Essays in behavioral industrial organization, corruption, and marketing

Abstract

In Chapter 1, I propose a model in which consumers base their purchasing decisions upon their recollections of the product quality, and in which firms can use persuasive advertising in order to change these recollections. Although consumers are aware that such advertising has occurred and take this into account when updating their beliefs about the product, they cannot prevent their memories from being affected. I analyze which firms engage in persuasive advertising as well as the price level that these firms choose. I show that persuasive advertising may be used in equilibrium even though consumers are fully aware of it, and that persuasive advertising does not always signal high quality products. The model is then extended to incorporate both persuasive and informative advertising, where firms reveal some verifiable information about their products. In that case, persuasive advertising may block the full unraveling of information, and high quality products are not promoted with only one type of advertising - in some cases, persuasive advertising can signal a product of either higher or lower quality than a product promoted with informative advertising. Chapter 2 is the product of joint work with Abhijit Banerjee and develops a model to study the effectiveness of complaints against corruption. A bureaucrat has to decide on a public infrastructure project in a village where a rich and a poor villagers live. A dishonest bureaucrat can be bribed not to choose the surplus maximizing project and instead to choose a project that favors the rich villager. Once the bureaucrat has chosen a project, the villagers can send a costly praising or complaining message to the bureaucrat's supervisor who does not know whether the bureaucrat is honest or dishonest.

(cont.) From his point of view the messages are anonymous; the supervisor does not know who is rich or poor in the village. The only leverage of the supervisor is to transfer the bureaucrat and replace him with another one who will repeat the game in the following period. In any relevant equilibrium no complaints happen and more generally there are no complaints in equilibrium without bribery. We find that complaints will be observed only when they should not be and that the government cannot necessarily get people to complain by cutting the message cost. In addition, lowering that cost may hurt since, when the share of honest bureaucrat is low, the poor are pessimistic about the benefit of complaints while the rich are optimistic and they respond more to a lower cost. Finally, the supervisor cannot fully decide to implement a particular equilibrium as multiple ones coexist. Chapter 3 is the product of joint work with Elie Ofek. We model a duopoly in which ex-ante identical firms need to decide where to direct their innovation efforts. The firms face market uncertainty with respect to consumers' preferences for innovation on two product attributes, and technology uncertainty with respect to the success of their R&D efforts. Firms can conduct costly research to resolve their market uncertainty before setting R&D strategy. We find that the value of market information to a firm depends on whether its rival is also expected to obtain this information in equilibrium. We show that, as a result, one firm may forgo market research even though its rival conducts such research and learns the true state of demand. We examine both vertical and horizontal demand structures. With vertical preferences, firms are a priori uncertain which attribute all consumers will value more.

(cont.) In this case, a firm that conducts market research will always innovate on the attribute it discovers that consumers prefer, and expend more on R&D than a rival that has not conducted market research. With horizontal preferences, distinct segments exist-each cares about innovation on only one attribute-and firms are a priori uncertain how many consumers are in each segment. In this case, a firm that conducts market research may follow a 'niche' strategy and innovate to serve the smaller segment to avoid intense price competition for the larger segment. Consequently, a firm that conducts market research may invest less in R&D and earn lower profits post-launch than a rival that has forgone such research.