Over the last ten years there has been a renewed interest in providing agricultural insurance in developing countries. However, voluntary demand for unsubsidised insurance products has been low, particularly from the poorest farmers. Chapter One presents a model of rational demand for hedging products, where there is a risk of contractual nonperformance. Demand is characterised and bounded for risk averse and decreasing absolute risk averse decision makers. For constant absolute and relative risk averse utility functions, demand is hump-shaped in the degree of risk aversion when the price is actuarially unfair, first increasing then decreasing, and either decreasing or decreasing-increasing-decreasing in risk aversion when the price is actuarially favourable. The apparently low level of demand for consumer hedging instruments, particularly from the most risk averse, is explained as a rational response to deadweight costs and the risk of contractual nonperformance. A numerical example is presented which suggests that some of the unsubsidised weather derivatives currently being designed for and marketed to poor farmers may in fact be poor products. Chapter Two presents experimental evidence collected from a framed microinsurance lab experiment using poor subjects in rural Ethiopia. In line with the theoretical model of Chapter One, demand for actuarially unfair index insurance is hump-shaped in wealth, first increasing then decreasing. In contrast with recent field experiments where it is not possible to demonstrate that low demand for indexed insurance is `too low', use of a laboratory experiment with an objectively known joint probability distribution allows normative statements to be made about the observed level of demand. The observed level of demand for index insurance in the experiment is higher than the decreasing absolute risk averse upper bound of Chapter One, suggesting that subjects bought `too much' index insurance. Chapter Three presents a vision of insurance design for the poor. Technically optimal arrangements involve insurance providers, such as microinsurers or governments, acting as reinsurer to groups of individuals who have access to cheap information about each other, such as extended families or members of close-knit communities, who in turn offer mutual insurance to each other.

Thesis (Ph. D.)--UCLA, 2008. / Vita. Includes bibliographical references (leaves 230-231).

This thesis is a collection of essays on the relevance of property right reforms on the wellbeing of poor households in developing countries; specifically titling programmes in urban Peru. The first essay assesses the effects of titling on housing investment. The availability of a unique dataset permits us to trace households' investment behaviour. This allows us to investigate if tenure security induces households to make sizeable investments, the evolution of this effect over time, and whether heterogeneous expectations about future tenure security matter for the estimated impact. Evidence reveals significant effects, especially on large investment; however, the response on this type of investment may take several years to become effective. The second essay contributes to the debate on the sustainability of property rights reforms by emphasising the importance of strong registration systems. Policymakers have focused on the process of granting titles but the conditions needed to maintain the formality of future plot transactions have been left unattended. The analysis exploits an exogenous variation in legislation to examine the impact of a change in the registration process on the registration rate of plot transactions. Evidence suggests a large negative effect, implying that a weak registration system could threaten the reform. The third essay explores the impact of titling on the risk preferences of slum dwellers. The analysis provides evidence that titled dwellers reported lower values of the risk aversion measure than their non-treated counterparts. Results also suggest that tenure security can influence slum dwellers' preference formation process. Evidence shows that beneficiaries who were exposed to tenure security during their youth report on average lower values of the risk aversion indicator than individuals titled at an older age.

This dissertation comprises three essays on the allocation of labour and capital in a large developing country, Indonesia. In the first essay, I examine the impact of the 1998 East Asian recession on child schooling outcomes in Indonesia. Using panel data on 7-15 year-olds, I exploit the heterogeneous impact of the recession across urban communities as measured by the variation in rice price increases, under the assumption that communities where rice prices increased the most were those where real wages declined the most. I find that for the youngest children (aged 7-12 years) there is a large negative impact of higher rice prices on school attendance and no effect on labour market participation. For older children (aged 13-15 years), schooling enrolment does not respond to rice prices but labour market participation declines sharply in the worst-hit communities. I find no evidence of adverse long-term consequences on human capital formation. In the second essay, I test the hypothesis that there exists a significant earnings differential between similar workers in the formal and informal sectors. Using panel data on salaried and self-employed individuals, I find that after controlling for firm size and individual-specific heterogeneity, there is no formal sector earnings premium, except in the public sector. The results are robust to the presence of unpaid family workers, measurement error, and non-random attrition in the survey. This questions the commonly held belief that labour markets in developing countries are segmented because of legal institutions that protect high formal sector earnings. In the third essay, I estimate the effect of a large exchange rate depreciation on the performance of importers. The ability to manage volatility in the cost of imported inputs is likely to depend on a firm's access to external sources of finance as well its ability to hedge against exchange rate movements. Using data from a census on Indonesian firms, I find that while domestic importers face lower value-added due to a rise in their costs of production, foreign-owned importers fare better: they are more likely to sustain higher value-added, hire more labour and use more materials than domestic owned firms. This suggests another channel through which FDI can add value to a firm in a developing country, particularly with the increasing importance of trade in intermediate goods.

Apart from the introduction (Chapter 1) and conclusion (Chapter 7), this thesis comprises five chapters organized into two parts: Part I studies promotion incentives in the public sector, and focuses on the case of teachers in rural China. All teachers in China compete with their colleagues for rank promotions. I aim to answer two questions: first, whether the promotion system for teachers in China elicits effort from teachers, and second, how the design features of the promotion system affect effort incentives. Part I includes four chapters. Chapter 2 introduces the topic and provides a background on promotions for teachers in China. It also discusses related work in this area, and introduces the data that will be used in Part I. Chapter 3 presents and tests a theoretical model of promotions as an incentive device. The model treats all teachers as identical in terms of their ability, and as such, focuses on average levels of teacher effort. It predicts that effort is exerted in response to potential promotions. In addition, the model also predicts that average effort incentives are higher in promotion contests in which the wage gap is higher, the promotion rate is closer to one half, the number of teachers competing for a promotion is higher (for promotion rates between 1/3 and 2/3), and the average age of teachers in the contest is lower, or the proportion of female teachers is lower. The model is used to derive an estimating equation by which to test predictions on average levels of teacher effort. An equation is estimated for the probability of promotion as a function of teacher effort, which is proxied by the teachers' annual performance evaluation scores. There is simultaneity present as effort increases the probability of promotion, but it is also the promise of promotion that motivates effort. As a result, effort is instrumented using wage changes, which are both informative (higher wage gaps are associated with higher effort) and valid (wages only affect promotions through effort). The second stage of the regression demonstrates that effort is indeed exerted by teachers in order to win promotions. The first stage confirms the predictions of the model with regards to wage gaps, the promotion rate, and the size and composition of the pool of competitors. Chapter 4 extends the model of Chapter 3 in two ways: teachers are now treated as heterogeneous in ability, and a multi-period model of teacher effort over time is also added. This chapter focuses on individual levels of teacher effort, and on how the parameters of the promotion system interact with teacher characteristics to affect teacher effort. The predictions include that teachers in the extremes of the skill distribution will have lower incentives, and as the contest size increases these teachers will have effort incentives that are lower still, that teachers who are five or more years from promotion eligibility will have zero effort, as will teachers in the highest rank, that teacher effort will increase in the five years leading up to promotion eligibility, and that teacher effort will decrease after a teacher is eligible for promotion but has been passed over several times. An effort equation is estimated that captures all of these components, and the predictions are largely affirmed by the data. Tests are conducted in order to alleviate concerns about selection, as well as measurement error in the performance evaluation scores. Chapter 5 concludes Part I. Part II of this thesis looks at teacher labour markets, social distance, and learning outcomes in Punjab, Pakistan. Chapter 6 explores the link between the distribution of teachers in the labour market, caste differences between teachers and students, and child learning outcomes. Using rich longitudinal data from Pakistan that allows me to convincingly identify the causal effects of caste on learning outcomes, I show how the distribution of teachers across public schools induces particular matches of high and low caste teachers and students, and that these matches are highly predictive of test score outcomes. Specifically, low caste male children perform significantly better when taught by high caste teachers than when they are taught by low caste teachers. Several possible channels are explored, including discrimination in the classroom, role model effects, teacher quality, patronage, peer effects, and returns to education. Although the channel cannot be proven, the data points to high caste teachers being able to raise the already high returns to education for low caste children because they are able to assist these children in getting educational benefits and employment later on using their patronage networks. Low caste children therefore work harder to impress high caste teachers, and this results in higher learning outcomes.

This thesis is a collection of three essays with contributions to the intergenerational and intra-generational mobility literature. The essay on full risk insurance and measurement error examines the likelihood that measurement error may reconcile observed departures from perfect rank immobility in insurable consumption with the mobility predictions of full risk insurance, by generating spurious rank-breaking transitions. The essay shows that under certain assumptions full risk insurance predicts perfect rank immobility and that there exists ranges of error covariance matrices for which the mobility predictions of full risk insurance plus measurement error can not be rejected in the Peruvian data. A novel approach to test these mobility predictions is presented. The essay on discrete time-states Markov chain models applied to welfare dynamics shows that models with higher order may fit data better than the popular first-order, stationary model, and that the order of the chain, in turn, affects the estimation of equilibrium distributions. A best-practice methodology to conduct homogeneity tests between two samples with different optimal order is proposed, and an index by Shorrocks, based on the trace of the transition matrix, is extended to discrete Markov chain models with higher order. The essay on cohort heterogeneity in intergenerational mobility of education shows how cohort heterogeneity affects the analysis of cross-group homogeneity and long-term prospects of a welfare variable, based on transition matrix analysis. The essay compares the transition matrices of Peruvian groups divided by gender and ethnicity and finds genuine reductions in heterogeneity of the mobility regimes between male and female and between indigenous and non-indigenous groups among the youngest cohorts. The essay proposes a methodology to conduct first-order stochastic dominance analysis with equilibrium distributions and shows that among the youngest cohorts past stochastic dominance of male over females and non-indigenous over indigenous disappear in the long term.

Accounting for output per worker differences across countries has been an ongoing topic of research in economics. This thesis expands upon standard approaches by allowing for technological heterogeneity and exploiting firm and worker level data to determine the microeconomic sources of variation in both productivity and earnings. An intercontinental comparison using production functions for the Ghanaian and South Korean manufacturing sectors in Chapter 2 finds, in contrast to the conclusions of much of the macroeconomic literature, that there is no difference in total factor productivity (TFP). The microeconomic sources of the difference in value added per worker lie within the technology of firms, which is defined as the way in which inputs are used. Two important dimensions of this difference are the larger role of material inputs and the much lower rate of return to schooling in Ghana. In Chapter 3 a more general specification investigates intra-African variation in production, which is much smaller than the intercontinental difference. The pattern of cross-country heterogeneity is that, as GDP per capita rises, the relative input of materials falls, those of capital and labour rise and the returns to education increase. Differences in TFP are limited. Possible sources of the low returns to schooling in Ghana are investigated in Chapter 4 using earnings and production functions. Conditional upon selection into occupations, the only group of workers for whom education appreciably increases earnings are those employed in skilled jobs with more than ten years of education. The evidence is consistent with a lack of technological sophistication being the source of these low returns. Investment in new production processes by firms will increase the return to education and raise incomes and output. Reducing the share of intermediate inputs in production is key to the transition from low to high productivity activities. Technology is the critical element that can explain the performance of manufacturing firms across countries.

This thesis consists of five stand-alone chapters empirically evaluating questions relating to the life cycle of natural resource extraction. We use three different data sets to shed light on the local impacts of natural resource booms and busts. In chapter 2 to 4 we use the household budget survey of Kazakhstan to explore the impacts of the oil boom on the local population. In the second chapter, we explore the distributional effects of the oil boom and show that average household income increased and income inequality decreased. In the third chapter we study how the increase in average income was perceived by the local population and find that households' satisfaction with income decreased. In the fourth chapter we study how the boom affected households' expenditure and show that the likelihood that households pay tuition fees for tertiary education increased. In chapter 5, we explore the long-term impacts of a negative labour demand shock following the coal mine closures in the UK. To do this we construct a new data set containing the location of all active coal mines since 1981 and link it to the UK census. We find that the dramatic lay off of miners since 1981 was associated with a persistent reduction in female labour force participation in the affected districts. In chapter 6, we study the determinants of drilling costs and their impact on the real price of oil using a new global data set on the number of exploration wells drilled and costs of drilling. To do this, we propose a structural model of the upstream sector in the oil and gas industry. The model allows us to decompose the variation in the reduced form errors of the estimated VAR into three structural shocks, and estimate the dynamic responses of the variables in the system to these shocks. We confirm that the upstream sector of the oil and gas industry is subject to increasing costs. But we do not find that the real oil price is permanently affected by shocks to costs of drilling.

This thesis employs two new approaches to explain some of the important debates in two key economic fields: labour market economics and macroeconomic studies related to current account imbalances. Chapter 1, Chapter 2 and Chapter 3 begin a new strand of research by introducing the normal inverse Gaussian (NIG) distribution to describe unobserved heterogeneity in the labour market. The NIG distribution can be represented as a normal variance-mean mixture with the inverse Gaussian (IG) distribution as the mixing distribution. A 0.01% subsample of the 1980 US Census, comprising all men between 18 and 65 who are in the labour force, as well as a comparable sample from Ghana, is used to show that the NIG distribution provides a better fit of the log earnings function than the normal distribution. The prediction of right skewness of the log earnings distribution arising from the log normal skill Roy selection model is rejected in favour of left skewness. The thesis then extends the model to describe the distribution of log earnings conditioned on education. The same two datasets (US males and Ghanaian males) are used for the empirical analysis. We find that, once the unobserved heterogeneity is accounted for, the return to education is almost flat for lower levels of education in Ghana, and then increases for education levels greater than ten years. One of the key differences between the two datasets is that skewness and unobserved heterogeneity is a function of education for Ghana but not for the US. The NIG framework is found to be a useful tool to model this heterogeneity. Chapter 4 uses a model that allows for a rich structure of age effects similar to those predicted by the life cycle theories to argue that the demographic shifts are partly responsible for the sustained rise in the US current account deficit and the rapid increase in China's current account surplus in the last decade. However, demographics do not have an impact on the long run equilibrium or level of current accounts. Rather, they are important determinants of the short run adjustment of current accounts to their equilibrium levels. In the next twenty years, the demographic shifts are likely to push towards further current account positive adjustments in China and current account negative adjustments in the US. Developing the infrastructure, financial markets, policy tools and regulatory settings to be able to cope with the excess capital flow remains an urgent task.

Rural inhabitants of developing countries face extraordinarily risky environments, and decision-making under risk has crucial implications for the welfare of the rural poor. Therefore, obtaining a better understanding of the behaviour under risk of low-income populations is a vital step in the comprehension of human behaviour, and is important for effective policy design and evaluation, as well as for shedding light on production, investment and technology adoption decisions. In Chapter One, I analyze data collected from a laboratory experiment involving poor subjects in rural Ethiopia, in order to determine which decision models (and corresponding risk preferences) best describe the decision-making under risk of inhabitants. I find that expected utility theory (EUT) does not provide a good overall description of the decisions made by participants in the experiment; instead, there is evidence of probability weighting and loss aversion, implying that rank-dependent and reference-dependent choice models are more likely to represent the true latent decision-making process of subjects. In Chapter Two, I analyze combined experimental and survey data from rural Ethiopia in order to evaluate the determinants of risk preferences as well as assess the degree of asset integration in experimental decisions. Analyzing both EUT and non-EUT decision models and using an instrumental variable strategy, I find that household wealth negatively affects both risk aversion and loss aversion, but independent background risk has no effect on risk preferences. Further, I find evidence of narrow framing, as opposed to asset integration, suggesting that participants make decisions in the experiment in isolation from outside wealth. In Chapter Three, I analyze experimental data from Brazil to evaluate whether subjects understand decision problems that use the complex Multiple Price List (MPL) elicitation procedure, and to determine which decision models best describe observed choices. I find that the MPL decision problems of the experiment enable a finer characterization of risk preferences as compared to Ordered Lottery Selection problems (used in the Ethiopian experiment). However, I find that a significant fraction of choice patterns in the MPL problems are intransitive, and the evidence indicates that subjects did not properly understand the decision problems and thus observed choices do not reveal true risk preferences. Therefore, the relatively complex MPL procedure may not be suitable for experiments conducted with poorly-educated subjects in developing country settings. Chapter Four presents a theory outlining the relationship between rational demand for index insurance – for which the net transfer between insurer and policyholders depends only on a publicly verifiable index – and wealth. Further, the validity of this theory is tested using the experimental data from Ethiopia. In line with the theoretical model presented, due to basis risk and actuarially unfair premiums, demand for index insurance is hump-shaped – first increasing then decreasing – in wealth. The results indicate that the low take-up of this product observed among the poorest (and most risk averse) individuals in recent field studies may result from rational choice rather than credit constraints or poor decision-making.