-growing relative to its neighbouring countries, particularly Cambodia, Lao PDR, and Myanmar. As a result, Thailand has, for decades, attracted a large number of low-skilled cross-border migrants. The majority of these immigrants have passed the border without any valid travel document. However, most of the time, past governments did not impose strict deportation measures on these undocumented/illegal immigrants since they were considered a key contributor to the Thai economy. Measures often used by recent governments included granting them leniency for temporary residence, issuing work permits for certain jobs, and insuring them through public-oriented health insurance, namely, the 'Health Insurance Card Scheme' (HICS). The primary aim of this thesis is to evaluate (i) the enrolment of cross-border migrants in a public health insurance scheme, namely, the HICS, in Thailand through the viewpoints of various stakeholders, and (ii) the effects of insurance on use of services. Ranong province was selected as the study site since it had the largest proportion of migrants compared to other provinces. The main objectives are: (1) to explore how the HICS evolved over time in light of changes in surrounding policies, (2) to investigate the responses of local officers and relevant stakeholders towards the HICS and to examine how the policy affects migrants' health-seeking behaviour in practice, (3) to evaluate the outcomes of HICS in terms of utilisation numbers and financial implications for its insurees, and (4) to provide policy recommendations. A multimethods approach was employed. In-depth interviews, document review and facilitybased data analysis were undertaken. Policy makers, local healthcare providers, and migrants were interviewed. Thematic and analyses were applied. 4 The findings revealed conflicting ministerial objectives and gaps in both inter- and intraministerial policies. In addition, policy objectives were not clear from the outset. While the health sector aimed to insure ‘all’ migrants, this was constrained by the security and economic authorities where the focus was mainly only on migrant workers who registered with the government. Besides, in reality, the boundary between ‘legal’ and ‘illegal’ migrants was very fluid. Though the current government attempted to address policy gaps by overhauling the HICS and instigating a new measure, namely, 'One Stop Service', it is difficult to claim that the deep-rooted implementation problems were resolved. This situation was even more complicated at the local level as some frontline health officers adapted the policy in various ways, and occasionally made the policy diverge from its initial objectives. For users, the cost of registration was a significant barrier in obtaining the insurance card, and a reliance on private intermediaries (both legal and illegal) to help them obtain the insurance card was not uncommon. Besides, there were migrants who were neither insured, nor able to return to their home country. However, the HICS still had some merits in reducing out-of-pocket payment, and helping increase utilisation of services amongst insurees. It was noteworthy that the most important factor determining the number of visits was history of experiencing catastrophic illness, not insurance status, and this influence was even more apparent in Thai patients than in migrants. Evidence suggested that there might be insured migrants with catastrophic illness who still experienced difficulties in accessing services, let alone uninsured migrants. Unless policies to protect the health of this population are put in place, poor access to health services for the uninsured will continue being a serious public health problem, not only to migrant communities but also to Thai society as a whole. Both macro- and micro policy recommendations are provided, for example, integrating the different authorities’ information systems on migrants, amending some outdated laws and regulations, and strengthening the capacity of the insurance governing body.

This thesis consists of 3 chapters plus an introductory chapter and a concluding chapter. They are on three different topics, but they are all related to China’s health insurance system from 2000 to 2011. Chapter 1 is the introduction to the thesis, providing background to the Chinese insurance system, the theoretical underpinning of the three chapters, a description of the datasets used in the thesis, and an overview of the thesis. Chapter 2 investigates whether there is adverse or advantageous selection in China’s private health insurance market before 2003. We found evidence in favour of adverse selection in a pure private insurance market. For the public insurance group where people already got covered by a public insurance but face the choice of buying a supplementary private insurance, we found advantageous selection. Chapter 3 examines whether implementing nearly universal coverage in 2009 led to a decrease in individual preventive behaviour prior to illness, termed ex-ante moral hazard. We exploit the longitudinal dimension of data from 2006 and 2009 and use Coarsened Exact Matching methods. The results do not provide strong evidence for ex-ante moral hazard. Chapter 4 aims at evaluating whether there is ex-post moral hazard after the introduction of universal coverage. We measured ex-post moral hazard as the impact of co-payment rate on treatment cost, to assess the variation of total medical expenditure to patients due to the decrease of price. We conclude that there is ex-post moral hazard in outpatient services after the reform of universal coverage in China. Chapter 5 is the concluding chapter, including a summary of the findings, policy implications, strength and limitations of the thesis, and challenges for future research.

Why did South Korea integrate multiple health insurers into a single national health insurance in 2003 while Japan maintained its fragmented insurance system based on labour market status? Why did labour in South Korea support the integration of health insurance schemes whilst labour in Japan was opposed to it? The health insurance systems in Japan and South Korea were both based on the social insurance system and fragmented on the basis of occupation and labour market status. However, these two countries have taken different reform paths. This thesis argues that the two self-undermining effects and ideas were interwined and these led to different policy coalitions. Firstly, workers’ support for the consolidation reform was dependent on the inclusivity of the decision-making process at company-level health insurance schemes. Labour in Korea was not able to take part in the decision-making process in company-based health insurance societies while Japanese workers were. The absence of self-governance in the Korean health insurance system reduced incentives for the labour unions to protect their health schemes. Secondly, the Korean government conferred small credibility to support for the municipal health insurance. The subsidy for municipal health schemes in Korea was provided at the discretion of the central government and local government had no legal responsibility for its municipal health funds. These regulations were in stark contrast to the Japanese regulations. It made the friction with the idea of universal health care in Korea. Thirdly, the socially oriented unionism and dense network between trade unions and reformers in Korea contributed to the integration of the health insurance system through creating intensive policy learning for solidarity inside labour movements. In contrast, the cooperative labour-management relationship and their strong networks in the Japanese healthcare policy arena led to the coalition to protect their occupational health funds.

International experience shows that achieving universal coverage has been an important way to ensure equity of access to health care and to protect people from bankruptcy due to severe illness. One common approach to universal coverage has been to expand public health insurance to cover all people in a country. In Thailand, universal coverage of health care in Thailand was achieved by expanding public insurance to the uninsured. Before universal coverage, there were two main public insurance schemes, the Civil Servant Medical Benefit Scheme (CSMBS) and the Social Security Scheme (SSS). Merging two other schemes targeted at lower income groups, (the low income card scheme and health card scheme), and adding the uninsured population, produced the Universal Coverage Scheme (UC), a third and much larger scheme. The three schemes differ in a number of ways including funding, payment of provider and benefit package. There has been considerable concern that these characteristics might affect the performance of the insurance schemes. The aim of this study was to evaluate the three public health insurance schemes in terms of their performance in selected areas. The first objective was to assess and explain variation in performance in terms of utilization, length of stay (LOS), and early readmission. The second was to identify the quality of care provided in each Insurance scheme using Diabetes Mellitus (OM) as a tracer of performance and examining LOS, early readmission, and various other indicators of quality of care. To answer the first objective, the Health and Welfare Survey 2005 was used to analyse uti lization by scheme and national claims data were used to analyse LOS and early readmission of OM patients. To assess quality of care in detail, primary quantitative and qualitative data were collected on OM patients and providers in Samutsakhon province. The study indicates that the type of insurance scheme influences performance. The utilization review found that SSS members had a higher probability of using ambulatory services but a lower probability of being hospitalized. CSMBS members had a higher probability of being hospitalized. Members of the VC scheme had shorter LOS than CSMBS members and a higher probability of readmission relative to both SSS and CSMBS members. The empirical study found that CSMBS members were more likely to receive care consistent with standard guidelines. However, intermediate outcomes such as fasting plasma glucose, and Haemoglobin Ale level, were not as good as might be expected possibly due to the effects of other factors such as body mass index and patient behaviour. The qualitative study found that different patient groups had different expectations and perceptions of quality of service and that the insurance scheme and hospital policy influenced provider behaviour. The study demonstrates that, despite universal coverage, patients covered by different insurance schemes experienced variation in quantity and quality of care. Countries moving toward universal coverage should pay particular attention to the features of the insurance scheme design, especially with respect to management, organization, provider payment and the benefit package, as these features influence the performance of the scheme and the ability to achieve health system goals.

Public-Private Partnerships (PPPs) in the health sector are essential in light of the challenges the public sector is facing in healthcare finance, provision and management. Recognizing the need to provide insurance coverage to those below the poverty line (BPL), Rashtriya Swasthya Bima Yojana (RSBY) was introduced in 2008 by the Ministry of Labour and Employment in India. RSBY is a social health insurance scheme for the informal sector, where health care delivery and management involves a multitude of stakeholders from both public and private sectors who are governed by contractual agreements. A family of up to five pays INR 30/- (£0.30) annually for enrolment for a coverage of INR 30,000/- (£302). The balance of the premium is subsidized and shared by the Central (75%) and the State (25%) governments. This research aims to evaluate the availability, provision and management of health services under RSBY Public-Private Partnership contracts and factors that might influence them in order to inform policy makers on how to improve scheme implementation for the BPL beneficiary. The study was conducted in the districts of Patiala and Yamunanagar, in the States of Punjab and Haryana respectively. The study has both qualitative and quantitative components using primary and secondary data. The results of the study can be broadly categorized under the main pillars of scheme design and implementation. These include political, regulatory and institutional capacity; stakeholder contracting; enrolment of beneficiaries; empanelment of health facilities; and finally provision and utilization of services. RSBY has clearly attempted to address the existing gaps in the provision of health services by offering a balanced Public-Private Partnership model that provides some degree of financial protection to the end user. Despite the weaknesses identified, it is a robust and evolving model that needs to be continuously developed, on the basis of lessons learnt from implementation of the scheme.

Taiwan adopted its National Health Insurance (NHI) scheme in 1995. Presently, the scheme covers virtually all of the island’s citizens. However, it is under the threat of a serious imbalance between expenditure and revenue. As spending has become unsustainable, everyone has realised the need for financial reform. However, the reform process itself is beset by political confrontations. There is a need to deepen the understanding of the relationships and dependencies among the policy actors. With a view to helping address this problem, this study empirically examines the multiple types of ties prevailing between the policy actors and the resulting power distribution while the DPP government was working earnestly towards reforming the NHI’s financial system in the period 2000-2008. Apart from official documents, data are drawn from a network survey coupled with semi-structural interviews of 62 policy actors including government officials and related unofficial policy participants. Measures such as the in-degree centrality index and core/periphery model, betweenness centrality, structural hole index (effective size), density index, E-I index and CONOOR procedure (Blockmodeling and multidimensional scaling - MDS) are used to identify the major participants and network structures in the NHI domain and assess their relative influence-powers on the basis of information transmission patterns, resource exchanges, action-set coalition relationships and reputational attributions. It is shown that, although the public sector and the medical associations were at the helm of the NHI reform, financial reform remained unfulfilled mainly because of poor communications among societal actors. We then performed a social network analysis and systematically mapped the prevailing political conflicts among diverse policy stakeholders. We confirm that SNA is an effective research tool for political feasibility evaluation; it can facilitate smoother policy adoption by enhancing better interactions within networks.

Universal coverage is a core health system goal which can be met through a variety of health financing mechanisms. The focus of this PhD is on one of these mechanisms, community-based health insurance (CBHI). CBHI aims to provide financial protection from the cost of seeking health care through voluntary prepayment by community members; typically it is not-for-profit and aims to be community owned and controlled. Despite its popularity with international policymakers and donors, CBHI has performed poorly in most low and middle income countries. The overarching objective of this PhD is therefore to understand the determinants of low enrolment and high drop-out in CBHI. The PhD builds on the existing literature, which employs mainly economic and health system frameworks, by critically applying social capital theory to the analysis of CBHI. A mixed-methods multiple case study research design is used to investigate the relationship between CBHI, bonding and bridging social capital at micro and macro levels and active community participation. The study focuses on Senegal, where CBHI is a component of national health financing policy. The results suggest that CBHI enrolment is determined by having broader social networks which provide solidarity, risk pooling, financial protection and financial credit. Active participation in CBHI may prevent drop-out and increase levels of social capital. Overall, it seems CBHI is likely to favour individuals who already possess social, economic, cultural and other forms of capital and social power. At the macro level, values (such as voluntarism, trust and solidarity) and power relations inhering in social networks of CBHI stakeholders are also found to help explain low levels of CBHI enrolment at the micro level. The results imply the need for a fundamental overhaul of the current CBHI model. It is possible that the needed reforms would require local institutions to develop new capacities and resources that are so demanding that alternative public sector policies such as national social health insurance might emerge as a preferable alternative.