Many firms rely on salespersons to communicate with prospective customers. Such person-to-person interaction allows for two-way discovery of product fit and flexibility on price, which are particularly important for business-to-business transactions. In the first chapter, I model the sales process as a game in which a buyer and a seller discover their match sequentially while bargaining for price. The match between the product's attributes and the buyer's needs is revealed gradually over time. The seller can make price offers without commitment, and the buyer decides whether to accept or wait. Players incur flow costs and can leave at any moment. The discovery process creates a hold-up problem for the buyer that causes him to leave too early and results in inefficient no-trades. This can be alleviated by the use of a list price that puts an upper bound on the seller's offers. A lower list price encourages the buyer to stay while reducing the seller's bargaining power. But in equilibrium the players always reach agreement at a discounted price. The model thus provides a novel rationale for the pattern of ``list price - discount'' observed in sales. I examine whether the seller should commit to a fixed price or allow bargaining. When the seller's flow cost is high relative to the buyer's, both players are willing to participate in discovery if and only if bargaining is allowed. In such a case, bargaining leads to a Pareto improvement, which explains the prevalent use of bargaining in sales. If the buyer has private information on his outside option, the model predicts that, counter-intuitively, the buyer with a higher net value for the product pays a lower price. The chapter expands the bargaining literature by adding a discovery process that introduces a hold-up problem as well as making the product value stochastic.

The second chapter examines how counter-offers affects the hold-up problem in stochastic bargaining. Firms increasingly rely on collaboration for the development and marketing of products. The expected surplus from such collaboration can change stochastically over time due to evolving market conditions or the arrival of new information. For collaboration to happen, both firms have to agree to collaborate as well as agree on how the profit is to be split. In such cases, at what point do firms form the alliance and how do they agree on the profit split? To answer these questions, I study a model of bilateral bargaining with a surplus that follows a Brownian motion. One firm can make repeated-offers to the other, and they switch roles after some time to allow for counteroffers. The frequency of counteroffers determines relative bargaining power, and the model captures different bargaining procedures by varying this frequency. The chapter shows that, when there is no outside option, firms collaborate after efficient delay. If there is a relevant outside option, the outcome is inefficient due to the existence of a hold-up problem faced by the weaker party. Firms form the alliance too early, taking the outside options too early, and the ex-ante probability of alliance becomes sub-optimal. Increasing the frequency of counteroffers improves social efficiency by balancing bargaining power and reducing the severity of hold-up. Furthermore, bargaining with more frequent counteroffers can lead to Pareto improvements; the proposer benefits, too, because the increased efficiency outweighs losses in bargaining power. The essay makes a step in understanding the effect of bargaining procedures on collaborative outcome, and shows how collaborators should (not) bargain.

The third chapter studies the effect of product labelling on consumer behavior empirically. Cigarettes are sold in different strengths, commonly categorized as regular, light, or ultralight. In 2009, Congress passed Tobacco Control Act (TCA) which banned tobacco companies from communicating product strengths to consumers on any marketing or packaging materials. Cigarette companies continue to sell products of different strengths by using less informative color codes, i.e., relabeling Marlboro Light to Marlboro Gold or Camel Light to Camel Blue. Brands do not use the exact same color codes, creating room for confusion. This chapter investigates the effect of such change in label informativeness on consumer choice. Using a panel of smokers from 2007 to 2012, I find a sharp decline in price sensitivity after Tobacco Control Act was passed. The finding is robust in choice models that account for preference heterogeneity, state dependence, price endogeneity, and consideration sets. This result suggests that consumers perceive products as more differentiated when strength labels change to color codes. This essay provides new evidence on the linkage between product labeling and choice behavior.

The dissertation has two chapters. In the first chapter, called “Optimal Policies for Differentiated Green Products: Characteristics and Usage of Electric Vehicles”, I study the issue of policy design for electric vehicles. When designing policies for electric vehicles (EVs) policymakers need to decide how to allocate policy support among EVs with different characteristics, since different EVs are likely to have differences in attractiveness to consumers and usage patterns and, hence, differences in environmental impact. In this paper, I build and estimate a structural model of the U.S. auto market that is able to predict market shares and usage patterns of electric and traditional vehicles with different characteristics under various market conditions and is able to assess the effects of policies differentiating on characteristics of EVs. On the demand side, I introduce the concept of consumer inconvenience costs of charging, which depend on EV battery range, charging infrastructure development level, consumer's driving needs, and other individual-specific factors. On the supply side, I model firm choice of prices and battery ranges. The estimation results show that the inconvenience costs have a dramatic effect on consumer purchase decisions and usage patterns of EVs, and, hence, their environmental impact. Also, the results indicate that firms are more likely to adjust battery ranges when policy support depends on battery range. I use the model estimates to evaluate the effects of two major U.S. policies for EVs, the federal subsidy and California's Zero Emission Vehicle regulation, on the environment, consumer surplus, firm profits, and social welfare. I also experiment with alternative structures of the federal subsidy that differentiate on type and battery range of EVs. I find that more efficient structures can improve the environmental effect of the subsidy by 4.6% and the welfare effect by 1.6%. Interestingly, the more efficient structures result in fewer EVs sold, but in more electric miles traveled and more gasoline miles replaced.In the second chapter, called “Do Big Businesses Influence Media? The Case of Amazon.com and The Washington Post.”, I study whether media outlets bias their coverage of the news about their owners or companies the owners have vested interests in. To shed some light on this question, I look at how the acquisition of the Washington Post, a major U.S. daily newspaper, by Jeff Bezos, the founder and CEO of Amazon.com, affected the coverage of the news about Bezos and Amazon.com. Using data on news stories in several major newspapers, I document that the acquisition resulted in an increase in the number of mentions of Bezos, Amazon, or Amazon’s products in the Washington Post, relative to other newspapers and news stories about other big tech companies. From a simple sentiment analysis, however, I found no evidence of change in the sentiment of the stories. I discuss potential mechanisms that can explain the results, including a conflict of interest, a shift in preferences of the readership, improved access of the Washington Post to information about Bezos and Amazon, and a shift in the beliefs of the newspaper’s editors and journalists about the importance of news about Amazon and Bezos.

This dissertation studies the role of information in marketing strategy. Specifically, it focuses on two key questions: How can consumers optimize their information search behavior to make better decisions? And in response to this, how can firms increase their profits by influencing consumers' information acquisition processes? The dissertation consists of three chapters of independent work.

Consumers frequently search for information before making decisions. Since their search and purchase decisions depend on the information environment, firms have a strong incentive to influence it. In the first essay, I endogenize the consumer’s information environment from the firm’s perspective. We consider a dynamic model where a firm sequentially persuades a consumer to purchase the product. The consumer only wishes to buy the product if it is a good match. The firm designs the information structure. Given the endogenous information environment, the consumer trades off the benefit and cost of information acquisition and decides whether to search for more information. Given the information acquisition strategy of the consumer, the firm trades off the benefit and cost of information provision and determines how much information to provide. This paper characterizes the optimal information structure under a general signal space. We find that the firm only smooths information provision over multiple periods if the consumer is optimistic about the product fit before searching for information. Moreover, if the search cost for the consumer is high, the firm designs the information such that the consumer will be certain that the product is a good match and will purchase it after observing a positive signal. If the search cost is low, the firm provides noisy information such that the consumer will be uncertain about the product fit but will still buy it after observing a positive signal.

When considering whether or not to buy a product, the consumer can often evaluate different attributes of it. Sometimes, the consumer chooses which attribute to search for because of exogenous reasons (e.g., one attribute is more important than others). However, the consumer often is unclear which attribute is more important ex-ante. Assuming that a product has two symmetric attributes, the second essay characterizes the optimal search strategy of the consumer by endogenizing the optimal attribute to search, when to keep searching for information, and when to stop searching and make a decision. We characterize the search region by a set of ordinary differential equations for moderate beliefs and by a system of equations for extreme beliefs. We find that it is always optimal for the consumer to search the attribute about which she has the higher uncertainty due to the faster speed of learning. The consumer only searches for the more uncertain attribute if she holds a strong prior belief about one of the attributes, and may search for both attributes otherwise. We then show how firms can influence consumers' search behavior and increase profits by informative advertising. The firm does not advertise if the consumer's prior beliefs about both attributes are extreme. Otherwise, the firm advertises the better attribute if the consumer is optimistic enough about the worse attribute, and advertises the worse attribute if the consumer is less optimistic about it.

As consumers become increasingly concerned about their privacy, firms can benefit from committing not to sell consumer data. However, the holdup problem prevents them from doing so in a static setting. The third essay studies whether the reputation consideration of the firm can serve as a commitment device in a long-run game when consumers have imperfect monitoring technology. We find that a patient enough monopoly can commit because its reputation will be permanently destroyed if consumers observe the data sale. The persistent punishment provides the monopoly a strong incentive not to deviate. In contrast, reputation may fail to serve as a commitment device when there are multiple firms. The penalty for selling data is smaller as consumers cannot know which exact firm sold the data. Also, other firms can hurt the reputation of a particular firm even if it does not sell data. We find some sufficient conditions under which the incentive to deviate is so strong that firms lose the commitment power. Reputation failure in the presence of multiple firms persists when we consider endogenous or asymmetric monitoring.

Firms make optimal decisions on the level of various marketing-mix variables to maximize profits. Some of these marketing-mix variables are unobserved by researchers, yet they are important to understand when they are endogenous in that a firm has incentives to increase or decrease them depending on how they change its profit. This thesis is focused on understanding a firm's incentive in two unobservable, endogenous decisions: workforce diversity and service effort. This thesis further builds theoretical frameworks around these two concepts, validates/estimates the model using public and proprietary data, and draws implications in market competition and policy experiment.

The literature in business, psychology, and sociology shows that there is a trade-off in hiring diverse workforce: diversity brings creative ideas that help solving problems, but it also creates friction among members. Under competition, the decision on the level of diversity of workforce is an important strategic decision that gives firms a competitive edge. We build a theoretical model with symmetric firms and find that the better a firm can deal with a diverse workforce or the more competition there is in the market, the more diverse workforce the firm will hire. We also find that a firm's profit decreases with the intensity of market competition as expected, but increases with the firm's inability to deal with diverse workforce as the inability deters competition between firms. Finally, we extend the model to firms with asymmetric private marginal costs and empirically validate the positive relationship between the diversity of workforce and the intensity of competition (industry concentration ratio, in particular) using two datasets from 1997: National Organizations Survey and Economics Census.

Service effort, often referred to as customer service, is also another important unobservable, endogenous decision that firms must make. Empirically quantifying or measuring the service effort is difficult because it is often unobservable. This paper proposes an empirical framework of the role of service effort in demand, along with other traditional marketing mix instruments. This model allows us to measure the unobserved effort level without data on effort, which is hardly available in most empirical settings. The paper also presents an application to a unique data set obtained from a franchise operating in the car radiator market. This framework can be useful in examining various aspects of service-intensive industries. In particular, this study investigates a much-debated public policy question regarding resale price ceiling in franchising. A policy evaluation shows that resale price ceiling lowers franchisees' profits and weakens their incentive to exert effort, which reduces consumer welfare. However, I find that, overall, resale price ceiling enhance consumer welfare in the car radiator market due to the lower price generated by the price ceiling.

This dissertation is composed of three essays focused on strategic considerations for product development. In chapter I, I address the question of whether consumers equally weigh capital and operating costs when purchasing durable goods. This trade off is important to manufacturers as they determine how much of their product design and production costs should be dedicated to keeping operating costs low. I test this empirically using data from the automobile industry. Chapter II is also an empirical study which explores whether consumers are willing to pay for socially responsible products. The answer to this question is important for firms to address in their product development process as they decide whether they will gain more market share by producing a socially responsible product with somewhat higher costs or a low cost product which does not incorporate socially responsible practices. In this case, I use data on retail sales in coffee industry using fair trade practices as an exemplar of social responsibility. Chapter III addresses the question of how durable or reliable manufacturers of durable goods should make their products. Consumers will likely want to pay more for a more durable product, yet increased durability depresses replacement demand. I attempt to gain insight into this trade off by developing an analytical model of the interplay between consumers and a monopolist manufacturer of durable goods. The remainder of the abstract provides a more detailed summary of each of these chapters in turn.

Chapter I explores whether consumers behave as if they are optimally trading off capital and operating costs when purchasing a durable good. I study this question using data on gas prices and automobile sales over the 20 year period, 1971-1990. This question is important for three reasons. First, it is interesting from a theoretical basis if consumers make this trade off optimally. Many theoretical models in marketing and economics make the fundamental assumption that consumers equally weigh current and future events when making decisions today. Yet there is some evidence, mostly from laboratory experiments, that consumers underweight future events. I attempt to explore this question in a market setting where the stakes are considerably higher. This research adds evidence to the debate about how much weight consumers place on future events when making choices today. Second, it is interesting to firms making product design decisions. If consumers underweight future events, then when making purchase decisions, consumers will view operating costs as less important than the upfront capital cost of the product. Finally, the answer to this question informs public policy. Many have argued that there is a need for the US to reduce gasoline demand per capita. Lower gasoline consumption would reduce environmental pressures, potentially dampen inflation, and allow more foreign policy flexibility in dealing with antagonistic regimes in oil-exporting states. While these rationale for reducing gasoline consumption have a normative flavor, and reasonable people may disagree as to the validity and motivations of this goal, it will nonetheless be useful to know the relative effectiveness of different policy levers in curbing gasoline consumption. For example, if consumers underweight fuel costs during the vehicle purchase decision, then a gas tax will be relatively less effective than a tax on gas guzzling vehicles.

To study this question I develop a choice model of the automobile industry. I identify the weight the consumer places on capital v. operating costs by determining how much of the variation in automobile market share can be explained by variation in each of these two factors holding other product attributes constant. We use data from the period 1971-1990, a period over which gas prices and thereby operating costs experienced considerable variation. In order to model operating costs which are not known at the time of purchase, I account for the expectations of consumers about car usage and gas prices. I assume that consumers are aware that they will respond to changes in the gasoline prices with changes in their driving patterns. Consumers also know that gasoline prices are not stable, and need to form expectations about future gasoline prices at the time of the automobile purchase. To take account of this effect, I estimate an ARIMA model of US gasoline prices from 1960-1995, that is used as the by consumers in their expectations formation process. Taking car usage and gas price expectations together enables an estimate of future gasoline costs of operating the car in the future. I also account for consumer heterogeneity in miles driven, sensitivity to automobile price, and sensitivity to operating costs. Finally, I recognize that prices are not exogenously determined and attempt to model prices as market outcomes.

Based on the results of the model developed, I find no evidence to support behavioral theories that consumers systematically underweight the cost of future events in real market settings. However, I find significant evidence that large portions of the population are not making the trade-off optimally. Some consumers underweight future operating costs (SUV drivers) while others appear to overweight them (hybrid drivers). Conservatively, at least 30% of the population is either drastically underweighting or overweighting operating costs when purchasing a new car.

Chapter II addresses the question of whether consumers are willing to pay for corporate social responsibility(CSR). This question is important in an environment where CSR is ubiquitous, yet it is unclear that these programs actually pay off for the firms that sponsor them. For example, consider Target's program to donate 1% of all retail sales to United Way local charities. Do consumers really want their money spent this way? Are consumers happily paying 1% higher prices or are they switching to a competitor which does not donate a portion of revenues to charity? Who is making the donation in the end, Target's shareholders or customers? From a social planner's perspective, the point is largely moot, yet to the shareholders of Target and many other firms practicing CSR, the question is crucially important.

I endeavor to study this question within the context of the coffee industry, an important and sizeable commodity market. In particular, I explore the impact of Fair-Trade (FT) certification on the retail coffee market. FT is a social and ethical movement that supports the ethical production of coffee and other products largely in third world countries. Coffee can be FT certified by adhering to FT standards. Once certified, FT coffee is distinguished from non-FT by distinctive labeling visible to the consumer who is deciding which coffee product to select from the supermarket shelf.

The analytical strategy for this paper is to first estimate the price premium commanded by FT coffee over non-FT coffees through ordinary least squares (OLS) and Fixed Effects hedonic price regressions. However, these tools do not allow us to disentangle the portion of the price premium which is due to supply considerations (i.e., FT certification costs) from the portion which is due to consumers' willingness to pay for FT coffee because they want to support socially responsible coffee production. To parse out willingness to pay from the overall price premium, I specify a brand choice model similar to the model used in chapter I.

Using hedonic price regressions I establish that FT coffee carries a price premium of $1.74 per 12-16 oz. It seems likely that at least a portion of this premium is due to increased consumer willingness to pay for FT coffee. However, I cannot rule out the possibility that the price premium is a result of the added costs associated with that FT practices. The choice model specified in this paper should enable the allocation of the cause of the price premium we have now established for FT coffee to demand v. supply considerations. I hope to estimate this model in future research.

Chapter III addresses the problem of how durable or long lasting manufacturers should they make their products. On the one hand a more durable product will be more desirable to consumers, since it will provide benefits over a longer period. Thus, a longer-lived product will command a higher price. However, it seems likely that unit production costs will increase as a product is made more durable due to the increased cost of more reliable materials and more exacting quality standards. In addition, a product which is more durable will be replaced less frequently. Ceteris paribus, less frequent replacement is less desirable to manufacturers, as the periodicity of the revenue stream increases. Manufacturers can trade off the benefits of durability with the costs to determine the optimal reliability or life for the goods they produce.

In some sense, this problem is a classic trade-off between quality and cost. What distinguishes the durable goods reliability problem is that increasing quality depresses replacement demand. A common anecdote is that light bulbs could easily be manufactured to last longer, but are not in order to increase replacement sales.

This question is important for manufacturers to understand from several perspectives. First, a manufacturer of a product with technology that is fairly static (e.g., light bulbs), needs to consider replacement demand in developing product designs. When technology is not static (e.g., computers), it is important to understand how the rate of technology advance will stimulate replacement demand. Should the products be designed to be more or less durable in the face of technological advance? An additional complication arises when the rate of technological advance may only be partially observable to the consumer (e.g. golf clubs). Finally, manufacturers need to consider "buying back" used durable goods from their customers in order to stimulate replacement demand. Even if the manufacturer does not take physical possession of the old product, it may be optimal to offer price concessions on newer products in order to induce the consumer to replace her existing technology.

I study this question by developing a partial equilibrium model for durable goods where both manufacturers and consumers are forward looking over an infinite horizon. I make the simplifying assumptions that there is a single monopolist manufacturer and that consumers have homogeneous preferences.

I find that in all cases, it is optimal for manufacturers to incent consumers (buy back) to replace their existing durable goods before the end of the useful life of the product. I also find that under extremely rapid rates of technological advance, it is optimal for manufacturers to extend the life of the product and increase durability. This result runs somewhat counter to intuition in that one might think that rapid advances in technology would promote more frequent product replacement. However, if technology advances rapidly then patient consumers will want to be able to enjoy their new product (e.g., HDTV) and are willing to pay a high price for durability. This increased willingness to pay for an advanced long-lived product will more than offset the loss of replacement sales in the future. Finally, I show that when technology advances are uncertain and not directly observable by consumers, they are willing to pay more than when technology advances at a known and constant rate. The intuition for this result is that consumers are risk averse and do not want to miss out on a new breakthrough product. We believe this may be the reason that certain durable goods manufacturers (e.g. golf club manufacturers) appear to introduce new products at a far greater rate than technology is actually advancing.

Econometric models based on observational data are often endogenous due to measurement error, autocorrelated errors, simultaneity and omitted variables, non-random sampling, self-selection, etc. Parameter estimates of these models without corrective measures may be inconsistent. The potential high-dimensional feature of these models (where the dimension of the parameters of interests is comparable to or even larger than the sample size) further complicates the statistical estimation and inference. My dissertation studies two different types of high-dimensional endogenous econometrics problems in depth and develops statistical tools together with their theoretical guarantees.

The first essay in this dissertation explores the validity of the two-stage regularized

least squares estimation procedure for sparse linear models in high-dimensional

settings with possibly many endogenous regressors. The second essay is focused on the semiparametric sample selection model in high-dimensional settings under a weak nonparametric restriction on the form of the selection correction, for which a multi-stage

projection-based regularized procedure is proposed. The number of regressors in the

main equation, p, and the number of regressors in the first-stage equation, d, can grow

with and exceed the sample size n in the respective models. The analysis considers the

sparsity case where the number of non-zero components in the vectors of coefficients

is bounded above by some integer which is allowed to grow with n but slowly compared

to n, or the vectors of coefficients can be approximated by exactly sparse vectors.

Simulations are conducted to gain insight on the small-sample performance of these

high-dimensional multi-stage estimators. The proposed estimators in the second essay

are also applied to study the pricing decisions of the gasoline retailers in the Greater

Saint Louis area.

The main theoretical results of both essays are finite-sample bounds from which

sufficient scaling conditions on the sample size for estimation consistency and variable selection consistency (i.e., the multi-stage high-dimensional estimation procedures correctly select the non-zero coefficients in the main equation with high probability) are established. A technical issue regarding the so-called “restricted eigenvalue (RE) condition” for estimation consistency and the “mutual incoherence (MI) condition” for selection consistency arises in these multi-stage estimation procedures from allowing the number of regressors in the main equation to exceed n and this paper provides analysis to verify these RE and MI conditions. In particular, for the semiparametric sample selection model, these verifications also provide a finite-sample guarantee of the population identification condition required by the semiparametric sample selection models.

In the second essay, statistical efficiency of the proposed estimators is studied via

lower bounds on minimax risks and the result shows that, for a family of models with exactly sparse structure on the coefficient vector in the main equation, one of the proposed estimators attains the smallest estimation error up to the (n, d, p)−scaling among a class of procedures in worst-case scenarios. Inference procedures for the coefficients of the main equation, one based on a pivotal Dantzig selector to construct non-asymptotic confidence sets and one based on a post-selection strategy (when perfect or near-perfect selection of the high-dimensional coefficients is achieved), are discussed. Other theoretical contributions of this essay include establishing the non-asymptotic counterpart of the familiar asymptotic “oracle” type of results from previous literature: the estimator of the coefficients in the main equation behaves as if the unknown nonparametric component were known, provided the nonparametric component is sufficiently smooth.

In three essays, I present my work that uses mathematical modelling to analyse real-world problems in marketing. In the first chapter, I model the design of referral programmes and offer normative advice about improving the cost-effectiveness of a referral programme. The second chapter is joint work with my co-advisor, Yuichiro Kamada. In this work, we construct a tractable game-theoretic model for multiple-priority queue management. Lastly, the third chapter analyses the incentives of online bloggers that post product reviews to subscribers when the reviews may be sponsored.

Customer referrals have become an increasingly important way for firms to grow their customer base. Likely thanks to improvements to customer relationship management software as well as growing awareness of the potential of word-of-mouth marketing, referral programmes are becoming increasingly popular among firms. In the first chapter, I study the design of customer referral programmes by constructing a stylized static principal-agent model with hidden actions, in which a firm designs a referral programme to incentivize an existing customer to exert costly efforts to refer the customer's friends to the firm. In the baseline model, I find it optimal for the firm to pay the customer if and only if every friend of the customer is successfully referred. In a number of extensions that are important and relevant to referral programmes, although the optimal referral contract is no longer a threshold contract, this class of contracts still plays an important role in the optimal referral programme design. Overall, my work shows that it is cost-effective to use or include threshold contracts to incentivize efforts.

When access to a service facility is congested, service providers commonly implement a special type of queue called priority queue, where each person/entity in the queue has an associated priority such that those with a higher priority will be ahead of those with a lower priority in the queue. Previous studies into second-degree price discrimination and queue management suggest that the firm that manages the firm that manages the park should set a large number of different priorities in order to improve price discrimination. In the second chapter, my co-author and I construct a model in which an amusement park sells different priority passes to customers in a queue whose utilities depend on positions in the queue. A customer's valuation of a priority pass depends on the distribution of customers buying each priority pass, and hence other customers' purchase decisions have an externality on the customer's valuation, which differentiates our model from the standard screening models. Through the model, we discuss the implementability of selling multiple passes for different patterns of customer utility functions. The main result of our work is that the externality makes the implementation of multi-pass schemes difficult, an issue that persists even when customers have heterogeneous utilities of positions in a queue.

Thanks to the popularity of internet platforms such as YouTube and TikTok, more and more customers are turning to their favourite internet bloggers for product reviews. That the bloggers have many subscribers that rely on their reviews for purchase decision makes bloggers a potential sales channel. This type of marketing is often called influencer marketing. To incentivize the bloggers to help promote a firm's product, one common incentive is for the firm to sponsor a blogger's product review by offering sales commissions for purchases by the blogger's subscribers. In a sponsored review, since the blogger now benefits from higher sales, the blogger has a bigger incentive to review the product favourably, even though the blogger's own private signal about the product's quality says otherwise. On the other hand, the blogger cares about the accuracy of the review because accurate reviews attract viewership. In the third chapter, I construct a stylized model that analyses the incentives of a blogger posting reviews of products with uncertain quality on an internet platform to the blogger's subscribers. Specifically, the blogger receives a costless private signal about the quality of the product and then sends a review message to the subscribers. On the one hand, the platform rewards the blogger for posting accurate reviews. On the other hand, higher sales of the product lead to higher sales commission offered by the firm manufacturing the product. I find that the blogger has an incentive to truthfully communicate the received signal when the signal is informative; otherwise, the blogger has an incentive to pretend to receive a favourable signal even when the realized signal is unfavourable to the product. Under a symmetric signal structure, my model shows that the blogger has an incentive for an honest review when the product is perceived to be mid-end.

Credit cards are an important vehicle for spending and borrowing. Collaborating with a leading commercial bank in China, this dissertation consists of two chapters that explore how the design of credit card products can influence consumer behavior, alongside its implications for marketing strategies and consumer welfare.

The first chapter studies reward programs, often a prominent feature of credit cards. I combine proprietary consumer-level data and a survey to study the causal effect of rewards on consumption and consumers' subjective expectations. I leverage a fuzzy regression discontinuity (RD) design to show that a more generous reward design causes consumption increases across both reward-earning and non-reward-earning categories. Applying the fuzzy RD to the survey data, I find that consumers correctly understand the impact of reward design on reward-earning consumption but underestimate its effect on total consumption. Using a stylized model, I study the implications of this misperception for market structure and welfare. My calibration results show that consumer misperceptions incentivize banks to offer more generous rewards, which ultimately diminishes market efficiency and leads to a cross-subsidy from less to more sophisticated consumers.

The second chapter, coauthored with Xiao Yin, investigates the extent to which consumers misperceive the interest costs associated with credit card debt using a combination of administrative data and surveys. Through a randomized controlled trial with an information treatment, our results show that consumers are imperfectly informed about the interest cost of unsecured debt, and the resulting misperception induces excess debt-taking by 26\%, mainly originating from spending on luxury goods. To understand the formation of interest rate misperception, we uncover selective information acquisition to be a potential channel. We demonstrate that consumers tend to actively seek information about their borrowing status when they expect creditworthiness to be high, and they disproportionately pay more attention to favorable information when interest rates are low.