Insurance creates an important source of economic well-being by providing for beneficiaries in times of need. But because a variety of forces may inhibit the proper functioning of insurance markets, governments are deeply involved through regulation, subsidies, and direct provision of insurance. This dissertation studies insurance demand, supply, and the role of policy in two types of markets of direct interest to policymakers: health insurance and annuities. I highlight the importance of both traditional market failures (adverse selection and moral hazard) and less standard factors like limited competition (market power) and puzzlingly low insurance demand to influence insurance market outcomes. In the first chapter, I study how health insurers compete in individual insurance markets like those established in the Affordable Care Act. I focus on the role of an increasingly important benefit: plans’ networks of covered medical providers. Using data from Massachusetts’ pioneer insurance exchange, I show evidence of substantial adverse selection against plans covering the most expensive and prestigious academic hospitals. Individuals loyal to the prestigious hospitals both select plans covering them and are more likely to use these hospitals’ high-price care. Standard risk adjustment does not capture their higher costs driven by preferences for using high-price providers. To study the welfare implications of network-based selection, I estimate a structural model of hospital and insurance markets and use the model to simulate insurer competition on premiums and hospital coverage in an insurance exchange. I find that with fixed hospital prices, adverse selection leads all plans to exclude the prestigious hospitals. Modified risk adjustment or subsidies can preserve coverage, benefitting those who value the hospitals most but raising costs enough to offset these gains. I conclude that adverse selection encourages plans to limit networks and star academic hospitals to lower prices, with the welfare implications depending on whether those high prices fund socially valuable services. Chapter 2 also studies health insurance exchanges and the competitive effect of a policy design choice: how the level of subsidies is determined. In the Affordable Care Act exchanges and other programs, subsidies depend on prices set by insurers – as prices rise, so do subsidies. I show that these “price-linked” subsidies incentivize higher prices, with a magnitude that depends on how much insurance demand rises when the price of uninsurance (the mandate penalty) increases. To estimate this effect, I use two natural experiments in the Massachusetts subsidized insurance exchange. In both cases, I find that a $1 increase in the relative monthly mandate penalty increases plan demand by about 1%. Using this estimate, my model implies a sizable distortion of $48 per month (about 12%). This distortion has implications for the tradeoffs between price-linked and exogenous subsidies in many public insurance programs. I discuss an alternate policy that would eliminate the distortion while maintaining many of the benefits of price-linked subsidies. Chapter 3 studies demand for annuities – insurance products that protect retirees against outliving their assets. Standard life cycle theory predicts that individuals facing uncertain mortality will annuitize all or most of their retirement wealth. Researchers seeking to explain why retirees rarely purchase annuities have focused on imperfections in commercial annuities – including actuarially unfair pricing, lack of bequest protection, and illiquidity in the case of risky events like medical shocks. I study the annuity choice implicit in the timing of Social Security claiming and show that none of these can explain why most retirees claim benefits as early as possible, effectively choosing the minimum annuity. Most early claimers in the Health and Retirement Study had sufficient liquidity to delay Social Security longer than they actually did and could have increased lifetime consumption by delaying. Because the marginal annuity obtained through delay is better than actuarially fair, standard bequest motives cannot explain the puzzle. Nor can the risk of out-of-pocket nursing home costs, since these are concentrated at older ages past the break-even point for delayed claiming. Social Security claiming patterns, therefore, add to the evidence that behavioral explanations may be needed to explain the annuity puzzle. / Economics

The three essays in this dissertation explore how households in Southern Africa interact with and rely upon environmental resources. The first chapter examines the relationship between irrigation dam placement and local infant health outcomes. Irrigation dams can enable farms to harness considerable water resources, and this has been critical to increasing the global food supply. However, irrigation consumes 70 percent of global water resources and returns polluted water back into river systems. I examine the effect of irrigation dams on water pollution and infant health outcomes in South Africa. To remove bias associated with non-random dam placement, I utilize an instrumental variables approach that predicts dam placement using geographic features and time-varying policy changes. I find that each additional dam within a district increases both water pollution and infant mortality. In districts downstream from dams, alternately, dams generate smaller water pollution effects and reduce infant mortality, though magnitudes are much smaller. I argue that this pattern is consistent with pollution-induced health costs that outweigh economic benefits within the districts that receive dams. Downstream, however, pollution generates smaller costs and the economic benefits dominate. Exploring other plausible channels through which irrigation dams may affect infant mortality, I find that while irrigation dams generate substantial effects on district employment and small effects on migration, these factors do not appear to explain the health outcomes observed. Instead, the results suggest that water pollution and reduced water availability may contribute to higher infant mortality near agricultural activity. The second chapter in this dissertation explores technology adoption of trees with environmental benefits, in the context of a field experiment in Zambia. As context, many technology adoption decisions are made under uncertainty about the costs or benefits of following through with the technology after take-up. As new information is realized, agents may prefer to abandon a technology that appeared profitable at the time of take-up. Low rates of follow-through are particularly problematic when subsidies are used to increase adoption. This chapter uses a field experiment to generate exogenous variation in the payoffs associated with taking up and following through with a new technology: a tree species that provides fertilizer benefits to adopting farmers. Our empirical results show high rates of abandoning the technology, even after paying a positive price to take it up. The experimental variation offers a novel source of identification for a structural model of intertemporal decision making under uncertainty. Estimation results indicate that the farmers experience idiosyncratic shocks to net payoffs after take-up, which increase take-up but lower average per farmer tree survival. We simulate counterfactual outcomes under different levels of uncertainty and observe that subsidizing take-up of the technology affects the composition of adopters only when the level of uncertainty is relatively low. Thus, uncertainty provides an additional explanation for why many subsidized technologies may not be utilized even when take-up is high. Finally, the third paper in my dissertation explores the role that mineral wealth has on local economic outcomes. Mineral wealth has been central to the development of the South African economy. However, we have little evidence regarding how it has affected employment and poverty. This paper explores how within-country variation in mineral wealth affects district-level outcomes. Using data on mineral deposits and historical world prices, I construct a plausibly exogenous variable reflecting district-level aggregate mineral wealth, and I use variation in this index to evaluate economic outcomes over three Census rounds (1991, 1996, 2001). In the short run, I show that positive shocks to aggregate mineral wealth generate higher employment, largely driven by increased mining employment. This increase in mining is accompanied by reductions in agricultural employment and slight reductions in manufacturing employment. On average, adult age individuals also report working more hours and earning higher salaries in districts experiencing higher mineral prices. In sum, mineral wealth shocks generate benefits to districts, with less households below the poverty line. / Public Policy

This dissertation consists of three essays, two on estimating dynamic entry games and one on the inference for misspecified models with fixed regressors. Big box retail stores have large impact on local economies and receive large subsidies from local governments. Hence it is important to understand how discount retail chains choose store locations. In the first two essays, I study the entry decisions of those firms, examine the role of preemptive incentives, and evaluate the impact of government subsidies on those decisions. To quantify preemptive incentives, I model firms' entry decisions using a dynamic duopoly location game. Stores compete over the shopping-dollars of close-by consumers, making store profitability spatially interdependent. I use separability and two-stage budgeting to reduce the state space of the game and make the model tractable. Instead of adopting census geographic units, I infer market divisions from data using a clustering algorithm built on separability conditions. I introduce a `rolling window' approximation to compute the value function and estimate the parameters of the game. The results suggest that preemptive incentives are important in chain stores' location decisions and that they lead to loss of production efficiency. On average, the combined sum of current and future profits of the two firms is lowered by 1 million dollars per store. Finally, I assess the impact of government subsidies to encourage entry when one retailer exits, as happened in the recent crisis. I find that although the welfare loss such exits cause on local economies can be substantial, the average size of observed subsidies is not enough to affect firms' entry decisions. This study is organized as follows. In the first essay, I provide descriptive evidence of preemptive entry in the discount retail industry and explain how I model firms' entry decisions. In the second essay, I describe the estimation strategy and present the counterfactual analyses. The third essay is joint work with Alberto Abadie and Guido Imbens. Following the work by Eicker (1967), Huber (1967) and White (1980ab; 1982) it is common in empirical work to report standard errors that are robust against general misspecification. In a regression setting these standard errors are valid for the parameter that minimizes the squared difference between the conditional expectation and the linear approximation, averaged over the population distribution of the covariates. In this essay, we discuss an alternative parameter that corresponds to the approximation to the conditional expectation based on minimization of the squared difference averaged over the sample, rather than the population, distribution of the covariates. We argue that in some cases this may be a more interesting parameter. We derive the asymptotic variance for this parameter, which is generally smaller than the Eicker-Huber-White robust variance, and propose a consistent estimator for this asymptotic variance. / Economics

My dissertation consists of three chapters on topics in public finance. Chapter 1 studies optimal disability insurance (DI) in two parts. In the first part, I show that the well-established result that DI reduces labor supply is driven largely by making it feasible for the disabled to stop working, rather than by reducing effective wages. Therefore, DI is very valuable because it operates through a non-distortionary (and welfare enhancing) income effect, rather than a price or substitution effect. In the second part of my paper, I show that externalities and internalities create unique challenges for designing DI systems. I study an institutional setting where 80% of the population receives a payment on the same day each month. I find that the probability of an emergency room visit increases for DI beneficiaries, but not others, when monthly income is received, and this response to payments is present even in the years before they were granted DI benefits. The results imply that optimal policy may involve non-traditional policy tools. Chapter 2 presents evidence on one such alternative policy tool, in-kind transfers, in the context of food stamp benefits in the United States. In November 2013, temporary benefit increases in the American Recovery and Reinvestment Act expired, resulting in lower benefits for all Supplemental Nutrition Assistance Program (SNAP) households. I use scanner data from 400 grocery stores and over 2.5 million SNAP households in Atlanta, Los Angeles, and Columbus, OH to estimate the effect of the benefit cuts on household spending. I find that the impact per household was relatively small, but the aggregate impact was large because 23 million households were affected in November 2013. In contrast, subsequent legislation passed in February 2014 will impact relatively few households, leading to a much smaller aggregate impact. Chapter 3 shows that, in addition to ER visits, consumption of hard alcohol, crime, and traffic accidents all increase nationwide in Denmark when 80% of the population receives income each month. Deaths may also increase. This evidence runs counter to standard theories of how households make consumption decisions and sheds new light on potential explanations. / Economics

This thesis consists of three essays on corporate R&D and innovations in the United States. Utilizing the newly collected survey of Business Research Development and Innovations (BRDIS), the first chapter establishes several stylized facts regarding the distribution of R&D spending as well as innovation outcomes, highlighting the fact that businesses with very little reported R&D also produce a fair amount of innovations, measured both by patent filings and new product introductions. In addition, service industries have become a major contributor to R&D spending and patenting activities. As most of our traditional studies in innovations have focused on manufacturing firms, these newly documented facts suggest a new perspective for future innovation research, with a refreshed look at the traditional definition of industry and firm linkages, as the rigid definition based on a manufacturing-dominated economy becomes less and less relevant in a new era of service-dominated economy. The second chapter of the work validates the self-reported measures of innovations from the BRDIS survey by studying its relationship with real economic measures. More specifically, by matching information of publicly listed firms from Compustat to the BRDIS dataset, I have found that Tobin's q ratio positively predicts firms’ future innovative successes, even after controlling for R&D and patent information. On a separate account, I have matched the BRDIS data with the Annual Survey of Manufacturers, and found that intensity measures of innovations (percentage of sales from newly introduced product/services) are positively associated with productivity growth of manufacturing firms. The last chapter studies the impact of credit constraints on firms' investment decisions (both capital expenditures and R&D spending) during the recent financial crisis. Using exogenous shocks coming from the syndicated loan market, the paper finds that credit constraints caused sharper declines in public firms' capital expenditures in 2009, and the impact was long lasting. However, the adverse effect of credit constraint was confined to firms with low internal liquidity reserves. Firms with high internal capital reserve were able to use the liquidity buffer to smooth out external shocks. However, no adverse impacts on R&D spending were identified for the sample of public firms, which supports commonly-held views in the corporate finance literature, namely, adjustment costs of R&D are high and firms usually depend upon internal capital and equity finance to support R&D activities. / Economics

Economic development, more often than not, is tightly connected to good governance and adequate provision of public goods and services. My dissertation examines characteristics of developing countries that are relevant to the quality of government, as they affect the interaction between citizens and government. In chapter 1, I focus on squatters and the opportunities for political intermediation associated with the lack of property rights on land. In this chapter, I study an extensive land titling program in urban Mexico and its effect on clientelism. In chapter 2, I examine the costs of party turnover for the quality of public education in Brazilian municipalities, a context where the bureaucracy is not shielded from the political process. In chapter 3, I study the feedback effect of government work on citizens’ participation. To this end, I focus on maintenance work on the public space and citizens’ complaints in the City of Buenos Aires. / Economics

This dissertation consists of three essays, two in the field of international economics and one in macroeconomics. The first essay studies the interaction of manufacturing firms' sourcing decisions and their R\&D investment. I propose a two-country model of sourcing where heterogeneous firms choose the sourcing destination as well as the extent to which they would like to engage in productivity-enhancing R\&D investments. Firms self-select into different combinations of sourcing modes and R\&D engagements based on their pre-R\&D, or pure productivity level. This model generates novel predictions of sorting into four different sourcing/R\&D combinations in the equilibrium. In contrast to canonical models of sourcing, this paper incorporates both extensive and intensive margins of R\&D and shows the interaction of these choices with sourcing decisions through the cost function. The model suggests that foreign sourcing decision should be complementary to R\&D investment both on the extensive margin and the intensive margin. I find empirical support for the model predictions using the ESEE data for Spanish manufacturing firms. Using an instrumental variable approach, I show that foreign sourcing is causally linked to an increase in a firm's propensity to engage in R\&D extensively and the extent of the engagement intensively. The second essay investigates manufacturing firms' decisions to employ multiple sourcing modes simultaneously and their relationship to productivity, firm size, and capital intensity in a heterogeneous firm framework. Using a simple extension of the \cite{AntrasHelpman04} model where each firm can produce differentiated goods by sourcing from a multitude of suppliers, I show that more productive firms are able to use more varied sourcing modes, produce more products and shift a larger share of inputs away from the most basic form of sourcing into more advanced strategies. This is achieved by allowing supplier heterogeneity in the set of associated sourcing fixed costs. I show that the model predictions are consistent with stylized facts and empirical analysis with Spanish manufacturing firms. Using an instrumental analysis, the multiple sourcing patterns are found to be driven by firm-level productivity. Finally, the third essay examines the propagation of monetary policy through the banking system in China in the context of a major banking reform in the 1990s. Using bank-level data for commercial banks over twenty years from 1986 to 2008, I examine the supply side of the narrow credit channel: loan level responses of commercial banks to monetary policy tools of the central bank. I find that banks have disparate but strong responses to different policy instruments depending on their type and level of capitalization. Moreover, the major banking reform in the 1990s has changed some characteristics of the bank-dependent propagation mechanism without diminishing its central role in monetary transmission. / Economics

This dissertation consists of three independent chapters on econometric methods for macroeconomic analysis. In the first chapter, I propose to estimate structural impulse response functions from macroeconomic time series by doing Bayesian inference on the Structural Vector Moving Average representation of the data. This approach has two advantages over Structural Vector Autoregression analysis: It imposes prior information directly on the impulse responses in a flexible and transparent manner, and it can handle noninvertible impulse response functions. The second chapter, which is coauthored with B. J. Bates, J. H. Stock, and M. W. Watson, considers the estimation of dynamic factor models when there is temporal instability in the factor loadings. We show that the principal components estimator is robust to empirically large amounts of instability. The robustness carries over to regressions based on estimated factors, but not to estimation of the number of factors. In the third chapter, I develop shrinkage methods for smoothing an estimated impulse response function. I propose a data-dependent criterion for selecting the degree of smoothing to optimally trade off bias and variance, and I devise novel shrinkage confidence sets with valid frequentist coverage. / Economics

In the United States, the fraction of individuals with public insurance is growing and the policies and markets that serve them are changing. Over two-thirds of Medicaid recipients are now enrolled in managed care organizations (MCOs), but little is known about how these plans operate. To study this market, I use data from New York Medicaid where managed care recipients are randomly assigned to plans. In chapter one, I estimate how physician and hospital networks impact health care use and spending. In chapter two, I study how production and selection vary across the managed care plans participating in New York Medicaid. Finally, in the third chapter, I turn to the Medicare program and examine how health care use and spending change at age 65 as adults switch from private to public coverage. Taken together, my dissertation chapters provide new perspective on how differences within public programs and between public and private programs, impact consumers. / Health Policy

The three chapters in this dissertation examine aspects of the relationships between transparency, government accountability, and the quality of public services. In the first chapter, I ask how ethnic diversity, or lack thereof, among polling station officials affects voting outcomes. I exploit a natural experiment occurring in the 2014 parliamentary elections in India, where the government mandated the random assignment of state employees to the teams that managed polling stations on election day. I find that the presence of officers of minority identities within teams led to significant shifts in vote share toward the political parties associated with these groups. Results suggest that the magnitude of these effects is large enough to be relevant to election outcomes. Using large-scale survey experiments, I provide evidence of own-group favoritism in polling personnel and identify the process of voter identity verification as an important channel through which voting outcomes are impacted. The second chapter examines whether electronic procurement (e-procurement), which increases access to information and reduces personal interactions with potentially corrupt officials, improves procurement outcomes in India and Indonesia. We find no evidence of reduced prices but do find that e-procurement leads to quality improvements in both countries. Regions with e-procurement are also more likely to have winners come from outside the region. On net, the results suggest that e-procurement facilitates entry from higher quality contractors. The third chapter studies the effects of the enactment across U.S. states of open meetings laws which ostensibly increase the public availability of information on legislator behavior. As recent work shows that increased remoteness of capital cities in U.S. states is strongly associated with reduced accountability and worse government performance, I also investigate how the impacts of open meetings vary with state capital isolation. I find that open meetings increase spending on public goods and heighten confidence in state government on average. Heterogeneous impacts on incumbent vote share suggest that at both low and high levels of initial accountability, open meetings provide citizens with additional information that influences voting decisions. / Public Policy