Individuals living in society are bound together by a social network, the complex of relationships that brings them into contact with other agents. In many social and economic situations, individuals learn by observing the behavior of others in their local environment. This process is called social learning. Learning in incomplete networks, where different agents have different information sets, is especially challenging: because of the lack of common knowledge individuals must draw inferences about the actions others have observed as well as about their private information. Whether individuals can rationally process the information available in a network is ultimately an empirical question. This paper reports an experimental investigation of learning in three-person networks and uses the theoretical framework Gale and Kariv (2003) to interpret the data generated by the experiments. The family of three-person networks includes several nontrivial architectures, each of which gives rise to its own distinctive learning patterns. We find that the theory can account for the behavior observed in the laboratory in variety of networks and informational settings. To account for errors in subjects’ behavior, we adapt the model of Quantal Response Equilibrium of McKelvey and Palfrey (1995, 1998) and find that its restrictions are also confirmed. The ‘goodness of fit’ is better for the QRE model than for the game-theory model. This provides important support for the use of QRE to interpret experimental data.

This dissertation consists of three chapters, which study the identification of people's reasoning abilities and belief levels in strategic interactions. The first chapter examines the methodology to identify subjects' reasoning levels in the ring games proposed by Kneeland (2015). In the second chapter, an experiment is designed to identify whether it is limited reasoning ability or incorrect belief that prevents people from playing equilibrium strategies. The experiment builds on a series of ring games similar to the ones analyzed in Chapter 1. The third chapter links people's performance in games to a measure of their cognitive ability, which also validates the findings in Chapter 2.

The ring games are first used by Kneeland (2015) to separate different orders of belief in rationality. Her experiment finds rationality levels higher than those in the literature. The first chapter points out two features in the ring games used by Kneeland (2015) that could potentially bias type classification, and reports the experimental results using two new sets of games with these features removed. Similar type distribution as seen in Kneeland (2015) is reproduced when the rows with the largest sums in the payoff matrices do not coincide with the equilibrium strategies. However, when some rows that have the lowest sums and contain zeros are set to be the best responses, a considerably larger fraction of low types are found. Additionally, the deviation rates from the theoretical prediction are higher in my data, partly due to the lesser performance of the identifying assumption.

The second chapter reports an experiment to identify the decisive factor that prevents people from using more depth of reasoning: their incorrect belief of others' rationality or their limited reasoning ability. The analysis first classifies subjects into different Lk types by the reasoning steps they use in the games. It then distinguishes between the "Lkb" players, who have high ability and best respond to Lk belief, and the "Lka" players, who could use, at most, k steps of reasoning, and thus could not respond to L(k+1) or higher-order belief. The separation utilizes a combination of simultaneous and sequential ring games. In the sequential games it requires more than k reasoning steps to respond to Lk belief, so Lkb players still best respond but Lka would fail. I find that around half of the L2 and L3 subjects are best responding to L2 or L3 belief, while the rest have reached their upper boundaries of reasoning. The findings suggest that both belief and reasoning ability could be the decisive factors of players' observed levels.

The third chapter finds positive relationship between the Cognitive Reflection Test (CRT) scores, an indicator of cognitive ability, and the reasoning levels in the ring games in Chapter 2. The CRT scores increase with levels among the lower types (from L0 to L2b), and but do not differ significantly from L2b to L4. Striking difference is found between the CRT scores of L2a and L2b subjects, which suggests that cognitive ability are strongly related to reasoning skills.

This dissertations includes three (3) chapters, each adding to the growing network games literature that incorporates incomplete information. Financial over-the-counter markets give motivating applications. (1) "Trading Networks and Equilibrium Intermediation" studies the efficiency of trade in networks. A network of intermediaries facilitates exchange between buyers and a seller. Intermediary traders face a private trading cost, a network characterizes the set of feasible transactions, and an auction mechanism sets prices. Stable networks, which are robust to agents' collusive actions, exist when cost uncertainty is acute and multiple, independent trading relationships are valuable. A free-entry process governs the formation of equilibrium networks. Such networks feature too few intermediaries relative to the optimal market organization and they exhibit an asymmetric structure amplifying the shocks experienced by key intermediaries. (2) "Interdealer Trade: Risk, Liquidity, and the role of Market Inventory" further studies traders facing private shocks, placed in a dynamic setting. Trades between ex ante symmetric, inventory carrying intermediaries ("dealers") are motivated by divergent liquidity needs of the counter parties. Market prices and asset flows are pinned by dealers' indifference between providing intermediation services and retaining liquidity to be utilized in subsequent interdealer markets. More active interdealer markets simultaneously increase the value to intermediation and the option-value to providing these services. Under infrequent shocks, interdealer trade boosts the availability of liquidity in the broader market. This boost decays with market inventory, which serves as a constraint on interdealer activity. Through this market mechanism, prices vary inversely with both search frictions between dealers and on their total current holdings. (3) "Information Acquisition and Response in Peer-effects Networks" endogenizes the quality of information that market participants carry in a general peer effects model. When pairwise peer effects are symmetric, asymmetries in acquired information are inefficiently low relative to the utilitarian benchmark. And with information privately acquired, all players face strictly positive gains to overstating their informativeness as to strategically influence the beliefs and behaviors of neighbors. If strategic substitutes in actions are present and significant, low centrality players move against their signals in anticipation of their neighbors' actions. A blueprint for optimal policy design is developed. Applications to market efficiency in financial crises and two-sided markets are discussed.

This dissertation consists of three chapters exploring the role of social norms and social image in economic decision-making. A combination of theory and experiments is used to understand how certain social preferences might form, how they influence our individual choices through both internal and external motivation, and how they might aggregate into societal norms when individuals are externally pressured to appear to be doing the right thing.

The first chapter examines how individual beliefs, or personal norms, interact with the desire to maintain a positive social image when individuals disagree about what should be done in that situation. This commonly occurs in settings such as partisan politics or when moral norms are in flux. Traditional notions of social norms cannot describe these situations, and there is correspondingly no unambiguously ``good'' action that social pressure can promote. I develop two alternative, psychological game theoretic models that explain how social pressure affects behavior even if individuals do not agree on the norm. One channel is ``approval seeking'', in which individuals want their peers to approve of their actions. Another channel is ``respect seeking'', in which individuals want to be known for strict adherence to their personal norms. Approval seekers pool on one option as social pressure increases, thus creating the illusion of societal consensus. This can potentially leading to destructive posturing, in which damaging norms are perpetuated. Respect seekers, on the other hand, are less hypocritical the more social pressure increases, and are accordingly less willing to compromise. Respect is thus a less likely force to produce a societal consensus norm. These results demonstrate that using social pressure to promote a certain behavior may backfire if it targets the wrong kind of social image.

In the second chapter, Ulrike Malmendier, Roberto Weber and I explore the case of the norm of reciprocity, both positive and negative. Reciprocal behavioral has been found to play a significant role in explaining outcomes in many important economic domains. However, despite mounting empirical evidence, economists still struggle to converge on the correct model of the underlying motives. Existing theories posit internal preferences for the welfare of others, inequality aversion, or utility from repaying others' kindness. Recent evidence reveals that `one-sided' acts of kindness, not involving reciprocity, exhibit a large degree of reluctance, with people trying to avoid opportunities to act generously, suggesting that external factors such as social image, self image, or social pressure are important determinants of unilateral sharing. However, this revision of conventional theoretical motives for sharing has had little spillover to `two-sided' reciprocity environments, where one individual responds to the actions of another. We review the literature on reciprocity and point to the relative lack of attention paid to external factors. We then present a novel experiment that explores the importance of internal versus external factors in driving reciprocal behavior. We find that, in a laboratory reciprocity setting (the double-dictator game), failure to account for external motives leads to a significant overestimation of internal motives such as fairness and altruism. We use the experimental data to illustrate the importance of combining reduced-form and structural analyses in disentangling internal and external determinants of pro-social behavior.

In the third chapter, Pamela Jakiela, Edward Miguel and I look deeper into the origins of norms. We combine data from a field experiment and a laboratory experiment to measure the causal impact of human capital on respect for earned property rights, a component of social preferences with important implications for economic growth and development. We find that higher academic achievement reduces the willingness of young Kenyan women to appropriate others' labor income, and shifts players toward a 50-50 split norm in the dictator game. This study demonstrates that education may have long-run impacts on social preferences, norms and institutions beyond the human capital directly produced. It also shows that randomized field experiments can be successfully combined with laboratory experiment data to measure causal impacts on individual values, norms, and preferences which cannot be readily captured in survey data.

Consider a first-price, sealed-bid auction where participants have affiliated valuations and private budget constraints; that is, bidders have private multidimensional types. We offer sufficient conditions for the existence of a monotone equilibrium in this environment along with an equilibrium characterization. Hard budget constraints introduce two competing effects on bidding. The direct effect depresses bids as participants hit their spending limit. The strategic effect encourages more aggressive bidding by participants with large budgets. Together these effects can yield discontinuous equilibrium strategies stratifying competition along the budget dimension. The strategic consequences of private budget constraints can be a serious confound in interpreting bidding behavior in auctions. Evidence from an experimental auction market lends support to the strategic importance of budget constraints.

Theories of justice in the spirit of Harsanyi and Rawls argue that fair-minded people should aspire to make choices for society — that is, for themselves and for others — as if in the original position, behind a veil of ignorance that prevents them from knowing their own social and economic positions in society. While the original position is a purely hypothetical situation, developed as a thought experiment, the main result of this paper is that (under certain assumptions) preferences — hence choices — behind the veil of ignorance are determined by preferences in front of the veil of ignorance. This linkage between preferences behind and in front of the veil of ignorance has implications for distributive theories of justice and for theories of choice.

Recent experimental economics research documents substantial gender differences in the willingness to compete in a tournament, suggesting that women may be underrepresented in positions of power because they lack competitive drive. The second chapter of this dissertation examines whether competitive inclination measured in the lab is predictive of the subsequent propensity to take a competitive and highly consequential high school entrance exam in rural China. Using a discrete choice mixed logit model, I estimate individual parameters of competitive inclination from lab data for my sample of ethnically diverse middle school students. I find that a middle school student with a taste for competition one standard deviation above the mean is 7.2 percentage points more likely to take the exam, controlling for prior test scores. Contrary to results from studies in adult populations, no gender differences were found in competitive inclination. Likewise, no gender difference was found in exam taking behavior, controlling for prior test scores.

The underrepresentation of women in leadership roles could also reinforce gender differences in competitive inclination in a "low" equilibrium. The third chapter of this dissertation studies adult competitive inclination in China, a country which has deliberately sought to move to a "high" equilibrium through massive coordinated government action. Competitive inclination is elicited from three ethnic groups in a "cultural laboratory": the majority Han Chinese, a matrilineal minority group, and a patrilineal minority group. The Han Chinese exhibited no statistically significant gender differences in competitive inclination. Women in the patrilineal minority, where clan leaders arrange marriages, are less competitively inclined than the men, and the men are as competitively inclined than the Han Chinese. Women in the matrilineal minority are as competitively inclined as the Han Chinese but the men are more competitively inclined. This non-intuitive result may be explained by the special non-exclusive nature of "marriages" in the matrilineal group. These findings suggest that coordinated government action can shape competitive preferences, which have the potential to lead to organic increases in gender diversity in leadership roles. More broadly, they support the case for quotas for women in parliament and corporate boardrooms.

This paper reports an experimental test of individual preferences for giving. We use graphical representations of modified Dictator Games that vary the price of giving. This generates a very rich data set well-suited to studying behavior at the level of the individual subject. We test the data for consistency with preference maximization, and we recover underlying preferences and forecast behavior using both nonparametric and parametric methods. Our results emphasize that classical demand theory can account surprisingly well for behaviors observed in the laboratory and that individual preferences for giving are highly heterogeneous, ranging from utilitarian to Rawlsian to perfectly selfish.

This dissertation studies individual decision making over risky assets in the context of choices from continuous budget sets.

The first chapter presents theoretical work modeling deviations from rational decision-making as errors in an individual's choices. In this chapter, I address the fundamental question, how can we identify mistakes from irrational behavior? Numerous studies in both the lab and field have found evidence of behavior that is inconsistent with rational choice. In a departure from existing indices of goodness-of-fit, this paper develops an intuitive measure of rationality estimated on a revealed preference relation. The proposed metric is a set estimate of the error rate necessary to explain deviations from economic rationality. The error rate arises as a parameter in a contaminated choice model and appears again in the resultant revealed preference relation, which is also contaminated with error. This random revealed preference is a new type of random graph with a unique dependency structure. The random graph is distributed on a space of feasible revealed preference relations and the distribution is identified by the unknown error rate and some ``true'' deterministic relation. Applying weak assumptions on this random graph model equates acyclicity with the general axiom of revealed preferences. A proposed hypothesis test estimates the unknown error rate conditional on the true relation begin acyclic using the expected number of cycles of length two. The result is a set estimate bounding the unknown error rate under the null. A proposed subsampling approach is shown through simulations to be computationally efficient. % emperical results from a lab experiment verify the simulations.

The second chapter presents empirical work measuring decision-making outside the lab, using a smartphone lab-in-field design. This is based on joint work with Shachar Kariv and Raja Sangupta. At the heart of Economics is the principle of rationality, namely that an individual's choices are determined by a fixed, if unknown, set of preferences over alternatives. Numerous lab experiments have found that most people demonstrate rational decision making, even when allowing for individual heterogeneity in preferences. However, questions of external validity remain- is the observed rationality due to the lab setting itself and the short time period over which choices are typically observed? To answer this we leverage the recent widespread adoption of smartphones and the resulting unprecedented access to individuals as they go about their daily lives. We present an experiment that takes a computer-based experiment measuring rationality out of the lab and onto the smartphone. Individuals participate in the experiment on their smartphones, while going about their lives, over a period of five days. We measure the rationality of each individual's choices over this period and find that the distribution of rationality across the sample does not differ from that found in lab experiments. The results of our experiment provide evidence for temporal and environmental external validity of lab experiments. Additionally we collect background data from built-in phone sensors and find no significant correlations between observable behavior and rationality.