No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
An Empirical Model of Heterogeneous Consumer Search for Retail Prescription Drugs
Abstract
This paper uses detailed data on retail pharmacy transactions to make inferences about the nature and intensity of consumer search for prescription drugs. Prescription prices exhibit patterns that should, in principle, induce search: in particular, prices vary widely across stores, and stores' price rankings are inconsistent across drugs (so the low-price pharmacy is different for one prescription vs. another). Estimates from a model of pharmacy choice suggest that search intensities are generally low: I estimate that for a typical prescription, the fraction of consumers that price-shops is approximately 5-10 percent. However, variation in this estimated search intensity across drugs is substantial and appears to be consistent with explanations based on rational search; for instance, price-shopping is more prevalent for maintenance medications than for one-time purchases, presumably because the benefits of finding a low price are magnified for prescriptions that are purchased repeatedly. Under some relatively strong assumptions imposed by the empirical model, the data also identify parameters of a search cost distribution, suggesting that the cost of conducting an exhaustive price search is approximately $15 for the average consumer.
... 9 Such an explanation conceives of patients first 'shopping' at the public pharmacy and, if they cannot obtain the drugs they need there, they resort to the private pharmacy as a second choice. 15 This sequence of behaviour makes sense when: (1) we believe patients are price-sensitive; and (2) drugs are free or highly subsidised in the public sector versus full price in the private sector, 16 which is the case in our study setting. From such a perspective, private retail drug stocks serve as a backstop, backfilling the public drug supply. ...
... This indicates that at least some patients in hospitals do not directly bypass the public pharmacy at the hospital but rather 'shop' at the public pharmacy before resorting to the private sector. 16 We do not find evidence of private drug purchases responding to drug stocks in primary facilities. These differences are consistent with the patient perceptions of higher quality of care in hospitals versus primary facilities extending to drug stocks, 22 but there are other potential explanations discussed below. ...
Introduction
In India, public sector patients purchase drugs from private pharmacies instead of obtaining them for free from public pharmacies—a phenomenon we call public patient forwarding to private pharmacies. This behaviour results in substantial financial hardship. We examine whether low public drug stocks, patient preferences for private drugs or the presence of private pharmacies nearby explain this behaviour.
Methods
We collected cross-sectional data from 7567 households, 523 health facilities and 1036 private pharmacies in Odisha, India. We linked 917 outpatient visits to facilities based on patient reports and linked public facilities to the nearest private pharmacy using Global Positioning System coordinates. We used ordinary least squares regression to assess whether the behaviour of facilities and patients was associated with drug stocks and pharmacy proximity, and whether patient satisfaction was associated with private drug purchases.
Results
Among public patients prescribed drugs, more than 70% purchased private drugs. In hospitals, for each 10% increase in drug stocks, 4.8% fewer patients purchased private drugs (p=0.047). In primary facilities, the same share of patients purchased private drugs across stock levels. Regardless of facility level, when more than 75% of drugs were in stock, 60% or more of patients still obtained drugs from the private sector. Patients were more likely to purchase private drugs when private pharmacies were near public facilities, but were not more satisfied with their visit when they obtained private drugs.
Conclusion
The results suggest that private pharmacies are both secondary and complementary suppliers of drugs for hospitals, but may act more like substitutes for primary facilities, consistent with evidence that private pharmacies provide advice and other services akin to primary care in Odisha. Improving public facility drug stocks alone is unlikely to fully address drug-driven financial hardship in India. Provider prescribing practices should be investigated to identify additional policy options.
... More generally, the challenge in predicting consumer choice with search cost is to simultaneously identify consumers' heterogeneous preferences and search costs. As pointed out by Sorensen (2001) and Hortacsu and Syverson (2004), explaining search decisions by consumers with heterogeneous preferences imposes an identification problem. A consumer may stop searching either because of a high valuation for the products already found or because of a high search cost. ...
... One of the major challenges in this analytical study is how to simultaneously identify consumers' heterogeneous preferences and search cost. As pointed out by Sorensen (2001) and Hortacsu and Syverson (2004), explaining search decisions by consumers with heterogeneous preferences imposes an identification problem. A person may stop searching either because she has a high valuation for the products already found or because she has a high search cost. ...
The overload of social media content today can lead to significant latency in the delivery of results
displayed to users on product search engines. We propose a dynamic structural model whose output can facilitate digital content analytics by search engines by helping predict consumers' online search paths. Such predictive prowess can facilitate web caching of the "most likely-to-be-visited" web pages and reduce latency. Our model combines an optimal stopping framework with an individual-level random utility choice model. It allows us to jointly estimate consumers' heterogeneous preferences and search costs in the context of product search engines, and predict a probability-based search path for each consumer. We estimate the parameters of the model using a dataset of approximately 1 million online search sessions resulting in room bookings in 2117 U.S. hotels. We find that search engine ranking can polarize search costs incurred by users. A good ranking saves consumers, on average, 18.54. Our model prediction results demonstrate that the proposed dynamic structural model provides the best overall performance in predicting the probabilities of consumers' online search paths compared to several baseline models that do not include a formal structure or dynamics in them.
... Another challenge in estimating product demand with search cost is how to simultaneously identify consumers' heterogeneous preferences and search cost. As pointed out by Sorensen (2001) and Hortacsu and Syverson (2004), explaining search decisions by consumers with heterogeneous preferences imposes an identification problem: A consumer may stop searching either because of a high valuation for the products already found or because of a high search cost. The same observed search outcome can be explained either by the preferences for product characteristics or by the moments of the search cost distribution (Koulayev 2010). ...
... One of the major challenges in the dynamic search demand estimation is how to simultaneously identify consumers' heterogeneous preferences and search cost. As pointed out by Sorensen (2001) and Hortacsu and Syverson (2004), explaining search decisions by consumers with heterogeneous preferences imposes an identification problem. A person may stop searching either because she has a high valuation for the products already found or because she has a high search cost. ...
With the proliferation of social media, consumers’ cognitive costs during information-seeking can become non-trivial during an online shopping session. We propose a dynamic structural model of limited consumer search thatcombines an optimal stopping framework with an individual-level choice model. We estimate the parameters of the model using a dataset of approximately 1 million online search sessions resulting in bookings in 2117 U.S. hotels. The model allows us to estimate the monetary value of the the search costs incurred by users of product search engines in a social media context. On average, searching an extra page on a search engine costs consumers 6.24, respectively. A good recommendation saves consumers, on average, 18.54. Our policy experiment strongly supports this finding by showing that the quality of ranking can have significant impact on consumers’ search efforts, and customized ranking recommendations tend to polarize the distribution of consumer search intensity. Our model-fit comparison demonstrates that the dynamic search model provides the highest overall predictive power compared to the baseline static models. Our dynamic model indicates that consumers have lower price sensitivity than a static model would have predicted, implying that consumers pay a lot of attention to non-price factors during an online hotel search.
... We acknowledge that inputting the value of time is rather arbitrary because we do not know if people value time away from work differently relative to the value given by their employers. For this reason, we include as controls, socio-demographic characteristics that have been found correlated with search costs such as age, education level, and gender, and we further control for purchasing habits (De los Santos, (2008); Sorensen, (2001)). This does not allow us to quantify the differences in the value of time, but it does allow us to control for it. ...
... The results on the control variables are presented inTable 6, in Appendix I. Most control variables are not significant. Consistent with Sorensen (2001), as purchasing frequency decreases, consumers are less likely to be willing to search. Respondents who purchase gasoline once a week are less likely to search than the reference category (i.e. ...
We use a choice experiment on gasoline consumers to investigate whether respondents exhibit limited attention to the way different costs enter their search decision. The search cost is a function of the amount of gasoline consumed while driving and the time spent searching for the lowest price. The gasoline used to drive is a disbursement the consumer has made in the past, whereas the time spent searching is a cost that is incurred at the time of search. We randomize the amount of information we provide respondents about search costs in one of 3 ways: (1) time, (2) gasoline spent driving or (3) both. The results indicate that consumers exhibit inattention, leading them to overestimate the cost of the gasoline used while driving, which in turn results in consumers not searching when the expected gains from search exceed the costs, forgoing consumer surplus. The differences in the magnitude of the effect of inattention to time costs, however, can be attributed to cognitive costs of computing the value of the amount of time it takes to drive one mile; it is stronger among consumers that were not reminded of either search cost.
... There are two features that must be present in a market for search to be profitable: there must be price dispersion, or else the opportunities to find a different (lower) price would be diminished, and consumers must be unable to perfectly classify retailers as high-or low-priced (Sorensen, (2001)). In the gasoline retail market, price dispersion can be partially attributed to the unique characteristics of the industry which contribute to product differentiation 2 , and partially to the lack of consumer search (Tappata, (2006Tappata, ( , 2009 Lewis, (2008); Hastings, (2004); Shepard, (1993)). ...
... Results on control variables are presented in Appendix I. Most of the control variables do not significantly influence willingness to search (income, age, education). Consistent with Sorensen (2001), as purchasing frequency decreases, consumers are less likely to be willing to search. The reference category corresponds to respondents who purchase gasoline twice a week or more. ...
We use an internet survey conducted among a random sample of 490 drivers in the State of Ohio to answer the question, “When are consumers more likely to search?” The internet survey affords us the opportunity to overcome endogeneity difficulties with market observation data by imposing exogenous price changes in a random sample of gasoline consumers to examine the decision-making process behind intended search decisions. Results indicate that among the respondents who faced prices below their expected price, only 12% chose to search, whereas 45% searched when prices were above. Results suggest that asymmetric search can be explained by prospect theory, in the sense that consumers evaluate current prices compared to a reference price, and as a consequence they value price increases differently from price decreases. Our findings indicate that in the gasoline retail market, consumers are allowing retailers to extract consumer surplus by exhibiting loss aversion because this behavior deters search when the probability of finding a lower price is highest.
... For a given card, there was also no prior commitment for prices to be the same across all pharmacies that offered discounts under the card. Second, in addition to the usual consumer search and switching costs that contribute to price dispersion in drug retail markets (see, e.g., Sorensen (2000 Sorensen ( , 2001), Scott-Morton (1997)), prohibitive consumer switching costs were erected by the very design of the program. 3 Once enrolled in a card program, a consumer was not allowed to switch to another card, except in certain special cases, such as when a consumer moves to a new location or if a card sponsor exited from the market. ...
... Consumer search is an important source of price dispersion in retail drug markets (e.g. Sorensen (2000 Sorensen ( , 2001). Static models of search are abundant in the literature (see, e.g., Salop and Stiglitz (1977), Reinganum (1979), Burdett and Judd (1983), Stahl (1989)). ...
In early 2004, the U.S. government initiated the Medicare Drug Discount Card Program (MDDCP), which allowed card subscribers to obtain discounts on prescription drugs. Pharmacy-level prices were posted on the program website weekly with the hope or promoting competition among card sponsors by facilitating consumer access to prices. A large panel of pharmacy-level price data collected from this website indicates that price dispersion across cards persisted throughout the program. Prices declined initially when consumers were choosing cards, but rose later when subscribers were restricted to commit to their card choices. In contrast, contemporaneous prices from online drug retailers, which were unrelated to the program, rose steadily over time, indicating that program prices evolved in a way different from the general evolution of prices outside the program.
... Thus, prices could potentially have evolved di¤erently in di¤erent pharmacies and locations. Second, in addition to the usual consumer search and switching costs that contribute to price dispersion in drug retail markets (see, e.g., Sorensen (2000 Sorensen ( , 2001 ), Morton (1997)), prohibitive consumer switching costs were erected by the very design of the program. 3 Once enrolled in a card program, a consumer was not allowed to switch to another card, except in certain special cases, such as when a consumer moves to a new location or if a card sponsor exited from the market. ...
... Consumer search is an important source of price dispersion in retail drug markets (e.g. Sorensen (2000 Sorensen ( , 2001). Our focus is on the implications of models of consumer search and better access to price information on …rms'pricing behavior, subject to the institutional aspects of the program. ...
In early 2004, the U.S. Government initiated the Medicare Discount Drug Card Program (MDDCP), which created a market for drug cards that allowed elderly and handicapped subscribers to obtain discounts on their prescription drug purchases. Pharmacy-level prices for many drugs were posted on the program website weekly from May 29, 2004 to December 31, 2005, as the largest undertaking in the history of government-sponsored information release began with the hope of promoting competition by facilitating access to prices. A large panel of pharmacy-level drug price data collected from the Medicare website indicates that there was significant and persistent dispersion in prices across cards throughout the program. Moreover, the time-path of prices was non-monotonic; the prices declined initially when consumers were choosing cards but rose later when subscribers were unable to switch from one card to another. In contrast, contemporaneous control prices from on-line drug retailers, which were unrelated to the program, rose steadily over time, indicating that MDDCP prices evolved in a way different from the general evolution of prices outside the program. In view of the fact that the program rules prevented consumers from changing their cards at will, the evolution of MDDCP prices is consistent with certain models of dynamic price competition with consumer switching costs, such as Klemperer’s (1987a,b). Estimates of potential savings from purchasing at program prices are also provided.
... Com o mecanismo de seguro saúde há indícios que ocorra um sobre-consumo de medicamentos, além disto durante o processo de busca pelo medicamento, ou seja o processo de compra, os consumidores não fazem pesquisa de preço em farmácias ou preocupam-se em adquirir o medicamento de marca mais barata, pois não têm incentivos financeiros para isto (SORENSEN, 2001). ...
... Kessler and McClellan (2000) report that the average patient will travel approximately five miles for inpatient care. 8 Retail pharmacy markets are typically local markets since consumers tend not to travel long distances to purchase prescription drugs (Sorensen, 2001). 9 6 I match the ICD codes in the definition to the diagnosis codes for each discharge to construct the main outcome variable. ...
This paper analyzes how prices in the retail pharmaceutical market affect health care utilization. Specifically, I study the impact of Walmart’s $4 Prescription Drug Program on utilization of antihypertensive drugs and on hospitalizations for conditions amenable to drug therapy. Identification relies on the change in the availability of cheap drugs introduced by Walmart’s program, exploiting variation in the distance to the nearest Walmart across ZIP codes in a difference-in-differences framework. I find that living close to a source of cheap drugs increases utilization of antihypertensive medications by 7 percent and decreases the probability of an avoidable hospitalization by 6.2 percent.
... So far, studies of these markets have either ignored transaction prices and abstracted from the price-setting mechanism actually used in the market (see for instance Berry et al. (2004) in their study of the demand for new automobiles), or assumed monopoly pricing (see Adams et al. (2009) in their analysis of sub-prime used-car loans). The focus of the empirical search literature on the other hand has been on posted-price markets and/or assumes exogenous price distributions (see for instance Sorensen (2001), Hortaçsu and Syverson (2004), Hong andShum (2006), De Los Santos et al. (2011), and Honka (2012)). Finally, there is also a growing empirical literature on the relationship between bargaining and price dispersion. ...
This paper develops and estimates a search and bargaining model designed to measure the welfare loss associated with frictions in oligopoly markets with negotiated prices. We use the model to quantify the consumer surplus loss induced by the presence of search frictions in the Canadian mortgage market, and evaluate the relative importance of market power, inefficient allocation, and direct search costs in explaining the loss. Our results suggest that search frictions reduce consumer surplus by almost 100, 000 loan, and that 17% of this reduction can be associated with discrimination, 30% with inefficient matching, and the remainder with the search cost. In addition, we find that product differentiation attenuates the effect of search frictions by reducing the cost of gathering quotes and improving efficiency, while posted prices do so through the ability of the first-mover to price discriminate. In contrast , competition amplifies the welfare effect of search frictions. Despite this, the overall effect of competition is to increase aggregate consumer surplus and drive prices down, but these effects are not spread equally across consumers: those with low search costs benefit more from competition.
... The paper also contributes to the search literature by explicitly introducing observed product differentiation into a search problem. Important related papers include Sorensen (2001), Hortacsu and Syverson (2004), and Hong and Shum (2006). Similarly, a related literature in Marketing and IO is concerned with heterogeneous consideration-sets (e.g. ...
We propose a model of bank choice and price negotiation that incorporates three key features of the market: differentiated services, heterogeneous bank valuations of consumers, and search costs. More generally, our goal is to build and estimate a tractable empirical model to analyze differentiated markets in which prices are negotiated rather than posted.
... While the previous section has an intuitive outcome regarding the dependence of markup on the cost of a good, it would be more realistic to allow consumers to choose an effort according to the amount of money involved. For instance Sorensen (2001), while related to search costs, states that consumers put a higher search effort for pharmaceutical products that they buy more often. Here the standard deviation of the recall error will be associated to the effort they will be putting in remembering the exact prices, which is denoted by Σ. ...
... Structural signaling models have been developed to solve the signal extraction process to determine how much to revise consumer estimate of product quality (e.g., Erdem et al. 2006). Consumer search models have also emerged to investigate how (rational) consumers weigh the costs and benefits of search and how they examine its implications on price dispersion (Sorensen 2001). With the rise of the Internet, a number of authors have adopted these models to test the implications of lower search costs and higher penetration rate of Internet search engines. ...
Structural models integrate behavioral and psychological decision theory into economics models and are more aligned with the true underlying economic primitives of the consumers. This allows researchers to investigate more behavior-driven and process-oriented customer decision processes such as learning of product attributes, formation of a consideration sets, stockpiling, and flexible consumption that cannot be easily handled by traditional marketing models. Chintagunta et al. (2006) gives an excellent review on the development and applications of structural models in marketing for modeling both consumer demand and firm competition.
Recent interests in consumer research have diversified from frequently purchased packaged goods to services, high-tech, Internet, and information industries. Consumers are observed to be more sophisticated, long-term oriented, risk averse, and rational when making purchase and consumption decisions in these product categories. Structural models are better choices to capture the nature of the sophisticated decision process under the new marketing environment and engineering. Given the thorough review of Chintagunta et al. (2006), I will focus only on the dynamic structural demand models that have the components of information processing, rational expectation, and/or endogenous decisions to trade off current and future utilities, and I discuss some current marketing issues that can be most appropriately addressed by these models.
... Stahl, 1989). Sorensen (2000 Sorensen ( , 2001) considers the retail market for prescription drugs. Sorensen (2000) concludes that less than one-third of the price dispersion can be attributed to pharmacy heterogeneity. ...
This paper looks into the search behavior of consumers in the market for health insurance contracts. We consider the recent health insurance reform in The Netherlands, where a private-public mix of insurance provision was replaced by a system based on managed competition. Although all insurers offer the same basic package (determined by the government), there is substantial premium dispersion. We develop a simple consumer search model containing the main features of the Dutch health insurance system. This model provides us with a number of hypotheses, which we test using data from the Dutch Health Care Consumer Panel. The data confirm the standard predictions on consumer choice (i.e. there is adverse selection and a lower premium increases coverage). We also find that consumers with lower search costs are more likely to receive a group contract offer. This generates a situation of price discrimination where individuals without group contracts and higher search costs pay higher premiums and buy lower insurance coverage.
... Structural signaling models have been developed to solve the signal extraction process to determine how much to revise consumer estimate of product quality (e.g., Erdem, Keane and Sun 2005). Consumer search models have also emerged to investigate how (rational) consumers weigh the costs and benefits of search and examine its implications on price dispersion (Sorensen 2001). With the rise of the Internet, a number of authors have adopted these models to test the implications of lower search costs and higher penetration rate of Internet search engines. ...
Structural models integrate behavioral and psychological decision theory into economics models and are more aligned with the true underlying economic primitives of the consumers. This allows researchers to investigate more behavior-driven and process-oriented customer decision processes such as learning of product attributes, formation of a consideration sets, stockpiling, and flexible consumption that cannot be easily handled by traditional marketing models. Chintagunta et al. (2006) gives an excellent review on the development and applications of structural models in marketing for modeling both consumer demand and firm competition. Recent interests in consumer research have diversified from frequently purchased packaged goods to services, high-tech, Internet, and information industries. Consumers are observed to be more sophisticated, long-term oriented, risk averse, and rational when making purchase and consumption decisions in these product categories. Structural models are better choices to capture the nature of the sophisticated decision process under the new marketing environment and engineering. Given the thorough review of Chintagunta et al. (2006), I will focus only on the dynamic structural demand models that have the components of information processing, rational expectation, and/or endogenous decisions to trade off current and future utilities, and I discuss some current marketing issues that can be most appropriately addressed by these models.
... Stahl, 1989). Sorensen (2000 Sorensen ( , 2001) considers the retail market for prescription drugs. Sorensen (2000) concludes that the data are in agreement with search theory and that less than one-third in prize dispersion can be attributed by pharmacy heterogeneity. ...
The Netherlands introduced a new health insurance system in January 2006, a system based on managed competition. Such a system critically hinges on consumers that search. It is for this reason we think it is important to investigate the extend to which consumers search, how they search and why they search ´or don’t search. The price dispersion observed in the insurance market after the reform suggests the number of consumers that searches is low. We set up a search model for insurance that includes the main features of the Dutch health insurance market after the reform and test the hypotheses from this model on the data.
... This can potentially be explained by the Lal and Matutes model in which two firms 80 simultaneously use the same good as a loss leader. Sorensen (2001) looks at price dispersion in the retail market for prescription drugs. He fits the data to a model of costly search and indeed finds evidence that dispersion is an outcome of imperfect search. ...
These four essays concern the theory of games and its application to economic theory. The first two, closely linked, chapters are an investigation into the foundational question of the sensitivity of the predictions of game theory to higher-order beliefs. Impact of Higher-Order Uncertainty with Muhamet Yildiz In some games, the impact of higher-order uncertainty is very large, implying that present economic theories may be misleading as these theories assume common knowledge of the type structure after specifying the first or the second orders of beliefs. Focusing on normal-form games in which the players' strategy spaces are compact metric spaces, we show that our key condition, called "global stability under uncertainty," implies a variety of results to the effect that the impact of higher-order uncertainty is small. Our central result states that, under global stability, the maximum change in equilibrium strategies due to changes in players' beliefs at orders higher than k is exponentially decreasing in k. Therefore, given any need for precision, we can approximate equilibrium strategies by specifying only finitely many orders of beliefs. Finite-Order Implications of Any Equilibrium with Muhamet Yildiz Present economic theories make a common-knowledge assumption that implies that the first or second-order beliefs determine all higher-order beliefs.
... These features, along with the potential unreliability of retailers to deliver products when promised, can enhance the importance of seller reputation in online markets, thereby leading to retailer differentiation and price dispersion even in the absence of consumer search costs and explicit product heterogeneity. A challenging extension would be to combine the equilibrium search models considered in this article with rich individual-level datasets (such as supermarket scanner panel datasets or the individual-level drug purchase dataset used in Sorensen (2001)). It would be interesting to investigate how to exploit the equilibrium restrictions of the theoretical search models in identifying consumer search costs with more detailed data. ...
We show how the equilibrium restrictions implied by standard search models can be used to estimate search-cost distributions using price data alone. We consider both sequential and nonsequential search strategies, and develop estimation methodologies that exploit equilibrium restrictions to recover estimates of search-cost heterogeneity that are theoretically consistent with the search models. We illustrate the methods using online prices for several economics and statistics textbooks. Ordering information: This article can be ordered from http://gemini.econ.umd.edu/cgi-bin/rje_online.cgi?action=buy&year=2006&issue=sum&page=257&tid=30492&sc=1869P1N9 .
... This is a remarkable result, given that no data on the search process itself is available. Sorensen (2001) is perhaps the closest paper to what we do. she estimates a model where consumers go around local farmacies looking for the best price of a particular drug. ...
In this paper we estimate a structural model of search for differentiated products, using a unique dataset of individual search histories for hotels online. We propose and implement an identification strategy that allows to separately estimate consumer's beliefs, search costs and preferences. Learning plays an essential role in this strategy: it creates variation of posterior beliefs across consumers that's orthogonal to variation in search costs. We obtain two kinds of results. First, we estimate consumer's demand from the search model and compare it to results from the static model. We find that ignoring the endogeneity of choice sets leads to biased estimates: in particular, the aggregate price elasticity is over-estimated by about 80%. Second, we attempt to evaluate an empirical performance of a model of rational search. The mean search cost is estimated to be around 40 dollars, and median is 30 dollars; however, there is also a significant variation of search costs among population. A test between models of search from known (Stigler 1967) and from unknown (Rothschild 1974) distribution favors the second one: we find a statistically significant amount of Bayesian learning, even though it doesn't seem to affect demand estimates in an economically meaningful way.
... On the theoretical side, Massa [2000] and Mamaysky and Spiegel [2001] explore the driving forces of fund creation. 8. Sorensen [2001] is a recent attempt in the industrial organization literature to estimate search costs using consumer-level product choice data. Examples of structural estimation of search models in labor markets include Flinn and Heckman [1982], Eckstein and Wolpin [1990], and van den Berg and Ridder [1998]. ...
Two salient features of the competitive structure of the U.S. mutual fund industry are the large number of funds and the sizeable dispersion in the fees funds charge investors, even within narrow asset classes. Portfolio financial performance differences alone do not seem able to fully explain these features. We investigate whether non-portfolio fund differentiation and information/search frictions also play a role in creating these observed industry characteristics. We focus on their impact in a case study of the retail S&P 500 index funds sector. We find that fund proliferation and price dispersion also exist in this sector, despite the funds' financial homogeneity. Furthermore, there was a marked shift in sector assets to more expensive (often newly entered) funds throughout our sample period. Our analysis indicates that these observations are consistent with the presence of both non-portfolio differentiation and information/search frictions. Structural estimation of a novel search-over-differentiated-products model reveals that reasonable magnitudes of investor search costs can explain the considerable price dispersion in the sector, and consumers seem to value funds'' observable attributes such as fund age and the number of other funds in the same fund family in largely plausible ways. The results also suggest that the substantial increase in mutual fund market participation observed during our sample, and the corresponding purchase decisions of novice investors, drove the shift in assets toward more expensive funds. We also find evidence consistent with the presence of switching costs, as distinct from search costs. Using structural estimates of demand parameters and search costs, we investigate the possibility that there are too many sector funds from a social welfare standpoint. The results of this exercise indicate that restricting entry would yield nontrivial gains from reduced search costs and productivity gains from scale economies, but these may be counterbalanced by losses from increased market power and reduced product variety.
... All N 2 uninformed beneficiaries know only a subset of plan characteristics as well as the maximum utility available to them if they decide to search, while the N 1 informed beneficiaries have full information on all plan characteristics. This type of information structure is similar to Sorensen (2001), where buyers know the maximum utility available to them but must engage in costly search to match plans with utility values. ...
There is increasing evidence suggesting that Medicare beneficiaries do not make fully informed decisions when choosing among alternative Medicare health plans. To the extent that deciphering the intricacies of alternative plans consumes time and money, the Medicare health plan market is one in which search costs may play an important role. To account for this, we split beneficiaries into two groups--those who are informed and those who are uninformed. If uninformed, beneficiaries only use a subset of covariates to compute their maximum utilities, and if informed, they use the full set of variables considered. In a Bayesian framework with Markov Chain Monte Carlo (MCMC) methods, we estimate search cost coefficients based on the minimum and maximum statistics of the search cost distribution, incorporating both horizontal differentiation and information heterogeneities across eligibles. Our results suggest that, conditional on being uninformed, older, higher income beneficiaries with lower self-reported health status are more likely to utilize easier access to information.
... This finding is in line with the results reported by Johnson et al. (2004), who studied consumer click-through behavior online. In a study of the market for prescription drugs, Sorensen (2001) reports a similar finding that only between 5% to 10% of the consumers conduct an exhaustive search. On the basis of this evidence, our results for the case that search costs are relatively high may be important for the way mergers are analyzed by anti-trust authorities. ...
We study mergers in a market where N firms sell a homogeneous good and consumers search sequentially to discover prices. The main motivation for such an analysis is that mergers generally affect market prices and thereby, in a search environment, the search behavior of consumers. Endogenous changes in consumer search may strengthen, or alternatively, offset the primary effects of a merger. Our main result is that the level of search costs are crucial in determining the incentives of firms to merge and the welfare implications of mergers. When search costs are relatively small, mergers turn out not to be profitable for the merging firms. If search costs are relatively high instead, a merger causes a fall in average price and this triggers search. As a result, non-shoppers who didn’t find it worthwhile to search in the pre-merger situation, start searching post-merger. We show that this change in the search composition of demand makes mergers incentive-compatible for the firms and, in some cases, socially desirable.
... These features, along with the potential unreliability of retailers to deliver products when promised, can enhance the importance of seller reputation in online markets, thereby leading to retailer differentiation and price dispersion even in the absence of consumer search costs and explicit product heterogeneity. A challenging extension would be to combine the equilibrium search models considered in this article with rich individual-level datasets (such as supermarket scanner panel datasets or the individual-level drug purchase dataset used in Sorensen (2001)). It would be interesting to investigate how to exploit the equilibrium restrictions of the theoretical search models in identifying consumer search costs with more detailed data. ...
How large would consumer search costs have to be in order for them to rationalize online price dispersion? We evaluate the ability of equilibrium search models to explain observed patterns of price dispersion in online markets by developing a methodology for estimating equilibrium search-based price dispersion models. We consider models of both sequential and nonsequential search strategies, and exploit equilibrium restrictions to recover estimates of search cost heterogeneity which are theoretically consistent with the search models. For a number of online electronics and new book markets, our estimates show that the nonsequential search model uniformly yields more realistic results for the search costs. However, the large magnitudes for the search costs suggests that the search models may be incomplete as a descriptor of consumer behavior and firm pricing in online markets.
Using contract‐level data for the Canadian mortgage market, this paper provides evidence of an “invest‐and‐harvest” pricing pattern. We build a dynamic model of price negotiation with search and switching frictions to capture key market features. We estimate the model and use it to investigate the effects of market frictions and the resulting dynamic competition on borrowers' and banks' payoffs. We show that dynamic pricing and the presence of search and switching costs have important implications for public policies.
Objective
To characterize geographic variability of generic benign prostatic hyperplasia (BPH) medications in order to improve drug price transparency and improve patient access to affordable medication sources. This is of interest because BPH is one of the most common chronic diseases in men and contributes to individual healthcare cost. Medical therapy is the main treatment modality for BPH, burdening patients with lifelong medication expenses which may impact adherence and subsequent outcomes. With an aging population, this is compounded by many older individuals requiring multiple daily medications.
Methods
All pharmacies within a 25-mile radius of our institution were identified and classified as chain, wholesale or independent. The out-of-pocket price for a 30-day supply of tamsulosin (0.4 mg), finasteride (5 mg), oxybutynin (5 mg TID), and oxybutynin 10 mg XL were obtained using a scripted telephone survey. Multivariable linear regression assessed the association between census-tract level demographic and socioeconomic factors and disparate generic out-of-pocket drug-pricing.
Results
The response rate was 93% with 255 pharmacies across 173 census tracts providing data. By pharmacy type, there was up to 5.5-fold variation in median out-of-pocket drug prices for the most common BPH medications. Demographic and socioeconomic factors were not significantly associated with generic BPH drug price variation.
Conclusion
The out-of-pocket price of generic medications for BPH varies significantly between pharmacies in a geographically-confined area. This study highlights the need for quality improvement initiatives that empower patients to price-compare and improve drug price transparency.
In this paper, a general framework is built up to model the dynamic of consumer health plan choice and individual health insurance market competition. A primary goal is to identify driving forces to individual health insurance equilibrium market coverage and premium. In the baseline model, we introduce plan quality information search cost as an additional determinant to consumer’s plan choice. Health insurers compete under the Hotelling’s game theory framework. Equilibrium solutions of the baseline model highlight the importance of budget limit and information search cost to health plan enrollment. The more important objective is to examine the impact of market entrants on equilibrium insurance market coverage and plan prices. In the model with market entry, we add an additional dimension to the baseline model. Equilibrium solutions and numerical studies show positive impact of higher insurance market coverage and lower health plan prices. The Affordable Care Act (ACA) brought multiple unprecedented changes to the health insurance market and provided opportunities to study market dynamics and driving forces. The ACA health insurance exchange market experience shows consistency with our model findings even at the early stage of implementation. More importantly, market observations suggest that entry barriers of claim costs and information search cost are high for entrants.
We use an internet survey conducted among a representative random sample of drivers in the State of Ohio consisting of a choice experiment designed to examine the mechanism driving asymmetric search. The internet survey affords us the opportunity to overcome endogeneity difficulties by imposing exogenous price changes on gasoline consumers to examine the decision-making process behind intended search decisions. We randomly assigned participants to one of five price treatments (either 2.5 or 5% above or below their reported expected price, or no change). We provide a simple empirical model to derive testable implications under prospect theory and use the internet survey to test them. Results indicate that among the respondents who faced prices below their expected price, only 12% chose to search, whereas 45% searched when prices were above. Further, we find results consistent with asymmetric search being driven by prospect theory. The change in consumers’ willingness to search is twice as large when prices exceed expectations by 2.5% relative to when prices exceed them by 5% suggesting that consumers derive utility of finding a good deal evaluated relative to a reference price. We show that this result is inconsistent with standard utility theory or consumers using alternative reference prices.
This paper considers consumer behavior in the Dutch market for health insurances, which is characterized by managed competition. We observe substantial price dispersion for the same basic insurance package. We describe the market by a simple consumer search model, which we empirically test using data from the Dutch Health Care Consumer Panel. We argue that insurers use group contracts to target discounts to better informed consumers, reducing their incentives to search. Since search is essential for competition, this increases premiums and premium dispersion.
Background
Monitoring prescription drug utilization is important for both drug safety and drug marketing purposes. However, access to utilization data is often expensive, limited and not timely.
Objectives
To demonstrate and validate the use of web search engine queries as a method for timely monitoring of drug utilization and changes in prescribing behaviors.
Methods
Drug utilization time series were obtained from the Medical Expenditure Panel Survey and normalized search volume was obtained from Google Trends. Correlation between the series was estimated using a cross-correlation function. Changes in the search volume following knowledge events were detected using a cumulative sums changepoint method.
Results
Search volume tracks closely with the utilization rates of several seasonal prescription drugs. Additionally, search volume exhibits changes following known major knowledge events, such as the publication of new information.
Conclusions
Search volume provides a first order approximation to pharmaceutical utilization in the community and can be used to detect changes in prescribing behavior.
This paper reviews the recent contributions on the structural estimation of search costs. We first discuss some of the theoretical and empirical literature on price disper-sion and consumer search. We then argue that optimal design of competition policy needs the development of methods to identify and estimate search costs. We finally discuss the methods that have been proposed to date and the results obtained.
There is a large body of work that documents a strong, positive correlation between education and measures of health, but little is known about the mechanisms by which education might affect health. One possibility is that more educated individuals are more likely to adopt new medical technologies. We investigate this theory by asking whether more educated people are more likely to use newer drugs, while controlling for other individual characteristics, such as income and insurance status. Using the 1997 Medical Expenditure Panel Survey (MEPS), we find that more highly educated people are more likely to use drugs more recently approved by the FDA. We find that education only matters for individuals who repeatedly purchase drugs for a given condition, suggesting that the cost of searching for higher-quality treatments is lower for more educated people.
Internet shopbots allow consumers to almost instantly compare prices and other characteristics from dozens of sellers via
a single website. We estimate the magnitude of consumer search costs and benefits using data from a major shopbot for books.
For the median consumer, the estimated benefit from simply scrolling down to search lower screens is $6.55. This amounts to
about 60% of the observed price dispersion and suggests that consumers face significant search costs, even in this “nearly-perfect”
market. Price elasticities are relatively high compared to offline markets (−7 to −10 in our base model). Furthermore, contrary
to the common assumption, search intensity is not correlated with greater price sensitivity. Instead, consumers who search
multiple screens put relatively more weight on non-price factors like brand.
In a recent paper Hong and Shum [2006. Using price distributions to estimate search costs. Rand Journal of Economics 37, 257–275] present a structural method to estimate search cost distributions. We extend their approach to the case of oligopoly and present a new maximum likelihood method to estimate search costs. We apply our method to a data set of online prices for different computer memory chips. The estimates suggest that the consumer population can be roughly split into two groups which either have quite high or quite low search costs. Search frictions confer a significant amount of market power to the firms: Despite more than 20 firms operating in each of the markets, we estimate price-cost margins to be around 25%. The paper also illustrates how the structural method can be employed to simulate the effects of the introduction of a sales tax.
In this paper, we emphasize that choice sets generated by a search process have two properties: first, they are limited; second, they are endogenous to preferences. Both factors lead to biased estimates in a static demand framework that takes choice sets as given. To correct for this bias, we estimate a structural model of search for differentiated products, using a unique dataset of consumer online search for hotels. Within a nested logit utility model, we show that the mean utility function and the search cost distribution of a representative consumer are non-parametrically identified, given our data. Using our model's estimates, we quantify both sources of bias: they lead to overestimation of price elasticity by a factor of five and four, respectively. The median search cost is about 38 dollars per 15 hotels; we also present some evidence on multi-modality of search cost distribution.
This paper studies the consequence of an imprecise recall of the price by the consumers in the Bertrand price competition model for a homogeneous good. It is shown that firms can exploit this weakness and charge prices above the competitive price. This markup increases for rougher recall of the price. If firms have different production costs, those with higher costs are not driven out of the market. However they choose to have a higher price in equilibrium, therefore price dispersion arises. It is shown that firms behave on average as a monopolist with stricter demand and that price dispersion increases with the price recall errors. If bigger recall errors happen, then both consumers and firms on the aggregate level are worse off, for some parameter choices. Furthermore being given the irrational choice that some consumers make, there are situations where the protection of a monopolist against entrants is a welfare maximizing policy. The introduction of more firms in the market does not have a significant impact on the prices. Even though the presented model is static, it can be interpreted as a stage game of an infinitely repeated game where a Nash Equilibrium is played in every stage. The intuition is that consumers do not actually seek information before every purchase, but have a vague idea of the price they faced in previous purchases.
We investigate the role that nonportfolio fund differentiation and information/search frictions play in creating two salient features of the mutual fund industry: the large number of funds and the sizable dispersion in fund fees. In a case study, we find that despite the financial homogeneity of S&P 500 index funds, this sector exhibits the fund proliferation and fee dispersion observed in the broader industry. We show how extra-portfolio mechanisms explain these features. These mechanisms a lso suggest an explanation for the puzzling late-1990s shift in sector assets to more expensive (and often newly entered) funds: an influx of high-information-cost novice investors. © 2004 MIT Press
It has long been recognized that brand name prescription medications are more expensive in the US than in Canada. However, non-peer-reviewed reports contend that this is not the case for generic medications.
To compare prices for generic prescription medications in the US and Canada.
A telephone and internet survey of pharmacies was conducted for the top 19 dispensed generic medications available in both countries. Twelve pharmacies in total were selected: six from the US (three online and three 'walk-in') and six from Canada (three online and three walk-in). Data were collected from March to April 2007.The main outcome measure was the total purchase price in $US at the day's exchange rate in 2007, obtained from each of the 12 pharmacies, for 100 doses of each of the 19 selected drugs.
Using the lowest quote for each selected drug, 12 of the 19 (63%) generic medications were least expensive in the US, with an average saving of 47% per drug for these 12 drugs. Seven of the 19 (37%) drugs were least expensive in Canada, with an average saving of 29% per drug for these seven drugs. Overall, there was a sizable variation in prices for the same generic medications within and between the US and Canada.
The lowest priced generic medications were not consistently found in either the US or Canada. The price controls and ensuing savings applied in Canada to prescription patented medications do not fully extend to generic medications.
Price dispersion among commodity goods is typically attributed to consumer search costs. We explore the magnitude of consumer search costs using a data set obtained from a major Internet shopbot. For the median consumer, the benefits to searching lower screens are 2.03. Interestingly, in our setting, consumers who search more intensively are less price sensitive than other consumers, reflecting their increased weight on retailer differentiation in delivery time and reliability. Our results demonstrate that even in this nearly-perfect market, substantial price dispersion can exist in equilibrium from consumers preferences over both price and non-price attributes
This paper builds a macroeconomic model of equilibrium unemployment in which firms persistently face difficulties in selling their production and this affects their decisions to create jobs. Due to search-frictins on the product market, equilibrium unemployment is a U-shaped function of the ratio of total demand to total supply on this market. When prices are at their Competitive Search Equilibrium values, the unemployment rate is minimized. Yet, the Competitive Search Equilibrium is not efficient. Inflation is detrimental to unemployment.
This paper presents an empirical examination of oligopoly pricing and consumer search. The theoretical model allows for sequential and non-sequential search and using the theoretical restrictions firm and consumer behavior impose on the data we study the empirical validity of the models. Two equilibria arise: one with costless search and the other with costly search. We find that the costless search equilibrium works well for products with a relatively low value, and, by implication, a small number of sellers. By contrast, the costly search equilibrium explains the observed data in a manner that is consistent with the underlying theoretical model for almost all products (for 86 out of 87!).
This research compares several approaches to inference in the multinominal profit model, based on two Monte Carlo experiments for a seven choice model. The methods compared are the simulated maximum likelihood estimator using the GHK recursive probability simulator, the method of simulated moments estimator using the GHK recursive simulator and kernel-smoothed frequency simulators, and posterior means using a Gibbs sampling-data augmentation algorithm. Overall, the Gibbs sampling algorithm has a slight edge, with the relative performance of MSM and SML based on the GHK simulator being difficult to evaluate. The MSM estimator with the kernel-smoothed frequency simulator is clearly inferior. Copyright 1994 by MIT Press.
It is shown that wquilibria with dispersed prices exist in environments with identical and rational agents on both sides of the market. In particular, the original Stigler model of nonsequential search often has many equilibria, some with price dispersion. Also, price dispersion holds in equilibrium in general if search is "noisy," i.e., there is some chance of learning two or more prices when an agent is looking for one price.
A finite number of identical stores sell a homogeneous good to consumers with heterogeneous search costs who search sequentially with perfect recall and without replacement. Consumers' search rules are optimal with respect to the stores' pricing strategies, and consumers must visit a store to obtain information about its actual price. A symmetric Nash equilibrium (SNE) always exists. The model exhibits a variety of equilibrium behaviors between monopoly pricing and marginal-cost pricing. The competitiveness of a market depends crucially on the shape of the search cost distribution, moreso than on the number of competing firms.
This study seeks to establish the empirical importance of price dispersion due to costly consumer search by examining retail prices for prescription drugs. Posted prices in two geographically distinct markets are shown to vary considerably across pharmacies within the same market, even after one controls for variation due to pharmacy differences. Pharmacy heterogeneity accounts for at most one-third of the observed price dispersion. The empirical analysis hinges on the observation that consumers' incentives to price-shop depend on characteristics of the drug therapy. Cross-sectional patterns in price distributions across drugs are consistent with the predictions of a search model: prices for repeatedly purchased prescriptions (for which the expected benefits of search are highest) exhibit significant reductions in both dispersion and price-cost margins.
N identical stores compete by choosing prices for a homogeneous good with constant marginal costs. Consumers search sequentially with perfect recall for the lowest price. One class of consumers, called shoppers, have zero search costs, while all other consumers have a positive search cost, c. There is a unique symmetric Nash equilibrium price distribution with the property that it changes smoothly from "marginal cost pricing" when all consumers are shoppers and/or c = 0 and "monopoly pricing" when no consumers are shoppers. Remarkably, as the number of stores increases, the Nash equilibrium becomes more monopolistic. Copyright 1989 by American Economic Association.
This paper proposes a simple modification of a conventional generalized method of moments estimator for a discrete response model, replacing response probabilities that require numerical integration with estimators obtained by Monte Carlo simulation. This method of simulated moments does not require precise estimates of these probabilities, as the law of large numbers operating across observations controls simulation error, and, hence, can use simulations of practical size. The method is useful for models such as high-dimensional multinomial probit, where computation has previously restricted applications. Statistical properties are established using empirical process methods that can handle discontinuities introduced by simulation. Copyright 1989 by The Econometric Society.
Equilibrium price distributions (for a homogeneous product) consistent with individual incentives are investigated. They arise
in informationally imperfect markets in which the only primitive datum is the distribution of search costs. It is shown that
single, multi- and continuous price distributions are all viable long-run phenomena depending on the nature of search costs.
A method for computing equilibrium price distributions is also provided.
: This paper considers mixed, or random coefficients, multinomial logit (MMNL) models for discrete response, and establishes the following results: Under mild regularity conditions, any discrete choice model derived from random utility maximization has choice probabilities that can be approximated as closely as one pleases by a MMNLmodel. Practical estimation of a parametric mixing family can be carried out by Maximum Simulated Likelihood Estimation or Method of Simulated Moments, and easily computed instruments are provided that make the latter procedure fairly efficient. The adequacy of a mixing specification can be tested simply as an omitted variable test with appropriately defined artificial variables. An application to a problem of demand for alternative vehicles shows that MMNL provides a flexible and computationally practical approach to discrete response analysis. Acknowledgments: Both authors are at the Department of Economics, University of California, Berkeley CA 94720-3880...
The Internet has the potential to significantly reduce search costs by allowing consumers to engage in low-cost price comparisons online. This paper provides empirical evidence on the impact that the rise of Internet comparison shopping sites has had for the prices of life insurance in the 1990s. Using micro data on individual life insurance policies, the results indicate that, controlling for individual and policy characteristics, a 10 percent increase in the share of individuals in a group using the Internet reduces average insurance prices for the group by as much as 5 percent. Further evidence indicates that prices did not fall with rising Internet usage for insurance types that were not covered by the comparison websites, nor did they in the period before the insurance sites came online. The results suggest that growth of the Internet has reduced term life prices by 8 to 15 percent and increased consumer surplus by $115-215 million per year and perhaps more. The results also show that the initial introduction of the Internet search sites is initially associated with an increase in price dispersion within demographic groups, but as the share of people using the technology rises further, dispersion falls. We would like to thank Eric Anderson, Judy Chevalier, Mark Duggan, James Garven, Robert Hartwig, Thomas Hubbard, Ken Isenberg, Kent Jamison, John Johnson, Peter Klenow, Olivia Mitchell, Jim Poterba, Todd Sinai, Alan Sorensen, Mark Warshawsky, Alwyn Young and seminar participants at Wharton, University of South Carolina, and the 2000 ARIA meetings for helpful comments, and Jeffrey Butler, Andrew Lee, and Soojin Yim for excellent research assistance. We would also like to thank Ken Isenberg and LIMRA International for assistance with data from the LIMRA Buyer Studi...