Using empirical methods, this paper examines household schooling and child labor decisions in rural Bangladesh. The results suggest the following: poverty and low parental education are associated with lower schooling and greater child labor; asset-owning households are more likely to have children combine child labor with schooling; households choose the same activity for all children within the household, regardless of gender; there is a weak association between direct costs and household decisions; finally, higher child wages encourage households to practice child labor. © 2007 Elsevier Inc. All rights reserved.

This study estimates the returns to boys' education for rural Bangladeshi households by accounting for some conventionally neglected items: direct costs of education, foregone child labour earnings, and option value. The estimated returns are 13.5% for primary education, 7.8% for junior-secondary education, 12.9% for higher-secondary education, and 9.7% for higher education; the resulting option value from primary education is 5.3%. These results suggest that there is economic rationale for non-poor rural households to invest in boys' education, especially at the primary level.

This dissertation examines the reasons for which a seller may decide to conduct a multi-unit auction sequentially rather than simultaneously. It analyzes the manner in which the information generated during a sequential auction can affect bidding to the seller's benefit and demonstrates the requirement of intertemporal commitment to the auction rules.When the seller cannot commit not to alter the reserve price over time, the bidders are reluctant to reveal their valuations. Therefore, with single-unit demands, a symmetric monotone equilibrium exists only in a sequential Dutch auction. In the earlier rounds of this auction, because of the anticipation of lower reserve prices in the future, some buyers prefer not to submit a bid, although their valuations exceed the requested reserve price. Furthermore, any buyer submitting a bid shades it sharply. Consequently, under imperfect commitment, the optimal sequential auction results in lower expected revenue than its simultaneous counterpart.In the presence of allocative and informational externalities, in particular, in a sale of two oligopoly licenses, a sequential auction succeeds in eliminating some of the payoff uncertainty by allocating the licenses in an ordered manner, according to the bidders' strength. Therefore, the weaker oligopolist can acquire his license at a lower price than the one he would pay in a simultaneous auction. In addition, he can avoid overpaying. Conversely, the stronger oligopolist pays a higher price, so that, when the bidders' production costs are independent, the two auction schemes generate the same expected revenue. Therefore, without affecting the seller's revenue or efficiency objectives, the sequential auction results in a more equal distribution of the wealth generated by the oligopoly.Finally, when the preparation or submission of bids is costly, so that a buyer will not enter the auction unless he expects a substantial gain from it, low prices in the earlier rounds of a sequential sale trigger stronger participation, and, consequently, higher prices in the later rounds. A sequential auction, therefore, may result in higher expected revenue than a simultaneous sale, especially when the number of potential bidders is large, the participation cost small, or the distribution of valuations convex.

Adaptive learning is important in dynamic models since it is a process that shows the improvement in the understanding of the agents of the model. Whenever there is a dynamic environment, there is a room for improvement through learning. In this thesis I analyze the adaptive learning of the agents in different setups. In my first paper I show that adaptive learning does not eliminate the multiplicity of stationary equilibria in the Diamond overlapping generations model with money and productive capital; both dynamically efficient and inefficient equilibria are found to be stable under adaptive learning. In my second paper I show that the two agents of a natural-rate model, with different beliefs, learn the economy which leads to convergence or endogenous fluctuations of the inflation rate under different conditions. And in my last paper I show that a central bank with an extraneous instrument, "cheap talk" announcements, can influence the private sector to achieve better outcomes than could be obtained by manipulating the nominal interest rate alone with full knowledge of private sector expectation formation and in anything less than full knowledge, the private sector learns to discount announcements.

In the first chapter, ¡°All-Pay Auctions with Resale¡±, I study equilibria of first- and second-price all-pay auctions with resale when players' signals are affiliated and symmetrically distributed. I show that existence of resale possibilities introduces an endogenous element to players' valuations and creates a signaling incentive for players. I characterize symmetric bidding equilibria for both first- and second-price all-pay auctions with resale and provide sufficient conditions for existence of symmetric equilibria. Under those conditions I show that second-price all-pay auctions generate no less expected revenue than first-price all-pay auctions with resale. The initial seller could benefit from publicly disclosing her private information which is affiliated with players' signals.Outcome in all-pay auctions is deterministic since the highest bidder wins the prize with probability one. However, many realistic contests have in-deterministic outcome and no player can guarantee winning the prize. The second chapter, ¡°Rent-Seeking Contest with Private Values and Resale¡±, studies rent-seeking contests with private values and resale possibilities. With an in-deterministic success function, the resulting possible inefficiency creates a motive for aftermarket trade. Players' valuations are endogenously determined when there is an opportunity of resale. I characterize symmetric equilibria. I assume that the winner has full bargaining power; however, the results extend to other resale mechanisms. I show that resale enhances allocative efficiency ex post at the expense of more wasted social resources since players compete more aggressively with resale possibilities.In the third chapter, ¡°The Imperfectly Discriminating Contests with Incomplete Information¡±, I study the existence of monotone pure-strategy equilibria in imperfectly discriminating contests with incomplete information. Sufficient conditions under which equilibria exist are provided for both finite-action and continuum-action cases. Using a two-bidder example, we derive some properties of equilibria and show a special case of revenue equivalence between contests with incomplete information and contests with complete information.

This dissertation consists of two chapters on different topics in macroeconomics. The first chapter studies the economic effect of privately-issued banking notes (private money). I build a model in which money is divisible and price is endogenous. The analysis shows that the private banknotes serve as short term credit for bankers. Given that the Friedman rule, which suggests a deflationary monetary policy, is not available, privately-issued banking notes improve resource allocation in the economy. The welfare improvement is not restricted to bankers. Nonbankers, though they have no right to issue private money, also enjoy welfare improvement. The second chapter is based on a paper co-authored with Ying Fang, who was my fellow graduate student at the University of Pittsburgh and now is a professor at Xiamen University. In this chapter we use crossing-city data to estimate the effect of property rights protection on China's economic performance. We adopt the historical enrollments of Protestant lower primary schools in the early 20th century as the instrumental variable for current property rights protection in China. We find that property rights protection dominates others potential determinants of economic performance, such as geography or government policy.

The fundamental question I address in the dissertation is how the behavior of economic agents interacts with networks of relationships which underlie a wide set of economic situations. In Ch. 2, entitled "Decentralized Information Sharing in Oligopoly," I analyze the incentives of firms for information sharing in a decentralized environment when firms face a stochastic demand. In order to do that, I develop a two stage model of strategic network formation, where a cooperative network formation stage is played in the first stage and a noncooperative Bayesian Cournot game is played in the second stage. I derive pure strategy mixed cooperative and noncooperative equilibria that are subgame perfect and stable, and characterize the resulting network structures. Ch. 3, entitled "A War of Attrition in Network Formation," investigates the strategic behavior of agents when they face a decision on the formation of relationships. I apply a war of attrition to the dynamic network formation process when links among agents have characteristics of public goods. Agents are randomly but exogenously matched in each stage. Based on Bala and Goyal's (2000) two-way flow model, I characterize the subgame-perfect equilibrium outcomes and discuss their efficiency. Finally, Ch. 4, entitled "Social Norms and Trust among Strangers," (with Huan Xie) studies the development of trust and reciprocity among strangers in the indefinitely repeated trust game with random matching. If reputation is attached to the community as a whole and if a single defection leads to the destruction of the cooperative social norm through contagious punishments, the cooperative social norm can be sustained by the self-interested community members in the equilibrium. We provide sufficient conditions that support the social norm of trust and reciprocity as a sequential equilibrium.

Limited dependent variable (LDV) panel data models pose substantial challenges in maximum likelihood estimation. The likelihood function in such models typically contains multivariate integrals that are often analytically intractable. To overcome such problem in a panel probit model with unobserved individual heterogeneity and autocorrelated errors, in Chapter 1 - co-authored with Roman Liesenfeld and Jean-François Richard - we perform classical and Bayesian analysis of the model based on the Efficient Importance Sampling (EIS) technique (Richard and Zhang, 2006). We apply our method to the product innovation activity of a panel of German manufacturing firms in response to imports and foreign direct investment confirming their positive effects. Nonetheless, our key coefficient estimates are smaller than found in previous literature which can be explained by our flexible model assumptions. The remaining two chapters present my work on new estimation methods for models based on conditional moment restrictions. Such models are frequently stipulated by economic theory but only a few estimators based directly on them have so far been analyzed in the literature. Indeed, estimation of parameters therein poses a difficult ill-posed inverse problem. Rather, these models are typically converted into unconditional moment restrictions that are easier to handle. However, such conversion results in a loss of information compared to the original specification. Using the information-theoretic framework of so-called Generalized Minimum Contrast (GMC) estimation, in Chapter 2 I propose a new class of estimators based directly on conditional moment restrictions that encompasses the entire GMC family. Moreover, I show that previous literature covering a few special cases of the GMC class use an arbitrary uniform weighting scheme over the space of exogenous variables that can be improved upon with optimal local weighting. All currently available GMC estimators are based on moments containing finite-dimensional Euclidean parameters. To alleviate a potential misspecification problem resulting from strong parametric assumptions, in Chapter 3 I propose a new Sieve-based Locally Weighted Conditional Empirical Likelihood (SLWCEL) estimator containing also infinite dimensional unknown functions, thus extending a special case of Chapter 2 to the semiparametric environment. Much of Chapter 3 is devoted to analysis of SLWCEL's asymptotic properties.

This dissertation consists of three theoritical chapters. In the first chapter, I study anevolutionary model with a finite population of boundedly rational agents, who do not haveaccess to the same amount of information. Time is discrete, and in each period two agentsare paired to play a 2 × 2 symmetric coordination game. Each player can cross paths withtwo kinds of opponents: Neighbors or Strangers. If a player faces a Neighbor, she can accesssome information about her opponents past plays, and plays using a myopic best-response.But if she faces a Stranger, she does not have access to any information, and therefore playsaccording to a casebased decision rule. I show that in the short run, segregated localitiesemerge, to finally disappear in the long run, in favor of the Pareto Efficient convention. Themain contribution of this chapter is that I show that agents coordinate in an evolutionaryframework on an efficient outcome, even when information is asymmetric, without assumingany pre-play communication or mobility of the agents.In the second chapter (with Alexander Matros) we consider K finite populations ofboundedly rational agents whose preferences and information differ. Each period agents arerandomly paired to play some coordination games. We show that several special (fixed)agents lead the coordination. In a mistake-free environment, all connected fixed agents haveto coordinate on the same strategy. In the long run, as the probability of mistakes goes tozero, all agents coordinate on the same strategy. The long-run outcome is unique, if all fixedagents belong to the same population.The last chapter (with Alexander Matros) considers a public good game similar to the oneiiiin Eshel, Samuelson and Shaked [14], which benefits are only local. We find some sufficientconditions which when applied to a particular set of graphs ensure the survival of Altruism.

<p>The first chapter of my thesis develops and estimates a dynamic<br />structural partial equilibrium model of schooling and work<br />decisions. The estimated model explicitly accounts for the<br />simultaneous choice of enrolling in school and working. It also<br />allows for endogenous leisure choices, intertemporal<br />nonseparabilities in preferences, aggregate skill specific<br />productivity shocks, aggregate consumption price effects, and<br />individual heterogeneity. Times spent on schooling, working, and<br />leisure are treated as continuous choice variables. The estimated<br />model is solved and two counterfactual simulation exercises are<br />performed. The first is the case where a subsidy is given to<br />individuals who enroll in school and do not participate in the labor<br />market. The second is the case where the demands of the school<br />curriculum are increased so that a young man enrolled in school<br />necessarily spends more time studying. The conclusion is that the<br />latter policy is more effective in enhancing educational<br />achievements and future wages.</p><p>The second chapter of my thesis develops a semiparametric estimator<br />for a dynamic nonlinear single index panel data model. Flexibility<br />of the model is achieved by assuming that the index function is<br />unknown. Flexibility in individual heterogeneity is achieved by<br />assuming that the individual effect is an unknown function of some<br />observable random variable. The paper proposes an algorithm that<br />estimates each of the finite and infinite dimensional parameters. In<br />particular, the full data generating process is estimated. This is<br />important if the predicted outcomes are used as plug-in estimators,<br />as in the multistage estimation of dynamic structural models.</p><p>The final chapter of my thesis develops a powerful new algorithm to<br />solve single object first price auctions where bidders draw<br />independent private values from heterogeneous distributions. The<br />algorithm allows for the scenario in which groups of symmetric and<br />asymmetric bidders may collude, and for the auctioneer to set a<br />reserve price. The paper also provides operational univariate<br />quadratures to evaluate the probabilities of winning as well as the<br />expected revenues for the bidders and the auctioneer. The expected<br />revenue function is used to the compute optimal reserve under<br />asymmetric environments.</p>

The objective of this study is to look at factors affecting the decision of early retirement for Egyptian government sector employees. The empirical analysis is based on a 2005 nationally representative sample of 3437 government sector workers, ranging in age between 50 and 57 years. The study’s main findings are: the age of 55 is shown to be the age of choice for early retirement for both men and women; women are more likely to retire earlier than men; good health status is associated with longer stay in the job; women are more educated than men; the level of education is not a determining factor for women, but it is for men; men without a university degree are more likely to retire earlier than others with a university degree; men plan to work after their early retirement; the presence of the working wife has a positive effect on her husband’s decision to retire early, yet a working husband discourages his wife to retire early. Meanwhile, the policy implications discussed are: timing of the announcement of the plans; the question of voluntary vs. compulsory early retirement plans; the potential outcome of excessive payoffs and curtailing adverse selection by targeting redundant employees and preserving the most productive; and the need for the government to provide early retirees with the necessary skills and training to face their post retirement years.