In two essays, I present my work that applies empirical method to real-world problems in quantitative marketing and economics. In the first chapter, I use structural econometrics to investigate airlines' dynamic price competition. I show how widely-used dynamic pricing techniques affect firms profits under a competitive equilibrium. The second chapter is a joint work with Zemin Zhong. We focus on how China's anticorruption effort impact its economy, measured by car consumption and new business registration.

Dynamic pricing is becoming a common practice in many industries, but its effect under competition is uncertain due to the potential for the Prisoner's dilemma. The paper studies profit and welfare implications of competitive dynamic pricing in the context of the airline industry. The paper develops a structural dynamic oligopoly model where firms compete in selling limited capacities when facing demand fluctuations. The supply and demand are jointly estimated using a unique daily-level data on airfares and capacity utilization. The identification leverages a natural experiment of carrier exit. The estimates show that air travel demand exhibits a large degree of temporal heterogeneity and stochastic variability. The counterfactuals show that the ability to perform dynamic pricing increases total welfare. In particular, (i) price discrimination (charging late-arriving consumers higher prices) softens competition in the late market and increases profits substantially and (ii) revenue management (pricing on remaining capacities) intensifies competition and does not increase profits.

Corruption could either benefit economic growth by “greasing the wheel,” or distort supply of public goods and create inefficiency. Empirically testing the impact of corruption is difficult due to its evasive nature. We take an alternative approach by investigating the economic impacts of anti-corruption policies. We focus on China's recent anti-corruption campaign, the largest of its kind in recent history. As an important initiative of this campaign, the Communist Party's Provincial Committees of Discipline Inspection (PCDI) send inspector teams to investigate municipal governments for potential corruption. The variation in their timing allows us to use a difference-in-difference design to identify their impact on local economy. Using two unique administrative datasets of vehicle and business registration, we find that PCDI visits have a negative impact on both car sales and new business entry. For vehicles, the effect is surprisingly uniform across different price tiers: Luxury brands exhibit a similar drop as domestic brands, suggesting corruption's impact permeates households across a wide income spectrum. Over time, the effect is strengthening: We observe a 2\% drop in the first three months of PCDI visit and a 10\% drop one year afterward. The especially large impact cannot be explained by the decline in government officials' consumption behavior, suggesting anti-corruption efforts also affect the private sector. We test the idea using business registration data, and we found PCDI visits indeed discourage new business registration. We validate our empirical strategy by showing that (1) the timing of PCDI visits cannot be predicted by observable county characteristics and (2) car registrations exhibit parallel pre-treatment trends. Our results suggest there may be a trade-off in anti-corruption and economic growth.

The dissertation has three chapters. In the first chapter, I estimate consumer search cost with purchase outcome data. I analyze a model where consumers search sequentially for the best option based on advertised price and partial product information. I verify the model's ability to recover the structural parameters by a numerical experiment. Then using Nielsen Consumer Panel Data, I estimate the product-specific search costs for the top five brands in 32oz refrigerated yogurt market. I find that after controlling for the prices, the private label brand has the lowest search cost. Counterfactual analysis shows that eliminating the search cost increases overall purchase and decreases the price sensitivity. Incorrectly ignoring search frictions leads to an overestimation of own-price elasticity.

In the second chapter, we state conditions under which choice data suffices to identify preferences when consumers may not be fully informed about attributes of goods. Our approach can be used to test for full information, to forecast how consumers will respond to information, and to conduct welfare analysis when consumers are imperfectly informed. In a lab experiment, we successfully forecast the response to new information when consumers engage in costly search. In data from Expedia, our method identifies which attribute was not immediately visible to consumers in search results, and we then use the model to compute the value of additional information.

In the third chapter, we study consumers' variety-seeking preferences and explore their implications for targeted marketing using proprietary data from a food delivery platform. We document that a substantial fraction of consumers have variety-seeking preferences. Consumers, on average, are willing to pay 20% more to switch to a different seller. In the counterfactual analysis, we find that optimizing rankings by taking into account variety-seeking preferences increases revenue, consumer welfare, and purchase probability. Furthermore, we find that consumers' variety-seeking preferences soften price competition. Optimal targeted pricing implies an increase in prices for rival sellers' consumers and a decrease in prices for the sellers' own consumers.

The two essays explore two topics in marketing - consumer purchase behavior and competitive firm strategies. The first essay examines consumer stockpiling behavior in the retail gasoline market and aims to shed light on what factors affect consumer stockpiling. Past research on consumer stockpiling behavior such as Hendel and Nevo (2006) finds evidence of inter-temporal substitution by consumers that implies stockpiling behavior. However, they do not observe actual inventory or consumption and have to rely on simplifying assumptions about these quantities. I collect a novel data set of gasoline purchase history of consumers with actual inventory and consumption to test several hypotheses that relate consumer stockpiling to price, duration between purchases, and consumption. First, I find that consumers holding more inventory are more price sensitive. Higher inventory increases the impact of price on purchase decisions and those with higher inventory can afford to do more price search before making a purchase. I also find that all else equal, consumers will reduce consumption following a purchase made during high prices; that consumers with lower inventory have a higher probability of purchasing; and that duration from previous purchase is shorter for purchases made during low prices and longer during high prices.

The second essay examines competition in a dynamic setting between Wal-Mart and Target in the context of location choices and expansion strategies. Studying this topic sheds light on how an industry evolves and how the market structure is shaped by decisions such as when to enter and exit and where to locate new stores as well as the driving force behind different expansion strategies. One of the early papers by Bresnahan and Reiss (1991) study entry of retail and professional services into isolated markets in a one-shot game. In their model, firms are allowed to enter once and open one store. Collard-Wexler (2014) estimates a model of investment and entry in the ready-mix concrete industry. Again, each firm is assumed to own a single plant in the model. In my paper, I build a structural model in which firms can open multiple stores over a period of time. This setting is closer to reality as competition between firms lasts over many periods with firms making regular entry decisions and often opening multiple stores in the same market. I estimate how entry decisions are affected by competitor's presence and market characteristics and learn about the evolution of market structure and expansion strategies of the firms. The firms are forward-looking and engage in Markov perfect equilibrium (MPE) strategies. The results show that firm profits are affected asymmetrically by the competitor's presence. First, Wal-Mart is the dominant firm with higher profits regardless of Target's presence. On the other hand, Target's profits depend a lot on Wal-Mart's presence. It must have more stores than Wal-Mart (called store advantage) to profit. I also find asymmetry in the expansion strategies of the two firms. Wal-Mart tends to explore new and smaller markets by being the first and often the only firm to enter, while Target tends to focus on major markets with high population and GDP and strives to maintain store advantage over Wal-Mart by matching or outdoing Wal-Mart's decision to open new stores.