I present three essays discussing the impact of archaic institutions and technology on inequality in wages and political participation. First I examine a modern facet of the Indian caste system: political quotas for disadvantaged minorities and their impact on political participation. I find that aggregate turnout falls by 9% of the baseline and right-wing parties win 50% more often, but electoral competition is not significantly affected. Detailed individual-level data for one state suggests that voter participation falls among women and minorities. This suggests that restricting candidate identity to minorities may cause some bias in voter participation. Next, I study caste and human capital: specifically why workers remain in lowpaying hereditary occupations, providing an explanation for both occupational specialization and hereditary occupations. I use a simple model of insurance provision in which parents pass on human capital to their children in return for insurance in the event of sickness, and find that workers with low human capital are likelier to participate in the arrangement, and that a higher cost of sickness can sustain higher human capital transfers. I conclude by studying human capital and technology- the impact of information and communication technologies (ICT) on wage inequality. We tested the hypothesis that information and communication technologies (ICT) polarize labour markets, by increasing demand for the highly educated at the expense of the middle educated, with little effect on low-educated workers. Using data on the US, Japan, and nine European countries from 1980-2004, we find that industries with faster ICT growth shifted demand from middle educated workers to highly educated workers, consistent with ICT-based polarization. Trade openness is also associated with polarization, but this is not robust to controlling for Research and Development. Technologies account for up to a quarter of the growth in demand for highly educated workers.

The aim of my thesis is to investigate the importance of access to information for individuals in developing countries. In the first chapter, I describe an important channel through which improved access to market information could increase the prices that producers receive from middlemen. I develop a theoretical model of trade between a farmer and a middleman which allows for the existence of different types of middlemen and I provide an empirical test of the theory from an original framed field experiment carried out with farmers and middlemen in India. In chapter 2, I investigate the relationship between the decision to produce high-quality goods and two important characteristics of the product: the degree of observability of quality and the level of intermediation in the supply chain. I present a model which demonstrates that if quality is not perfectly observable, there will be a range of values of the price difference between high-quality and low-quality goods for which production of high-quality goods will occur with vertical integration but will not occur if the stages of production are carried out by separate agents. This chapter also presents some case studies of supply chains for various products in a number of developing countries that have characteristics which are consistent with the predictions of the model. In the final chapter, I try to understand how access to information could be improved for individuals in developing countries. I investigate the relationship between rates of mobile phone and Internet use and a number of geographic, institutional and economic variables in a sample of 164 countries from 1990 to 2009. The aim of this chapter is to identify the main characteristics of countries that have had success in adopting these new technologies in order to gain some insight into the barriers which may be faced by those countries that have been less successful.

This thesis consists of three independent chapters. The first chapter examines the strategic choices of the targets and the intensity of violence by rebel groups. The chapter presents a theoretical framework that links a rebel group’s targeting decisions to income shocks. It highlights that this relationship depends on the structure of the rebels’ tax base. The hypotheses from the model are tested in the context of India’s Naxalite conflict. The second chapter estimates the impact of military recruitment on human capital accumulation in colonial Punjab. In this context, I find that higher military recruitment was associated with increased literacy at the district-religion level. The final chapter presents a model that describes the optimal design of civil-military institutions in a setting where some control of the military over domestic politics is deemed desirable.

Two drawbacks of current empirical studies on late medieval financial market integration are: the use of low frequency data; and the lack of a benchmark for comparison. As a result, there is a tendency to underestimate the degree of integration and one has no clear idea about whether the estimated degree of integration is high or low by the standards of the time. Consequently, there is not yet a satisfactory answer as to how integrated and efficient financial markets were in the late Middle Ages and early modern era. In tackling these two problems, this thesis employs monthly and weekly exchange rates to measure the degree of exchange market integration and the results are judged using the speed of communication as a benchmark since the flow of information played a critical role in financial arbitrage. Therefore, this thesis is able to show that exchange markets were already well integrated in the late fourteenth century. From then to the late seventeenth century, the high speed of adjustment to profitable opportunities was maintained, but the transaction costs associated with arbitrage fell over time. The reduction in transaction cost may be attributed to the financial innovations that took place in the sixteenth century. This thesis also finds that the type of information related to shocks received by economic agents had a decisive impact on the speed of price adjustment. The more explicit the information, the more efficiently the market responded to shocks.

This thesis presents three empirical essays on the interaction of housing and labour markets, which generates academically meaningful social outcomes. The first essay looks at whether one’s tenure choice affects unemployment as this question has potential implications for homeownership subsidy schemes adopted in many advanced countries. The contribution of this essay is mainly methodological in that it rigorously deals with the endogeneity of homeownership by taking an IV approach with instruments not adopted previously for studies in the UK in conjunction with panel data models. Using the local homeownership rate and parental homeownership status as an instrument, it shows that homeownership does not increase the probability of being unemployed. The second essay highlights the role of local housing market information as a determinant of housing tenure. As the distance a mover wants to move increases, the costs of collecting information on the destination housing market rise and the quality and amount of the information collected fall. Therefore, it is hypothesised that the longer the distance moved, the more likely movers are to choose private renting over owner-occupation since homeownership decisions require a large amount of information on the target properties and their neighbourhoods. Empirical tests that control for relevant characteristics correlated with distance moved and tenure decisions provide supporting evidence for this hypothesis. The last essay is the first UK study to confirm that commuting time has a negative influence on worker effort. The topic has important implications for transportation policy, employer’s commuting welfare strategy and hiring decisions and individual worker’s location decisions. As commuting is physically and mentally tiring, it could influence worker effort negatively. The hypothesis turns out to be true when the absenteeism rate and unpaid overtime hours are used as proxy variables for work effort.

This thesis explores the real estate and happiness consequences of public investment in local public goods improvements by using unique micro-geographical data from Beijing; it focuses on the spatial variations in park amenity values, and on the impact of transport improvements on land prices and homeowners’ happiness. Despite intense public interest, little is known about these effects. This thesis aims to fill these gaps. First, I explore the impact and sources of variations of park proximities as capitalized into the residential land prices. This analysis, using geographically-coded data from Beijing, provides new insights on the ways in which land markets capitalize the values of proximity to parks and suggests that this is highly dependent on the parcel’s location and local contextual characteristics. Next, I examine the real estate consequence of public investment in transport improvements using a rich data set of vacant land parcels in Beijing. I use a multiple intervention difference-in-difference method to document opening and planning effects of new rail stations on prices for different land uses in affected areas versus unaffected areas. Residential and commercial land parcels receiving increased station proximity experience appreciable price premiums, but the relative importance of such benefits varies greatly over space and local demographics. Finally, I investigate the impact of transport improvements on happiness that altered the residence-station distance for some homeowners, but left others unaffected. My estimation strategy takes advantage of micro happiness surveys conducted before-and-after the building of new rail stations in 2008 Beijing. I deal with the potential concern about the endogeneity in sorting effects by focusing on “stayers”and using non-market housings with pre-determined locations. I find the significantly heterogeneity in the effects from better rail access on homeowners’ happiness with respect to different dimensions of residential environment. The welfare analysis results suggest strong social-spatial differentiations. In combination, the three papers of this thesis make important contributions to a growing literature on public infrastructure, land market and happiness.

The International Monetary Fund has often been accused of adopting a one-size-fits-all approach to the resolution of financial crises. However, its programs present substantial differences in terms of their relative size and conditionality among other characteristics. This dissertation examines the causes and consequences of this variation through the lenses of two cases in which the contrast between the Fund´s interventions was particularly marked: Argentina and Uruguay during the period that surrounds the financial crash of 2001-02. The first part of this study analyses the determinants of these multilateral interventions through an adaptation of Robert Puntam´s two level games, exploring the way in which national politics interacted with international priorities to produce distinct outcomes in Argentina and Uruguay. The two experiences confirm that domestic ratification processes impose significant constraints on the negotiation of IMF programs, potentially conferring localized bargaining advantages to borrowing governments. Beyond a certain point, however, an intensification of these ratification constraints can result in a suspension of the Fund´s support, after which borrowers´ bargaining position weakens dramatically. That this point of rupture was reached in Argentina but not in Uruguay was due primarily to the different propensity to cooperate exhibited by political actors in these two countries, itself the product of certain institutional conditions such as the strength of their systems of checks and balances or a varying distribution of veto power. In turn, the second part of this thesis applies a hypothetical counterfactual approach to assess the consequences of the multilateral decisions adopted during the 2001-02 crisis in the Southern Cone. Although the contrast between the suspension of the Argentine program in December 2001 and the Uruguayan bailout of August 2002 had surprisingly modest macroeconomic consequences, its impact on politics and institutions was profound both in the short and in the medium-term. As a result of these findings, this dissertation argues that a better understanding of the implications of multilateral crisis resolution loans on the political economy of the countries concerned is still needed.

Accounts of economic development during mid-twentieth century have been dominated by import-substituting industrialisation (ISI) and/or state-led industrialisation frameworks. is literature attaches considerable importance to such policy areas critical to manufacturing as: trade and tariffs, foreign exchange and the promotion of credit. According to this view, industrialisation became an official goal and in many developing economies governments committed to it seriously. Focusing on Colombia, this dissertation challenges conventional wisdom. It demonstrates that the Colombian state did not provide financial aid, or implement deliberate trade-protectionist support, for industrialists to the degree hitherto argued. A distinct political-economy configuration, in which small-scale agriculturalists, particularly coffee exporters, wielded significant power within the state, meant that the type of distortive pro-ISI macro policies pursued in other Latin American economies were eschewed. Industrialisation proceeded apace in Colombia, but this was chiefly a market- or private-led phenomenon. e methodology employed to substantiate this claim is not comparative, yet frequent references are made to other Latin American nations to serve as benchmarks and counterpoints. New archival material, both quantitative and qualitative, is combined in novel ways to substantiate the original, revisionist interpretations advanced in the thesis. Policy-makers, targeting the twin challenges of managing external-account pressures and sustaining fiscal revenue, rather than promoting inward-looking development, best explain moderate levels of tariffs and slight overvaluation of the currency observed in Colombian trade policy. e heretofore untold history of the Institute for Industrial Development, a direct supplier of venture capital, shows a government agency with major organisational weaknesses, incapable of fulfilling its legal mandate, least of conforming to the major role attached by the literature as key agent for industrialisation. Findings regarding credit demonstrate that neither ordinary nor subsidised credit flowed to manufacturing to the extent previously thought. Patterns of legislated credit, sector-targeted banking and privileged access to the Central Bank, all show that agrarian ventures, not industrialists, were the recipients of subsidised official financing. A growing incompatibility between the financial requirements of advanced industrialisation and the clientelistic nature of the domestic polity that had to cater for the needs of agrarian groups, prevented policy elites from adopting a pro-manufacturing stance in financial and credit policies, even had they so wished.

How do falling barriers to flows of trade and information both within and across countries affect economic livelihoods in developing countries? The three chapters presented in this PhD thesis aim to contribute to our understanding of this question.

This thesis contains three independent chapters aimed at increasing our understanding of the effects of Internet diffusion on politics and development. The first chapter proposes a novel methodology for measuring Internet penetration. Using IP geolocation data, a new measure of Internet access is created, which counts the number of IP addresses per person in a region. This is the first measure of Internet penetration that is comparable not only across countries but across sub-regions of countries such as states or even electoral districts. The second chapter applies this measure to test whether Internet diffusion can weaken incumbent power in a semi-authoritarian regime. Using Malaysia as a test case, I find that the Internet is responsible for a six point swing away from the incumbent party in the 2008 elections. In the third chapter, co-authored with Valentino Larcinese, we look at the effects of the Internet on U.S. presidential elections. In accordance with anecdotal evidence, we find that increased Internet penetration leads to an increase in small donations to the Democratic Party and a swing towards the Democratic presidential candidate.