Shocked by What People Write Online? How Uncovering Negative Reviews Impacts Online Search and Purchase

We investigate the impact of negative reviews on consumer online search and purchase decisions. More specifically, using a large data set from an online retailer that tracks when consumers scroll down and paginate over product reviews, we quantify the effect of a 1-star review on information search about the focal product and other products in the category, and on eventual purchase. We show that when consumers scroll down to a negative review, it significantly increases browsing for other products, leading to a larger consideration set. Based on this result, we quantify two effects of the negative review on choice: the direct effect on purchase because of negative information about the product; and an indirect effect because of an increase in the choice set. The magnitude of these effects depends on (i) how unexpected the negative rating was, (ii) when in the search sequence the negative review was discovered, (iii) the type of product category, and (iv) whether search occurs on weekend. Preliminary results show that although switching between products does happen, for the retailer the net result is either insignificant or positive. This is so because consumers start their search by looking at cheaper products and a negative review may lead them to consider more expensive alternatives.


Ordered consumer search

The paper discusses situations in which consumers search through their options in a deliberate order. Topics include: the existence of ordered search equilibria with symmetric sellers (all consumers first inspect the seller they anticipate will set the lowest price, and a seller which is inspected first by consumers will set the lowest price); the use of price and non-price advertising to direct search; how purchase history can guide future search; and the incentive a seller can have to raise its own search cost. I also show how ordered search can be reformulated as a simpler discrete choice problem without search frictions.




Vertical Information Restraints: Pro- and Anti-Competitive Impacts of Minimum Advertised Price Restrictions

We consider vertical contracts where the retail market may involve search frictions. Minimum advertised price restrictions (MAP) act as a restraint on customers' information and so can increase search frictions in the retail sector. Such restraints, thereby, soften retail competition|an impact also generated by resale price maintenance (RPM). However, by accommodating (consumer or retailer) heterogeneity, MAP can allow for higher manufacturer profi ts than RPM. We show that they can do so through facilitating price discrimination among consumers; encouraging service provision; and facilitating manufacturer collusion. Thus, welfare effects may be positive or negative compared to RPM or to the absence of such restrictions.


Michael (Yu-Fai) CHOI

Consumer Search and Price Competition with Anovia Yifan Dai and Kyungmin Kim


We consider an oligopoly model in which consumers engage in sequential search based on partial product information and advertised prices. We derive a simple condition that fully summarizes consumers’ shopping outcomes and use the condition to reformulate the pricing game among the sellers as a familiar discrete-choice problem. Exploiting the reformulation, we provide sufficient conditions that guarantee the existence and uniqueness of pure-strategy market equilibrium and obtain several novel insights about the effects of search frictions on market prices. Among others, we show that a reduction in search costs increases market prices, but providing more pre-search information raises market prices if and only if there are sufficiently many sellers.


Tobit GAMP

Deceptive Products and Competition in Search Market

We study a search market where firms have a choice between offering either efficient, high quality (“candid”) products or inefficient, low quality (“deceptive”) products which some (“naive”) consumers fail to recognize as such. We derive an equilibrium in which both business models co-exist and show that as search frictions vanish, high quality goods are entirely driven out of the market. We show that market share and price dynamics can be non–monotone in search frictions, and we argue that while policy interventions that reduce search frictions such as the standardization of price and package formats may harm welfare, a price floor regulation and a minimum quality standard can improve welfare.



Dynamic Pricing with Search Frictions

This paper studies dynamic pricing in markets with search frictions. Sellers have a single unit of a good and post prices in every trading period. Buyers have to incur a search cost to match with a new seller and upon matching they observe the price and the realization of some idiosyncratic match value. There is no discounting but trade ends at an exogenously given deadline. We show that equilibrium involves trading in finitely many trading periods and the volume of trade increases over time. Under mild conditions on the buyer-to-seller ratio and the distribution of valuations, prices decrease at increasing rates as the deadline approaches. We derive the gains from trade in equilibrium and their distribution between buyers and sellers. For the case in which the measures of buyers and sellers coincide, we provide a full characterization of the (unique) equilibrium for a class of distribution functions. We fi nally discuss implications for market design, including the use of platform fees and cancellation policies.


Elisabeth HONKA

The Effects of Advertising on Awareness, Consideration, and Choice: Evidence from the U.S. Auto Insurance Industry

How does advertising influence consumers' decision-making? Does advertising inform or persuade consumers? We investigate the relationship between both advertising content and quantity and each stage of the purchase funnel, namely, awareness, consideration, and choice. Understanding how the amount and content of advertisements affects consumers' decision-making is crucial for companies in effectively and efficiently using their advertising budgets. Our unique individual-level data contain information on consumers' awareness and consideration sets, purchase decisions, demographic variables, and perceived prices for consumers between 2008 and 2015. We supplement these data with data on advertising quantities for all and advertising content for four media channels (newspapers, magazines, TV, and internet). We account for the endogeneity of the advertising decision using the regression discontinuity approach suggested by Shapiro (forthcoming). Our results reveal that advertising quantity affects consumer awareness, but not consideration or choice.


No Shopping in the U.S. Mortgage Market: Direct and Strategic Effects of Providing Information

We document and analyze price dispersion in the U.S. mortgage market. We find significant price dispersion in posted prices in the retail channel: for example, a consumer with a prime credit score and with a 20% down payment might see a spread in interest rates of 50 basis points, controlling for all relevant consumer/property characteristics, including discount points. We also show, from survey evidence, that close to half of consumers did not shop before taking out a mortgage, and worse, many consumers do not seem to realize that there is price dispersion. Using a proprietary dataset of lenders' ratesheets, we estimate an equilibrium model of costly search where a share of consumers holds incorrect beliefs regarding price dispersion. Whereas high search costs is one reason behind the lack of search, we show that non-price preferences also play an important role in preventing consumers from searching more; and so an effective policy would target both. In one of our counterfactuals, we show that eliminating non-price preferences results in savings of about $9 billion dollars a year. 



Informational Cycles in Search Markets

A cyclical equilibrium exists in a search market where buyers learn about the unknown state of the market from other buyers' actions. Each buyer observes whether a randomly chosen buyer traded in the previous period. In the cyclical equilibrium, the informational content of observing a trade oscillates: a trade is good news about the state in odd periods and bad news in even periods. The volume and probability of trading fluctuate more if the unknown state is bad rather than good. A steady state equilibrium where buyers are more likely to continue searching than in the cyclical equilibrium is less efficient than the cyclical equilibrium. The cyclical equilibrium is reached from a nondegenerate set of parameter values in a market that starts off at date one.


Multiproduct Intermediaries

This paper offers a framework for studying the optimal product range choice of a multiproduct intermediary, in an environment where consumers demand multiple products and face search frictions. We first demonstrate that the intermediary earns positive profit even if it is no more efficient than small firms at selling products. We then characterize its optimal stocking policy. The intermediary uses exclusively stocked high-value products as loss leaders to increase store traffic, and at the same time earns profit from non-exclusively stocked products which are relatively cheap to buy from manufacturers. We also show that relative to the social optimum, the intermediary tends to be too big and stock too many products exclusively. (Joint with Makoto Watanabe and Jidong Zhou)


Jurre H. THIEL

The Remuneration of Advisors in Markets for Complex Products

I develop a search-theoretic model to explain the entry of advisors in retail financial markets. When advisors are remunerated through kickbacks, more advisors enter when firms have a larger incentive to obfuscate (increase search costs). Thus, the intuitive link that advisors are more prevalent in more “complex” markets arises endogenously. Even though advisors facilitate consumer search, the effect of advisor entry on consumer welfare is ambiguous. However, as in the long-run free entry equilibrium there will generally be fewer advisors when kickbacks are banned than when they are not, banning such kickbacks may ultimately harm consumers. I show that the effectiveness of kickbacks in steering advisors’ recommendations and the value of additional services advisors offer are key parameters to understand whether banning kickbacks is beneficial to consumers.


Chengsi WANG

Search platforms: Showrooming and price parity clauses

We provide a model in which consumers search for fi rms directly or through platforms. Platforms lower search costs but charge firms for the transactions they facilitate. Platform fees raise the possibility of showrooming, in which consumers search on a platform but then switch and buy directly to take advantage of lower direct prices. In settings like this, search platforms like Amazon's marketplace and have adopted price parity clauses, requiring fi rms offer their best prices on the platform, arguing this is needed to prevent showrooming. We use our model to evaluate the implications of showrooming and price parity clauses.