Book Description

Consumer preferences are changing over time. In this dissertation, we provide three studies regarding changes in consumer preferences and methods of modeling time-varying preferences. In Chapter 1, we propose a Simulated Maximum Likelihood estimation method for the random coefficient logit model using aggregate data, accounting for heterogeneity and endogeneity. Our method allows for two sources of randomness in observed market shares - unobserved product characteristics and sampling error. Because of the latter, our method is suitable when sample sizes underlying the shares are finite. We show that the proposed method provides unbiased and efficient estimates of demand parameters. We also obtain endogeneity test statistics as a byproduct, including the direction of endogeneity bias. The model can be extended to incorporate Markov regime-switching dynamics in parameters and is open to other extensions based on Maximum Likelihood. The benefits of the proposed approach are achieved by assuming normality of the unobserved demand attributes, an assumption that imposes constraints on the types of pricing behaviors that are accommodated. However, we find in simulations that demand estimates are fairly robust to violations of these assumptions. We propose a structural model of market evolution and apply the proposed model to the South Korean cigarette market data in Chapter 2. In the South Korean cigarette market, consumers have shown dramatic changes in their cigarette preferences. While most consumers smoked high-tar cigarettes ten years ago, now most consumers prefer low-tar cigarettes. Another interesting trend in this market is Given the strong dynamics in the growing popularity of super-slim cigarettes. preferences, we raise two critical questions - 1) what are the sources of preference change, and 2) how does the firm (KT&G Corporation, a de-facto monopolist in the market) react to these preference changes. We answer these questions using a unique In the proposed demand structural model of consumer demand and firm behavior. model, evolution of consumers' preferences is driven by an exogenous effect and a new product introduction effect. On the one hand, the increasing preference for low- tar cigarettes can be explained by consumers' growing heath consciousness, an exogenous effect. Due to stringent government restrictions on promotion and advertising of tobacco products, new product introduction is an important marketing instrument for KT&G. We hypothesize that a new product carries critical This is the information that subsequently influences consumer preferences. introduction effect. We propose an aggregate random coefficient logit model wherein the parameters evolve as a function of the introduction and exogenous effects. This model allows us to separate the two effects and examine their relative significance. Another key research question we study is how the firm reacts to the To answer this question, we build two supply side models. preference changes. First, we specify the firm's pricing model which elucidates the influence of the timevarying preferences on the firm's pricing decisions. Second, we model the firm's decisions regarding new product design and introduction. This model clarifies the firm's decision process regarding the new product under the time-varying consumer preferences. This study provides valuable insights into the sources of preference Also, it changes, and how firms' decisions shape the fundamentals of the market. sheds light on the role and the value of new products design and introduction. proposed model can help a firm develop a new product strategy that will move consumer preferences in a preferred direction. In many categories consumers display cyclical buying: they repeatedly The purchase in the category for several periods, followed by several periods of not buying. One possible explanation for such cyclicality is the joint effect of habit and boredom on repeated purchasing. In Chapter 3, we propose a Markov regime-switching random coefficient logit model to represent these behaviors as stochastic switching between high and low category purchase tendencies. The main feature of the proposed model is that it divides the stream of purchase decisions of a consumer into distinct regimes with different parameter values that characterize high versus low purchase tendencies. In an empirical application of the model to purchases of yogurt-buying households we find that as many as 40.8% display cyclicality between high and low yogurt purchasing tendencies. We show (via simulation) that alternating between high and low purchase tendencies corresponds with changing levels of consumer inventory in a substitute category. If one ignores this phenomenon, a correlation between yogurt inventory and the unexplained part (or error term) in utility arises leading to biased estimates. Predictions from the proposed model track observed yogurt purchases of households over time closely, and the model also fits better than three benchmark models. Also, we show that cyclicality in buying has a key implication for a firm's price promotion strategies: a price reduction that is offered to a household during its high purchasing tendency period will result in greater increases in sales than one that is offered during its low purchasing period. This opens up a new dimension for enhancing the effectiveness of promotions - customized timing of price reductions.