This thesis is composed of four independent empirical essays that draw on and contribute to aspects of health, urban, public, and environmental economics. The chapters can be split into two distinct parts. The first part comprises two chapters that provide new quantitative evidence about the impacts of recent health care policies in the English National Health Service (NHS). While essentially describing policy evaluations, the essays provide insights into the underlying economic forces of health care demand and supply, and are linked to the urban economics literature by an explicit consideration of spatial issues. The second part comprises two further chapters that focus on a core urban economics topic — housing markets — placing particular emphasis on specific links between housing and environmental issues. The unifying theme, and overriding contribution, of the thesis is to bring fresh evidence to bear on policy-relevant issues in urban and public economics by the generation of new datasets and the application of econometric techniques.

The collection of three essays study how political factors can shape economic outcomes, with a particular regard to developing countries. My exploration in this direction begins with the government-firm connections and extends to the cause and prevention of internal armed conflicts. The first chapter examines the reciprocal relationship between governments and firms. The rent-seeking behavior of politically connected firms and its associated costs have long been recognized by economists. The existing literature has mainly focused on the favors that firms receive, but much less attentions has been paid to what politicians gain in return. In my first chapter, titled ”Can Governments Harvest Connections with Firms? Evidence from China,” I provide evidence on the reverse - firms providing favors to governments in a reciprocal relationship - exploiting a natural experiment in China. In October 2001, the tax revenue sharing rule between central and local governments was unexpectedly reformed: the higher the local tax revenue in 2001, the higher the share that local governments would get post-2001. From a newly collected dataset, I find that before the reform the governments that granted more favors to firms - access to credit and tax deductions - were able to mobilize more assistance from firms in order to raise the tax revenue in 2001. Furthermore, this reciprocation is not an institutional relationship, but hinges on a repeated interaction between firms and local leaders. Exploring the variation in leadership turnover, I find that firms who had previously received government favors provided no assistance to leaders who would soon leave office. These results are consistent with a theory of reciprocal relationships between governments and firms. My findings not only suggest that governments and firms can form dynamic relationships to exchange favors intertemporally, but also shed light on the government-business relationship in China. The last two chapters focus on political violence, which has commonly been regarded as among the first-order issues in developing countries. The second chapter examines whether the endowment of natural resources might lead to internal armed conflict. To examine this question, we ideally require exogenous variation in resource windfalls. This is a challenging task as the quantity of natural resources extracted is a choice and oil prices may be affected by violent conflict. To address this issue, in a published paper co-authored with Guy Michaels at the London School of Economics, titled ”Do giant oilfield discoveries fuel internal armed conflicts?,” we use new data to examine the effects of giant oilfield discoveries around the world since 1946. We show that the timing of giant oilfield discoveries is plausibly exogenous. We find that on average these discoveries increase per capita oil production and oil exports by up to 50 percent. But these giant oilfield discoveries also have a dark side: they increase the incidence of internal armed conflict by about 5-8 percentage points. This increased incidence of conflict due to giant oilfield discoveries is especially high for countries that had already experienced armed conflicts or coups in the decade prior to discovery. In the last chapter, I focus on the prevention of internal armed conflict. In order to lessen the likelihood of conflict, one important precaution is to ensure that military resources do not fall into the hands of non-state groups. Observing the recent conflicts in the Middle East, and, in particular, the issue of wide spread of military resources well beyond governments control, one main contributing feature is that governments sponsor weapons to non-state agents (militias) to fight the states enemies, but after the war ends these weapons cannot be called back. In my third chapter, titled ”Proxy Warriors: A Theory of Military Assistance”, I develop a theory to study government’s optimal strategy in sponsoring weapons to militias. Although giving militias more weapons may win the war more quickly, it puts more weapons into the hands of militias afterwards. Militias learn that they can benefit from holding more weapons when hostilities end. This incentivizes the militias to fight strategically in order to maximize their subsequent stock of weapons. Given this governmental dilemma, in a dynamic setting I rationalize a supply strategy featured with a stopgap proposed by the US security forces.

Dominant approaches in International Political Economy treat inflows of foreign direct investments (FDI) only as a material fact, a physical flow of capital. The analysis of the perceptions of inward FDI presented in this research, however, reveals that the meaning that policymakers and analysts attribute to FDI inflows goes far beyond that. What is more, the predominant interpretation of the meaning of FDI inflows has changed dramatically over time: While they were perceived primarily as a threat to national economic development from the 1950s to the 1980s, they came to be gradually re-interpreted as a sign of economic success in the 1990s. Focusing on these developments in the major OECD economies, this research aims to make sense of this stunning transformation in the social interpretation of inward FDI and to examine the implications of these ideational evolutions for policy outcomes. To do so, the research adopts a mixed methods research design, which combines quantitative approaches with the insights gained from qualitative historical analysis: After providing a nuanced theoretical discussion of the significance of economic narratives in international economic affairs and a broad overview of the key developments in FDI policies and relevant policy discourses in the six largest advanced economies during the post-war era, the research subjects the theoretical argument to two quantitative tests at large cross-national samples using data from public opinion surveys and general election results; finally, a qualitative comparison of relevant developments in the United Kingdom and France analyses the impact of these ideational changes on FDI policy-making processes in empirical depth.

This dissertation examines the Chinese Shanxi piaohao, arguably the most important Chinese indigenous financial institutions in the nineteenth century, which emerged in one particular province. Concentrating on their governance structure, business strategy, and historical development, my thesis explores the emergence, growth and decline of the piaohao, and asks how they solved the commitment problem and developed their business by means of modern economists’ tools, such as path dependency and the principal agent model. This study uses a wealth of microeconomic data, which, although published, had not been digitized and properly analysed before. My work reveals that the emergence of the Shanxi piaohao shows reactive sequence path dependence. In a situation which gave little protection to shareholders’ capital, it imposed a highly centralized management structure, and a tenure- and performancebased incentive structure to discipline distant employees. Moreover, through establishing a widespread branch network, providing services to a wide range of clients, inventing various types of draft, and pricing remittance fees on diverse factors, the Shanxi piaohao were soon successful and controlled the Chinese remittance market for many decades. However, as it expanded and as China’s social and business environment became more unpredictable, the piaohao’s head managers reached the limits of their competence and its centralized management structure began to show diminishing returns. High financial leverage and a narrowed profits margin also indicated the hidden risks to the piaohao. When the 1911 Revolution broke out and ended China’s last feudal dynasty, many piaohao branches encountered immense losses from looting, bad loans and deposit withdrawals. It was during this time that modern Chinese banks and foreign banks penetrated the Chinese remittance market and poached the piaohao’s staff. Because the piaohao shareholders shared unlimited liabilities, when these threats materialized, many of them went bankrupt, while others turned to investments elsewhere and abandoned the piaohao.

Since the 1960s the government of Burkina Faso has consistently sought to implement new development policies to improve the economic and social conditions of its people. Until the end of the 1990s these efforts have been disappointing and unsatisfactory. In the early 2000s there was a shift towards a knowledge-centred development policy and policy makers trusted that it would bring about the sough-after improvements. In 2006 Burkina Faso chose to adopt the National Innovation Systems (NIS) framework as a policy tool to implement this new policy. Drawing on a broader definition of technology that covers social technologies, this thesis used the ST-Systems analytical concept to chart the adoption and diffusion of the NIS policy tool at two levels, namely at strategic policy level and at the operational level, focusing here on the case of Bt cotton which was officially introduced to Burkina Faso in 2003. Ethnographic methods, including in-depth interviews with policy makers, farmers, Monsanto representatives, civil society actors and researchers, were used to gain new insights into the difficulties encountered by these actors when trying to implement the NIS policy tool. 60 interviews were analyzed against a backdrop of detailed historical studies, based on examining a large amount of grey literature, published between 1961 and 2016. Findings show that the implementation of the NIS policy for innovation diffusion for socio-economic development in Burkina Faso was shaped by local actors competing for control of financial resources and power positions. The new tool also had to compete with older, more familiar tools. In the end, it failed to bring about the expected improvements in policy design and practice at sectoral level. The thesis is among the first to have studied empirically the transfer processes of the NIS policy tool for innovation diffusion in an African country (Burkina Faso) through a case study focusing on the introduction of Bt cotton. The results achieved should contribute to more informed development policy-making in Burkina Faso.

The present collection of essays studies some of the ways in which the interaction of economic and political forces affects a country’s development path. The focus of the thesis is on Colombia, which is a fertile setting for the study of the political economy of development given its long-lasting internal conflict, the multiple reforms to the functioning of the state that have taken place in the last decades and the availability of high quality sub-national data. The first two chapters explore people’s tendency to use resources differently depending on their source. The first chapter shows how the source of public revenue affects a government’s incentives to provide public goods and services, while the second one studies people’s propensity to make risky choices when playing a game with easily-gotten house money. The third chapter contributes to our understanding of the international dimension of civil conflict by analyzing the effects of access to territory in a neighboring country on the intensity of an insurgent group’s activities. The idea that governments perform better when they are funded with tax revenue has a long history and surfaces often in debates regarding the origin of the natural resource curse, the effectiveness of foreign aid and the benefits from decentralization. However, the empirical evidence backing this claim is somewhat limited. In the first chapter, I try to fill this gap by comparing the effects of increases in internally-raised tax revenue and in royalties from the extraction of oil on local public good provision in a panel of Colombian municipalities. I find that tax revenue leads to an improvement in public services while oil royalties have no effect. Furthermore, I document a negative effect of royalties on the quality of government, as measured by the disciplinary prosecution of local public officials. One possible explanation for the results in chapter 1 is that taxation leads to greater accountability because voters value tax revenue more than revenue from an external source. The idea that people assign greater value to resources over which they have some sense of ownership is further explored in the second chapter. In that chapter, which is the result of joint work with Juan Camilo C´ardenas, Nicol´as De Roux and Christian Jaramillo, we show that the risk aversion displayed by participants in a lab experiment varies depending on whether they received the endowment on the same day of the session or one month in advance. We interpret this finding as evidence of people’s reduced risk aversion when allocating easily-gotten resources, also known as the ‘housemoney’ effect. In the third chapter, I turn my attention to Colombia’s internal armed conflict and I study the allegation that the administration of Hugo Ch´avez provided access to territory in neighboring Venezuela to Colombian insurgent groups FARC and ELN. I document a disproportionate increase in the intensity of insurgent activity (mainly by FARC) in Colombian municipalities next to the border with Venezuela after Ch´avez comes to power in 1999. This finding is consistent with the idea that the rebels had access to a safe haven across the border during the Ch´avez administration, but that the strategic advantage provided by this sanctuary decreased with distance to the border. This chapter contributes to our understanding of foreign support for insurgent groups by developing a novel data-driven method for the detection of the usually secretive activities of trans-national rebel groups. It additionally provides credible estimates of the causal effect of access to foreign territory on insurgent activities.

This thesis takes interest rates as the topic of interest, and studies financial market integration in China. Paper I studies structural determinants and temporal coefficients of interest rates scattered over 6th -20th century China. Detailed findings concern intrinsic loan features (such as maturity and creditor type, among others) and how they affected the interest rate level. Overall, interest rates decreased in the markets under study, with fluctuations corresponding with dynastic cycles, up-and-down. The two interest rate troughs are found around the 9th -11th century of the Song dynasty and the 19th century of the late Qing period. Significant events of political economy (wars and recovery, international relations and trade, etc.) significantly affected interest rates, but mostly through temporary shocks; economic development and its ensuing financial advancement (in institution, innovation, markets, etc.) tended to show qualitative and long-run impact on financial markets and interest rates. Part II estimates financial integration regarding 14th-20th century China. Firstly estimated is pair-wise integration based on time-series data in 18th-20th century China. Before 1840, distance was the major (but not the only) determinant of financial integration. The maximum range of financial integration at the time was up to 1,400 kilometres, which was slightly farther than that of commodity (grain) integration and confirms the macro-region theory of Skinner in that there was little cross-regional market interplay. However, the overall integration performance for the period before 1840 was limited, with large gaps between distance groups regarding both interest divergence and adjustment speed. A national financial market did not seem to emerge until the 19th century, when both local and cross-regional capital markets became more homogeneous (with converging interest rate gaps and synchronising arbitrage speed). However, the final wars (the 2nd anti-Japanese war and the 2nd civil war in China) before the People’s Republic of China (PRC) stopped this integration process. Secondly, overall integration among 23 provincial markets scattered over 57 years in the Ming and Qing period may be explained by three factors: education, population, and the relative position of the local market to all other markets. The spatial autoregressive coefficients were negative, suggesting that a local provincial market was negatively related to all other markets. Education and population represent the influence of innovation and commerce on overall integration respectively. They are positively associated with interest rate gaps, hence negatively connected to integration. However, such negative relationships might denote financial development in local markets, which lowered local interest rates and temporarily enlarged the interest rate gaps. Neither arable land nor warfare involvement was significant in explaining overall financial integration.

Statistics contribute to the understanding of events by objectifying phenomena, as they are perceived to reflect or be an approximation of reality. This perception is based on the premise that statistical tools are straightforward, apolitical facts. However, quantification and its results are not objective. Definitions are needed beforehand to determine the phenomenon to be measured and the aim of the quantification. Thus, statistics face debates on methods, interpretation and use. Using the Argentine cost of living indices released in 1918, 1924 and 1935 as a case study and following a process of de-construction/construction/re-construction of the series, this thesis studies how, why and by whom statistics are made and used. It suggests that the political economy plays a crucial role in the history of the Argentine cost of living index in the first half of the twentieth century. In the de-construction phase, the thesis analyses various reports to arrive at an understanding how the indices were originally estimated. The construction stage then discusses the people and institutions involved in the production of each index and the methodology that they used, placing both within the political, economic and social context. It looks at how and why each index was produced by analysing their context, uses, contemporary reception and significance. Moreover, the pitfalls that come from the assumptions and methods underlying the indices are demonstrated using data available to those who produced them. Lastly, each CLI estimate is re-constructed by correcting its main pitfalls using the information available when the series were initially developed to depict how different assumptions result in different series. This leads to an alternative cost of living index being presented for the period 1912-1943. The re-construction also comprises a comparison in tandem of the Argentine, US, British and German cost of living indices.

This thesis contends that the historiographical boundaries and focus of labour history, political history, of policy making and nationalisation have resulted in an incomplete understanding of trade unions attitudes towards, and influence upon, post-war British economic policy. In particular, the predominant concern of labour historians with strike patterns and their causes, particularly within the coal industry, has been at the expense of other forms of trade union activity. Whilst the more general historiography of the period and that of policy making address these issues, they do not tend to do so below the peak level organisation of the TUC and of Whitehall and Westminster. This has lead to miners unions being portrayed as a somewhat monolithic organisation predominantly concerned with disputes, strike prone with poor industrial relations, but politically conservative and generally supportive of the Labour Party and Government policy. In taking a multi-level analysis, with particular emphasis on Scotland, and examining the evidence from the NUM's interaction with Government, party, National Coal Board and the industry'S conciliation and consultative machinery, this thesis argues that a more diverse pattern of trade union attitudes and influence existed. It is suggested that the TUC had a relatively minor role to play in the development of coal nationalisation policy after 1947. Furthermore, the national level of the NUM was unable to adapt fully to its new-role under nationalisation because areas such as Scotland continued to exercise considerable power and influence. In this it is demonstrated that Scotland could take a divergent attitude to the national level of the union, particularly over wages, and ultimately meet with some success. The Scottish Area of the NUM also displayed poorer industrial relations to the national and local levels. In particular, the evidence from colliery level consultation demonstrates that there was a more positive and constructive side to local union activity within the nationalised industry than the focus on disputes hitherto suggested. Therefore, this thesis concludes that there is sufficient evidence from the experience of the NUM to suggest that a more complex and diverse pattern of trade union behaviour existed between 1945 and 1955 in the nationalised coal industry. However, this pattern is not so rooted in any Scottish cultural explanation, or contradictory to existing interpretations, as to preclude its broader applicability to other areas of the coal industry or unions in other nationalised industries.

Market based reforms to China’s agricultural sector, between 1978 and 1984, marked the start of the reform era. The reforms were enormously successful, resulting in dramatic increases in both agricultural productivity and output. The first two chapters of this thesis are an empirical exploration of the consequences of the agricultural reforms for the growth of China’s non-agricultural sector and the pattern of Chinese urbanisation. The third chapter uses the interaction between differential rural income growth and the One Child Policy, to shed light on how declining family size has fuelled the latent demand for sex selective abortion in China and beyond. The first chapter explores the link between agricultural productivity and industrialisation in the context of reform era China. A classic literature argues that, at low levels of development, improvements in agricultural productivity can provide an important stimulus to the nonagricultural sector, however empirical evidence of this is limited. Using a natural experiment provided by China’s agricultural reforms, I show that higher agricultural productivity growth had a substantial positive causal effect on non-agricultural output. I use the predictions of a simple two sector model, which nests the possibility of linkages through demand externalities, the supply of capital, and the supply of labour, to provide additional results indicating that the linkages I observe appear to be driven primarily by increases in the supply of capital. In the second chapter, I ask how higher agricultural productivity affected China’s urbanisation. In 1978, at the time of the reforms, more than 80% of China’s population lived in the countryside. By 2011, fewer than 50% did. Whether agricultural productivity increases the pace of urbanisation is theoretically ambiguous, and depends on whether the effect of higher rural incomes is more than offset by the increased demand for urban goods. I show that increased agricultural productivity not only increased the pace of urbanisation between 1978 and 1995, it also affected the type of cities that formed. Higher agricultural productivity increased the output of the urban service sector, at the expense of the tradable industrial sector. The results are consistent with a simple model where urbanisation and structural transformation are jointly determined. In the third chapter, I use China as a setting to explore the hypothesis that the increase in male-female sex ratios observed in China, and many other parts of the world, over the past fifty years is, at least in part, driven by the demographic transition to smaller family sizes. Since 1979, the One Child Policy has imposed economic, and sometimes non-economic sanctions, for breaching proscribed fertility levels. The economic sanctions have meant that, within a class of households, the 1CP may have constrained the fertility of poor households more than rich ones. Using plausibly exogenous variation in household income, I show that richer, less constrained households, had higher fertility in the wake of the implementation of the One Child Policy. I then show that this increase in fertility is associated with a subsequent decline in sex selection. The decline in sex selection is roughly contemporaneous with the emergence of pre-natal ultrasound which dramatically reduced the costs of sex selective abortion. Together, the results suggest that China’s dramatic decline in fertility in the 1970’s, may have played an important role in fuelling the demand for sex selection from the mid 1980’s onwards.