Jean-Pierre Dubé

New: Discrete-Choice Models and Representative Consumer Theory
Date Posted:Wed, 15 Jun 2022 06:16:10 -0500
We establish the Hurwicz-Uzawa integrability of the broad class of discrete-choice additive random-utility models of individual consumer behavior with perfect substitutes preferences and divisible goods. We derive the corresponding indirect uility function and then establish a representative consumer formulation for this entire class of models. The representative consumer is always normative, facilitating aggregate welfare analysis. These findings should be of interest to the literatures in macro, trade, industrial organization, labor and ideal price index measurement that use representative consumer models, such as CES and its variants. Our results generalize such representative consumer formulations to the broad, empirically-relevant class of models of behavior that are routinely used in the discrete-choice analysis of micro data, including specifications that do not suffer from the IIA property and that allow for heterogeneous consumer preferences and incomes. When products are ...

New: Intangible Marketing Capital
Date Posted:Wed, 15 Jun 2022 06:09:20 -0500
We present several empirical facts about trends in marketing investment in the US. We also present estimates of the private value of brands to firms and aggregate intangible brand capital stocks created by these investments. These investments include the creation and maintenance of a brand name and all its corresponding brand elements (e.g., awareness, reputation, image, etc.), along with the investments in the labor force used to manage and oversee the branding. We then discuss the private benefits to firms from their advertising and the established academic wisdom for the striking magnitude of marketing outlays. Finally, we explore the welfare implications of investments in brand capital.

REVISION: Discrete-Choice Models and Representative Consumer Theory
Date Posted:Wed, 01 Jun 2022 15:05:24 -0500
Abstract We establish the Hurwicz-Uzawa integrability of the broad class of discrete-choice additive random-utility models of individual consumer behavior with perfect substitutes preferences and divisible goods. We derive the corresponding indirect uility function and then establish a representative consumer formulation for this entire class of models. The representative consumer is always normative, facilitating aggregate welfare analysis. These findings should be of interest to the literatures in macro, trade, industrial organization, labor and ideal price index measurement that use representative consumer models, such as CES and its variants. Our results generalize such representative consumer formulations to the broad, empirically-relevant class of models of behavior that are routinely used in the discrete-choice analysis of micro data, including specifications that do not suffer from the IIA property and that allow for heterogeneous consumer preferences and incomes. When ...

REVISION: Non-parametric Estimation of Habitual Brand Loyalty
Date Posted:Fri, 22 Apr 2022 08:27:06 -0500
Testing for and measuring habitual brand loyalty (HBL) is one of the classic questions in quantitative marketing. We propose a nonparametric test for HBL using a ``dynamic potential outcomes' model that resolves the classic identification challenge of decoupling state dependence and unobserved heterogeneity. We then propose a semi-parametric test for forward-looking behavior to assess whether consumers rationally plan their HBL. Case studies of several large consumer packaged goods categories reveal a non-trivial extent of HBL in consumers’ brand choices. Semi-parametric evidence for forward-looking consumer behavior strongly rejects the myopic discrete-choice model in favor of the dynamic discrete-choice model with a freely-varying discount factor. However, consumers are found to be considerably more impatient than would be implied by the real rate of interest. Nevertheless, the long-run price elasticities from a dynamic discrete-choice model are found to be larger in magnitude ...

REVISION: Discrete-Choice Models and Representative Consumer Theory
Date Posted:Fri, 01 Apr 2022 10:39:25 -0500
We establish the Hurwicz-Uzawa integrability of the broad class of discrete-choice additive random-utility models of individual consumer behavior with perfect substitutes preferences and divisible goods. We derive the corresponding indirect uility function and then establish a representative consumer formulation for this entire class of models. The representative consumer is always normative, facilitating aggregate welfare analysis for this broad class of models. When products are indivisible, we show that integrability fails, indicating the lack of a representative consumer formulation.These findings should be of interest to the literatures in macro, trade, industrial organization, labor and ideal price index measurement that use representative consumer models, such as CES and its variants. Our results generalize such representative consumer formulations to the broad, empirically-relevant class of models of behavior that are routinely used in the discrete-choice analysis of micro ...

REVISION: The Persuasive Effect of Fox News: Non-Compliance with Social Distancing During the COVID-19 Pandemic
Date Posted:Tue, 07 Sep 2021 05:00:30 -0500
To what extent do mass media outlets influence viewers' trust in scientific evidence and compliance with behavior recommended by scientific experts? Exploiting the US lock-down period of the COVID-19 pandemic in early 2020, we analyze a large longitudinal database that combines daily stay-at-home behavior from approximately 8 million mobile phones and local viewership of cable news networks. Early in the pandemic, several of Fox News' hosts downplayed the severity of the pandemic and the risks associated with the transmission of the virus. A combination of regression analysis and a natural experiment finds that a 10% increase in viewership of Fox News in a zip code causes a 0.76 percentage point reduction in compliance with stay-at-home behavior. The results imply a media persuasion rate that is larger than typical advertising persuasion rates on consumer behavior. Similar analyses using viewership of MSNBC and CNN, which supported lock-down measures, were inconclusive but suggested ...

REVISION: The Persuasive Effect of Fox News: Non-Compliance with Social Distancing During the COVID-19 Pandemic
Date Posted:Mon, 23 Aug 2021 06:20:11 -0500
To what extent do mass media outlets influence viewers' trust in scientific evidence and compliance with behavior recommended by scientific experts? Exploiting the US lock-down period of the COVID-19 pandemic in early 2020, we analyze a large longitudinal database that combines daily stay-at-home behavior from approximately 8 million mobile phones and local viewership of cable news networks. Early in the pandemic, several of Fox News' hosts downplayed the severity of the pandemic and the risks associated with the transmission of the virus. A combination of regression analysis and a natural experiment finds that a 10% increase in viewership of Fox News in a zip code causes a 0.76 percentage point reduction in compliance with stay-at-home behavior. The results imply a media persuasion rate that is larger than typical advertising persuasion rates on consumer behavior. Similar analyses using viewership of MSNBC and CNN, which supported lock-down measures, were inconclusive but suggested ...

REVISION: Personalized Pricing and Consumer Welfare
Date Posted:Thu, 24 Jun 2021 06:03:45 -0500
We study the welfare implications of personalized pricing, an extreme form of third-degree price discrimination implemented with machine learning for a large, digital firm. Using data from a unique randomized controlled pricing field experiment we train a demand model and conduct inference about the effects of personalized pricing on firm and consumer surplus. In a second experiment, we validate our predictions in the field. The initial experiment reveals unexercised market power that allows the firm to raise its price optimally , generating a 55% increase in profits. Personalized pricing improves the firm's expected posterior profits by an additional 19%, relative to the optimized uniform price, and by 86%, relative to the firm's unoptimized status quo price. Turning to welfare effects on the demand side, total consumer surplus declines 23% under personalized pricing relative to uniform pricing, and 47% relative to the firm's unoptimized status quo price. However, over 60% of ...

REVISION: The Persuasive Effect of Fox News: Non-Compliance with Social Distancing During the COVID-19 Pandemic
Date Posted:Mon, 03 Aug 2020 03:56:04 -0500
We test for and measure the effects of cable news in the US on regional differences in compliance with recommendations by health experts to practice social distancing during the early stages of the COVID-19 pandemic. We use a quasi-experimental design to estimate the causal effect of Fox News viewership on stay-at-home behavior by using only the incremental local viewership due to the quasi-random assignment of channel positions in a local cable line-up. We find that a 10% increase in Fox News cable viewership (approximately 0.13 higher viewer rating points) leads to a 1.3 percentage point reduction in the propensity to stay at home. We find a persuasion rate of Fox News on non-compliance with stay-at-home behavior during the crisis of about 5.7%- 28.4% across our various social distancing metrics.

REVISION: The Persuasive Effect of Fox News: Non-Compliance with Social Distancing During the COVID-19 Pandemic
Date Posted:Thu, 30 Jul 2020 04:03:57 -0500
We test for and measure the effects of cable news in the US on regional differences in compliance with recommendations by health experts to practice social distancing during the early stages of the COVID-19 pandemic. We use a quasi-experimental design to estimate the causal effect of Fox News viewership on stay-at-home behavior by using only the incremental local viewership due to the quasi-random assignment of channel positions in a local cable line-up. We find that a 10% increase in Fox News cable viewership (approximately 0.13 higher viewer rating points) leads to a 1.3 percentage point reduction in the propensity to stay at home. We find a persuasion rate of Fox News on non-compliance with stay-at-home behavior during the crisis of about 5.7%-28.4% across our various social distancing metrics.

REVISION: The Persuasive Effect of Fox News: Non-Compliance with Social Distancing During the COVID-19 Pandemic
Date Posted:Tue, 23 Jun 2020 11:45:36 -0500
We test for and measure the effects of cable news in the US on regional differences in compliance with recommendations by health experts to practice social distancing during the early stages of the COVID-19 pandemic. We use a quasi-experimental design to estimate the causal effect of Fox News viewership on stay-at-home behavior by using only the incremental local viewership due to the quasi-random assignment of channel positions in a local cable line-up. We find that a 10% increase in Fox News cable viewership (approximately 0.13 higher viewer rating points) leads to a 1.3 percentage point reduction in the propensity to stay at home. We find a persuasion rate of Fox News on non-compliance with stay-at-home behavior during the crisis of about 11.9%-25.7% across our various social distancing metrics.

REVISION: The Persuasive Effect of Fox News: Non-Compliance with Social Distancing During the Covid-19 Pandemic
Date Posted:Fri, 29 May 2020 03:57:06 -0500
We test for and measure the effects of cable news in the US on regional differences in compliance with recommendations by health experts to practice social distancing during the early stages of the COVID-19 pandemic. We use a quasi-experimental design to estimate the causal effect of Fox News viewership on stay-at-home behavior by using only the incremental local viewership due to the quasi-random assignment of channel positions in a local cable line-up. We find that a 10% increase in Fox News cable viewership (approximately 0.13 higher viewer rating points) leads to a 1.2 percentage point reduction in the propensity to stay at home. It follows that one rating point increase in Fox News viewership in a zipcode reduces the propensity to stay at home by 8.9 percentage points compared to the pre-pandemic average. We find a persuasion rate of Fox News on non-compliance with stay-at-home behavior during the crisis of about 33.5%-50% across our various social distancing metrics.

REVISION: The Persuasive Effect of Fox News: Non-Compliance with Social Distancing During the COVID-19 Pandemic
Date Posted:Mon, 18 May 2020 12:14:42 -0500
We test for and measure the effects of cable news in the US on regional differences in compliance with recommendations by health experts to practice social distancing during the early stages of the COVID-19 pandemic. We use a quasi-experimental design to estimate the causal effect of Fox News viewership on stay-at-home behavior by using only the incremental local viewership due to the quasi-random assignment of channel positions in a local cable line- up. The average partial effect of Fox News viewership in a zipcode implies that 1 percentage point increase in cable viewership reduces the propensity to stay at home by 8.9 percentage points compared to the pre-pandemic average. We find a persuasion rate of Fox News on non- compliance with stay-at-home behavior during the crisis of about 33.5% - 50% across our various social distancing metrics.

REVISION: The Persuasive Effect of Fox News: Non-Compliance with Social Distancing During the Covid-19 Pandemic
Date Posted:Sun, 17 May 2020 08:52:30 -0500
We test for and measure the effects of cable news in the US on regional differences in compliance with recommendations by health experts to practice social distancing during the early stages of the COVID-19 pandemic. We use a quasi-experimental design to estimate the causal effect of Fox News viewership on stay-at-home behavior by using only the incremental local viewership due to the quasi-random assignment of channel positions in a local cable line-up. The average partial effect of Fox News viewership in a zipcode implies that 1 percentage point increase in cable viewership reduces the propensity to stay at home by 8.9 percentage points compared to the pre-pandemic average. We find a persuasion rate of Fox News on non-compliance with stay-at-home behavior during the crisis of about 33.5%-50% across our various social distancing metrics.

REVISION: Random-Coefficients Logit Demand Estimation with Zero-Valued Market Shares
Date Posted:Tue, 21 Apr 2020 04:27:32 -0500
Although typically overlooked, many purchase datasets exhibit a high incidence of products with zero sales. We propose a new estimator for the Random-Coefficients Logit demand system for purchase datasets with zero-valued market shares. The identification of the demand parameters is based on a pairwise-differencing approach that constructs moment conditions based on differences in demand between pairs of products. The corresponding estimator corrects nonparametrically for the potential selection of the incidence of zeros on unobserved aspects of demand. The estimator also corrects for the potential endogeneity of marketing variables both in demand and in the selection propensities. Monte Carlo simulations show that our proposed estimator provides reliable small-sample inference both with and without selection-on-unobservables. In an empirical case study, the proposed estimator not only generates different demand estimates than approaches that ignore selection in the incidence of zero ...

New: Random-Coefficients Logit Demand Estimation with Zero-Valued Market Shares
Date Posted:Mon, 09 Mar 2020 20:24:41 -0500
Although typically overlooked, many purchase datasets exhibit a high incidence of products with zero sales. We propose a new estimator for the Random-Coefficients Logit demand system for purchase datasets with zero-valued market shares. The identification of the demand parameters is based on a pairwise-differencing approach that constructs moment conditions based on differences in demand between pairs of products. The corresponding estimator corrects non-parametrically for the potential selection of the incidence of zeros on unobserved aspects of demand. The estimator also corrects for the potential endogeneity of marketing variables both in demand and in the selection propensities. Monte Carlo simulations show that our proposed estimator provides reliable small-sample inference both with and without selection-on-unobservables. In an empirical case study, the proposed estimator not only generates different demand estimates than approaches that ignore selection in the incidence of ...

REVISION: Random-Coefficients Logit Demand Estimation with Zero-Valued Market Shares
Date Posted:Wed, 04 Mar 2020 05:33:29 -0600
Although typically overlooked, many purchase datasets exhibit a high incidence of products with zero sales. We propose a new estimator for the Random-Coefficients Logit demand system for purchase datasets with zero-valued market shares. The identification of the demand parameters is based on a pairwise-differencing approach that constructs moment conditions based on differences in demand between pairs of products. The corresponding estimator corrects nonparametrically for the potential selection of the incidence of zeros on unobserved aspects of demand. The estimator also corrects for the potential endogeneity of marketing variables both in demand and in the selection propensities. Monte Carlo simulations show that our proposed estimator provides reliable small-sample inference both with and without selection-on-unobservables. In an empirical case study, the proposed estimator not only generates different demand estimates than approaches that ignore selection in the incidence of zero ...

REVISION: State-Dependent Demand Estimation with Initial Conditions Correction
Date Posted:Thu, 27 Feb 2020 10:06:50 -0600
We analyze the initial conditions bias in the estimation of brand choice models with structural state dependence. Using a combination of Monte Carlo simulations and empirical case studies of shopping panels, we show that popular, simple solutions that mis-specify the initial conditions are likely to lead to bias even in relatively long panel datasets. The magnitude of the bias in the state dependence parameter can be as large as a factor of 2 to 2.5. We propose a solution to the initial conditions problem that samples the initial states as auxiliary variables in an MCMC procedure. The approach assumes that the joint distribution of prices and consumer choices, and hence the distribution of initial states, is in equilibrium. This assumption is plausible for the mature consumer packaged goods products used in this and the majority of prior empirical applications. In Monte Carlo simulations, we show that the approach recovers the true parameter values even in relatively short panels. ...

REVISION: Personalized Pricing and Customer Welfare
Date Posted:Fri, 21 Feb 2020 11:57:13 -0600
Abstract We study the welfare implications of personalized pricing, an extreme form of third-degree price discrimination implemented with machine learning for a large, digital firm. We conduct a randomized controlled pricing field experiment to train a demand model and to conduct inferences about the effects of personalized pricing on firm and customer surplus. In a second experiment, we validate our predictions out of sample. Personalized pricing improves the firm's expected posterior profits by 19%, relative to optimized uniform pricing, and by 86%, relative to the firm's status quo pricing. On the demand side, customer surplus declines slightly under personalized pricing relative to a uniform pricing structure. However, over 60% of customers benefit from personalized prices that are lower than the optimal uniform price. Based on simulations with our demand estimates, we find several cases where customer surplus increases when the firm is allowed to condition on more customer ...

REVISION: State-Dependent Demand Estimation with Initial Conditions Correction
Date Posted:Sat, 24 Aug 2019 14:56:47 -0500
We analyze the initial conditions bias in the estimation of brand choice models with structural state dependence. Using a combination of Monte Carlo simulations and empirical case studies of shopping panels, we show that popular, simple solutions that mis-specify the initial conditions are likely to lead to bias even in relatively long panel datasets. The magnitude of the bias in the state dependence parameter can be as large as a factor of 2 to 2.5. We propose a solution to the initial conditions problem that samples the initial states as auxiliary variables in an MCMC procedure. The approach assumes that the joint distribution of prices and consumer choices, and hence the distribution of initial states, is in equilibrium. This assumption is plausible for the mature consumer packaged goods products used in this and the majority of prior empirical applications. In Monte Carlo simulations, we show that the approach recovers the true parameter values even in relatively short panels. ...

New: The Economics of Brands and Branding
Date Posted:Wed, 12 Dec 2018 10:37:23 -0600
Consumer goods industries tend to be dominated by a small number of nationally-branded goods. In many cases, brands are one of the only sources of differentiation between otherwise physically undifferentiated products. In this chapter we survey the literature on the role of consumer branding and brand preferences in driving consumer demand. We also discuss the incentives for firms to invest in and build brands on the supply side. Finally, we discuss the measurement of the economic value of brands as intangible ...

REVISION: Food Deserts and the Causes of Nutritional Inequality
Date Posted:Wed, 21 Nov 2018 05:49:37 -0600
We study the causes of "nutritional inequality": why the wealthy eat more healthfully than the poor in the U.S. Exploiting supermarket entry, household moves to healthier neighborhoods, and purchasing patterns among households with identical local supply, we reject that neighborhood environments contribute meaningfully to nutritional inequality. Using a structural demand model, we find that exposing low-income households to the same products and prices available to high-income households reduces nutritional inequality by only nine percent, while the remaining 91 percent is driven by differences in demand. These findings counter the common notion that policies to reduce supply inequities, such as "food deserts," could play an important role in reducing nutritional inequality. By contrast, the structural results predict that means-tested subsidies for healthy food could eliminate nutritional inequality at a fiscal cost of about fifteen percent of the annual budget for the U.S. SNAP ...

REVISION: Microeconometric Models of Consumer Demand
Date Posted:Fri, 26 Oct 2018 10:16:31 -0500
A long literature has developed econometric methods for estimating individual-consumer-level demand systems that accommodate corner solutions. The increasing access to transaction-level customer purchase histories across a wide array of markets and industries vastly expands the prospect for improved customer insight, more targeted marketing policies and individualized welfare analysis. A descriptive analysis of a broad, CPG database indicates that most consumer brand categories offer a wide variety of differentiated offerings for consumers. However, consumers typically purchase only a limited scope of the available variety, leading to a very high incidence of corner solutions which poses computational challenges for demand modeling. Historically, these computational challenges have limited the applicability of microeconometric models of demand in practice, except for the special case of pure discrete choice (e.g., logit and probit). Recent advances in computing power along with ...

New: Consumer Misinformation and the Brand Premium: A Private Label Blind Taste Test
Date Posted:Wed, 17 Oct 2018 14:28:52 -0500
We run in-store blind taste tests between a retailer's private label food brands and the leading national brand counterparts in three large CPG categories. In a survey administered during the taste test, subjects self-report very high expectations about the relative quality of the private labels relative to national brands. However, they predict a relatively low probability of choosing them in a blind taste test. Surprisingly however, an overwhelming majority systematically chooses the private label in the blinded test. During the week after the intervention, the tested private label product market shares increase by 15 share points, on top of a base share of 8 share points. However, the effect diminishes to 8 share points during the second to fourth week after the test, and to 2 share points during the second to fifth month after the test. Using a structural model of demand, we find that the intervention increases the preference for the private label brands, but decreases the ...

REVISION: Microeconometric Models of Consumer Demand
Date Posted:Tue, 07 Aug 2018 02:56:43 -0500
A long literature has developed econometric methods for estimating individual-consumer-level demand systems that accommodate corner solutions. The increasing access to transaction-level customer purchase histories across a wide array of markets and industries vastly expands the prospect for improved customer insight, more targeted marketing policies and individualized welfare analysis. A descriptive analysis of a broad, CPG database indicates that most consumer brand categories offer a wide variety of differentiated offerings for consumers. However, consumers typically purchase only a limited scope of the available variety, leading to a very high incidence of corner solutions which poses computational challenges for demand modeling. Historically, these computational challenges have limited the applicability of microeconometric models of demand in practice, except for the special case of pure discrete choice (e.g., logit and probit). Recent advances in computing power along with ...

REVISION: The Geography of Poverty and Nutrition: Food Deserts and Food Choices Across the United States
Date Posted:Mon, 15 Jan 2018 00:54:39 -0600
We study the causes of “nutritional inequality”: why the wealthy tend to eat more healthfully than the poor in the U.S. Using two event study designs exploiting entry of new supermarkets and households' moves to healthier neighborhoods, we reject that neighborhood environments have economically meaningful effects on healthy eating. Using a structural demand model, we find that exposing low-income households to the same food availability and prices experienced by high-income households would reduce nutritional inequality by only 9%, while the remaining 91% is driven by differences in demand. In turn, these income-related demand differences are partially explained by education, nutrition knowledge, and regional preferences. These findings contrast with discussions of nutritional inequality that emphasize supply-side issues such as food deserts.

REVISION: The Geography of Poverty and Nutrition: Food Deserts and Food Choices Across the United States
Date Posted:Fri, 05 Jan 2018 08:26:56 -0600
We study the causes of “nutritional inequality”: why the wealthy tend to eat more healthfully than the poor in the U.S. Using two event study designs exploiting entry of new supermarkets and households' moves to healthier neighborhoods, we reject that neighborhood environments have economically meaningful effects on healthy eating. Using a structural demand model, we find that exposing low-income households to the same food availability and prices experienced by high-income households would reduce nutritional inequality by only 9%, while the remaining 91% is driven by differences in demand. In turn, these income-related demand differences are partially explained by education, nutrition knowledge, and regional preferences. These findings contrast with discussions of nutritional inequality that emphasize supply-side issues such as food deserts.

REVISION: Scalable Price Targeting
Date Posted:Sun, 27 Aug 2017 13:44:36 -0500
Abstract We study the welfare implications of scalable price targeting, an extreme form of third-degree price discrimination implemented with machine learning for a large, digital firm. Targeted prices are computed by solving the firm's Bayesian Decision-Theoretic pricing problem based on a database with a high-dimensional vector of customer features that are observed prior to the price quote. To identify the causal effect of price on demand, we first run a large, randomized price experiment and use these data to train our demand model. We use l1 regularization (lasso) to select the set of customer features that moderate the heterogeneous treatment effect of price on demand. We use a weighted likelihood Bayesian bootstrap to quantify the firm's approximate statistical uncertainty in demand and profitability. We then conduct a second experiment that implements our proposed price targeting scheme out of sample. Theoretically, both firm and customer surplus could rise with scalable ...

REVISION: Scalable Price Targeting
Date Posted:Tue, 27 Jun 2017 05:57:45 -0500
We study the welfare implications of scalable price targeting, an extreme form of third-degree price discrimination, for a large, digital firm. Targeted prices are computed by solving the firm’s Bayesian Decision-Theoretic pricing problem based on a database with a high-dimensional vector of customer features that are observed prior to the price quote. To identify the causal effect of price on demand, we first run a large, randomized price experiment. These data are used to train our demand model. We use lasso regularization to select the set of customer features that moderate the heterogeneous treatment effect of price on demand. We use a weighted likelihood Bayesian bootstrap to quantify the firm’s approximate statistical uncertainty in demand and profitability. Theoretically, both firm and customer surplus could rise with scalable price targeting. We test the welfare implications out of sample with a second randomized price experiment with new customers. Optimized uniform pricing ...

REVISION: Scalable Price Targeting
Date Posted:Mon, 26 Jun 2017 07:40:51 -0500
We study the welfare implications of scalable price targeting, an extreme form of third-degree price discrimination, for a large, digital firm. Targeted prices are computed by solving the firm’s Bayesian Decision-Theoretic pricing problem based on a database with a high-dimensional vector of customer features that are observed prior to the price quote. To identify the causal effect of price on demand, we first run a large, randomized price experiment. These data are used to train our demand model. We use lasso regularization to select the set of customer features that moderate the heterogeneous treatment effect of price on demand. We use a weighted likelihood Bayesian bootstrap to quantify the firm’s approximate statistical uncertainty in demand and profitability. Theoretically, both firm and customer surplus could rise with scalable price targeting. We test the welfare implications out of sample with a second randomized price experiment with new customers. Optimized uniform pricing ...

REVISION: Self-Signaling and Prosocial Behavior: A Cause Marketing Experiment
Date Posted:Mon, 19 Jun 2017 21:23:04 -0500
We test an information theory of prosocial behavior whereby ego utility and self-signaling crowd out the effect of consumption utility on choice. The data come from two field experiments involving purchases of a consumer good bundled with a charitable donation. Across experimental cells, we randomize the price level and the donation level. A model-free analysis of the data reveals non-monotonic regions of demand when the good is bundled with relatively large charitable donations. Subjects also self-report lower ratings of “feeling good about themselves” when offered bundles with large donations and price discounts. The evidence suggests that price discounts crowd out consumer self-inference of altruism. Alternative motivation-crowding theories are rejected due to their inability to explain the non-monotonic data moments. The standard use of interaction effects and other falsification checks to explore the underlying choice mechanism in an experimental setting is complicated in our ...

Update: Do Pharmacists Buy Bayer? Sophisticated Shoppers and the Brand Premium
Date Posted:Sun, 16 Oct 2016 03:21:15 -0500
We estimate the effect of information on consumers’ willingness to pay for branded goods in physically homogeneous consumer packaged goods categories. In a case study of headache remedies, we find that college education, working in a healthcare occupation, and other proxies for product knowledge predict more purchases of private labels relative to brands. Pharmacists devote almost 90 percent of headache remedy purchases to private labels, against 71 percent for the average consumer. The effect of knowledge is similar across a broad set of health products, and in a set of relatively homogeneous food products, but smaller for food and drink products overall. We conclude that a significant share of the willingness to pay for brands in these categories would disappear in a world where consumers were fully informed.
New PDF Uploaded

New: The Formation of Consumer Brand Preferences
Date Posted:Thu, 22 Sep 2016 11:36:25 -0500
Brands and brand capital have long been theorized to play an important role in the formation of the industrial market structure of consumer goods industries. We summarize several striking empirical regularities in the concentration, magnitude and persistence of brand market shares in consumer goods categories. We then survey the theoretical and empirical literatures on the formation of brand preferences and how brand preferences contribute to our understanding of these empirical regularities. We also review the literature on how brand capital creates strategic advantages to firms that own established brands.

Update: Income and Wealth Effects on Private-Label Demand: Evidence from the Great Recession
Date Posted:Sun, 28 Aug 2016 13:35:21 -0500
We measure the causal effects of income and wealth on the demand for private-label products. Prior research suggests that these effects are large and, in particular, that private-label demand rises during recessions. Our empirical analysis is based on a comprehensive household-level transactions database matched with price information from store-level scanner data and wealth data based on local house value indices. The Great Recession provides a key source of the variation in our data, with a large and geographically diverse impact on household incomes over time. We estimate income and wealth effects using “within” variation of income and wealth at the household level. Our estimates can be interpreted as income and wealth effects consistent with a consumer demand model based on utility maximization. We establish a precisely measured negative effect of income on private-label shares. The effect of wealth is negative but not precisely measured. However, the estimated effect sizes are ...
New PDF Uploaded

Update: Do Pharmacists Buy Bayer? Sophisticated Shoppers and the Brand Premium
Date Posted:Tue, 23 Aug 2016 14:30:39 -0500
We estimate the effect of information on consumers’ willingness to pay for branded goods in physically homogeneous consumer packaged goods categories. In a case study of headache remedies, we find that college education, working in a healthcare occupation, and other proxies for product knowledge predict more purchases of private labels relative to brands. Pharmacists devote almost 90 percent of headache remedy purchases to private labels, against 71 percent for the average consumer. The effect of knowledge is similar across a broad set of health products, and in a set of relatively homogeneous food products, but smaller for food and drink products overall. We conclude that a significant share of the willingness to pay for brands in these categories would disappear in a world where consumers were fully informed.
New PDF Uploaded

Update: Brand History, Geography, and the Persistence of Brand Shares
Date Posted:Sun, 21 Aug 2016 03:38:55 -0500
We study persistence in the geographic variation in market shares of branded goods in consumer packaged goods industries across 50 U.S. city-markets. We match scanner data on local market shares and survey data on local quality perceptions for the largest brands in 34 consumer packaged goods industries. These data are then matched with historic information on the year and US city-market in which each brand was first launched. We find that these consumer brands have persistently higher market shares in markets closest to their respective cities-of-origin than in markets farthest from their respective cities-of-origin, where they were typically launched later. For 6 of the 34 industries, we collected more complete historic entry data with which we can determine the local order of entry among the top brands in each of the 50 U.S. city-markets. We find a persistent effect from differences in the order-of-entry of competing brands on their current relative brand shares and quality ...
New PDF Uploaded

REVISION: Income and Wealth Effects on Private-Label Demand: Evidence from the Great Recession
Date Posted:Fri, 12 Aug 2016 09:20:43 -0500
We measure the causal effects of income and wealth on the demand for private-label products. Prior research suggests that these effects are large and, in particular, that private-label demand rises during recessions. Our empirical analysis is based on a comprehensive household-level transactions database matched with price information from store-level scanner data and wealth data based on local house value indices. The Great Recession provides a key source of the variation in our data, with a large and geographically diverse impact on household incomes over time. We estimate income and wealth effects using “within” variation of income and wealth at the household level. Our estimates can be interpreted as income and wealth effects consistent with a consumer demand model based on utility maximization. We establish a precisely measured negative effect of income on private-label shares. The effect of wealth is negative but not precisely measured. However, the estimated effect sizes are ...

Update: Do Switching Costs Make Markets Less Competitive?
Date Posted:Wed, 10 Aug 2016 15:10:10 -0500
The conventional wisdom in economic theory holds that switching costs make markets less competitive. This paper challenges this claim. We find that steady-state equilibrium prices may fall as switching costs are introduced into a simple model of dynamic price competition that allows for differentiated products and imperfect lock-in. To assess whether this finding is of empirical relevance, we consider a more general model with heterogeneous consumers. We calibrate this model with data from a frequently purchased packaged goods market where consumers exhibit inertia in their brand choices, a behavior consistent with switching costs. We estimate the level of switching costs from the brand choice behavior in this data. At switching costs of the order of magnitude found in our data, prices are lower than without switching costs.
New PDF Uploaded

REVISION: Competitive Price Targeting with Smartphone Coupons
Date Posted:Fri, 05 Aug 2016 03:33:08 -0500
To investigate the potential returns of targeted mobile discounts, we design and implement a large-scale field experiment involving two competing movie theaters. In the experiment, we test different forms of targeting based on real-time consumer location and past consumer behavior. A novel feature of our experiment is that we test a range of relative ticket prices from both firms to assess their targeting incentives and the moderating effects of competitive targeting. The experiment reveals substantial profit gains from mobile discounts during an off-peak period of the day. In particular, both firms could create incremental profits by targeting their competitor's location. However, the returns to such “geo-conquesting” are reduced when the competitor also launches its own targeting campaign. To study optimal targeting incentives, we combine our experiment with a model of demand that can be used to predict the consumer choices at price levels that were not included on the ...

REVISION: Self-Signaling and Prosocial Behavior: A Cause Marketing Experiment
Date Posted:Tue, 08 Mar 2016 01:18:38 -0600
We test an information theory of prosocial behavior whereby ego utility and self-signaling crowd out the effect of consumption utility on choice. The data come from two field experiments involving purchases of a consumer good bundled with a charitable donation. Across experimental cells, we randomize the price level and the donation level. A model-free analysis of the data reveals non-monotonic regions of demand when the good is bundled with relatively large charitable donations. Subjects also self-report lower ratings of “feeling good about themselves” when offered bundles with large donations and price discounts. The evidence suggests that price discounts crowd out consumer self-inference of altruism. Alternative motivation-crowding theories are rejected due to their inability to explain the non-monotonic data moments. The standard use of interaction effects and other falsification checks to explore the underlying choice mechanism in an experimental setting is complicated in our ...

REVISION: Competitive Price Targeting with Smartphone Coupons
Date Posted:Tue, 23 Feb 2016 02:25:51 -0600
We conduct a large-scale field experiment to study competitive price discrimination in a duopoly market with two rival movie theaters. The firms use mobile targeting to offer different prices based on location and past consumer activity. A novel feature of our experiment is that we test a range of relative ticket prices from both firms to trace out their respective best-response functions and to assess equilibrium outcomes. We find an empirically large return on investment when a single firm unilaterally targets its prices based on the geographic location or historical visit behavior of a mobile customer. However, these returns can be mitigated by competitive interactions whereby both firms simultaneously engage in targeting. In practice, firms typically test only their own prices and do not consider the competitive response of a rival. In our study of movie theaters, competition enhances the returns to behavioral targeting but reduces the returns to geo-targeting. Thus, managers ...

REVISION: Self-Signaling and Prosocial Behavior: A Cause Marketing Experiment
Date Posted:Tue, 23 Feb 2016 02:22:57 -0600
We test an information theory of prosocial behavior whereby ego utility and self-signaling crowd out the effect of consumption utility on choice. The data come from two field experiments involving purchases of a consumer good bundled with a charitable donation. Across experimental cells, we randomize the price level and the donation level. A model-free analysis of the data reveals non-monotonic regions of demand when the good is bundled with relatively large charitable donations. Subjects also self-report lower ratings of “feeling good about themselves” when offered bundles with large donations and price discounts. The evidence suggests that price discounts crowd out consumer self-inference of altruism. Alternative motivation-crowding theories are rejected due to their inability to explain the non-monotonic data moments. The standard use of interaction effects and other falsification checks to explore the underlying choice mechanism in an experimental setting is complicated in our ...

REVISION: Competitive Price Targeting with Smartphone Coupons
Date Posted:Fri, 15 Jan 2016 01:51:08 -0600
We conduct a large-scale field experiment to study competitive price discrimination in a duopoly market with two rival movie theaters. The firms use mobile targeting to offer different prices based on location and past consumer activity. A novel feature of our experiment is that we test a range of relative ticket prices from both firms to trace out their respective best-response functions and to assess equilibrium outcomes. We find an empirically large return on investment when a single firm unilaterally targets its prices based on the geographic location or historical visit behavior of a mobile customer. However, these returns can be mitigated by competitive interactions whereby both firms simultaneously engage in targeting. In practice, firms typically test only their own prices and do not consider the competitive response of a rival. In our study of movie theaters, competition enhances the returns to behavioral targeting but reduces the returns to geo-targeting. Thus, managers ...

REVISION: Competitive Price Targeting with Smartphone Coupons
Date Posted:Mon, 11 Jan 2016 10:25:24 -0600
We conduct a large-scale field experiment to study competitive price discrimination in a duopoly market with two rival movie theaters. The firms use mobile targeting to offer different prices based on location and past consumer activity. A novel feature of our experiment is that we test a range of relative ticket prices from both firms to trace out their respective best-response functions and to assess equilibrium outcomes. We find an empirically large return on investment when a single firm unilaterally targets its prices based on the geographic location or historical visit behavior of a mobile customer. However, these returns can be mitigated by competitive interactions whereby both firms simultaneously engage in targeting. In practice, firms typically test only their own prices and do not consider the competitive response of a rival. In our study of movie theaters, competition enhances the returns to behavioral targeting but reduces the returns to geo-targeting. Thus, managers ...

REVISION: Competitive Price Targeting with Smartphone Coupons
Date Posted:Tue, 15 Dec 2015 05:15:56 -0600
We conduct a large-scale field experiment to study competitive price discrimination in a duopoly market with two rival movie theaters. The firms use mobile targeting to offer different prices based on location and past consumer activity. A novel feature of our experiment is that we test a range of relative ticket prices from both firms to trace out their respective best-response functions and to assess equilibrium outcomes. We find an empirically large return on investment when a single firm unilaterally targets its prices based on the geographic location or historical visit behavior of a mobile customer. However, these returns can be mitigated by competitive interactions whereby both firms simultaneously engage in targeting. In practice, firms typically test only their own prices and do not consider the competitive response of a rival. In our study of movie theaters, competition enhances the returns to behavioral targeting but reduces the returns to geo-targeting. Thus, managers ...

REVISION: Self-Signaling and Prosocial Behavior: A Cause Marketing Experiment
Date Posted:Mon, 14 Dec 2015 03:04:56 -0600
We test an information theory of prosocial behavior whereby ego utility and self-signaling crowd out the effect of consumption utility on choice. The data come from two field experiments involving a consumer good bundled with a charitable donation. The evidence consists of non-monotonic regions of demand for the bundle combined with lower self-reported consumer ratings of “feeling good about themselves” when offered bundles with large donations and price discounts. Price discounts crowd out consumer self-inference of altruism. Alternative motivation-crowding theories do not fit the non-monotonic data moments. Results from a structural model of self-signaling quantify the economic role of ego utility. These findings contribute to a growing literature on self-signaling and consumer behavior by quantifying the magnitude of self-signaling on preferences and choices. The results also have implications for the design of a cause marketing campaign and the potential negative synergies ...

REVISION: Competitive Price Targeting with Smartphone Coupons
Date Posted:Mon, 14 Dec 2015 02:48:51 -0600
We conduct a large-scale field experiment to study competitive price discrimination in a duopoly market with two rival movie theaters. The firms use mobile targeting to offer different prices based on location and past consumer activity. A novel feature of our experiment is that we test a range of relative ticket prices from both firms to trace out their respective best-response functions and to assess equilibrium outcomes. We find an empirically large return on investment when a single firm unilaterally targets its prices based on the geographic location or historical visit behavior of a mobile customer. However, these returns can be mitigated by competitive interactions whereby both firms simultaneously engage in targeting. In practice, firms typically test only their own prices and do not consider the competitive response of a rival. In our study of movie theaters, competition enhances the returns to behavioral targeting but reduces the returns to geo-targeting. Thus, managers ...

REVISION: Competitive Price Targeting with Smartphone Coupons
Date Posted:Wed, 02 Dec 2015 12:21:14 -0600
We conduct a large-scale field experiment to study competitive price discrimination in a duopoly market with two rival movie theaters. The firms use mobile targeting to offer different prices based on location and past consumer activity. A novel feature of our experiment is that we test a range of relative ticket prices from both firms to trace out their respective best-response functions and to assess equilibrium outcomes. We find an empirically large return on investment when a single firm unilaterally targets its prices based on the geographic location or historical visit behavior of a mobile customer. However, these returns can be mitigated by competitive interactions whereby both firms simultaneously engage in targeting. In practice, firms typically test only their own prices and do not consider the competitive response of a rival. In our study of movie theaters, competition enhances the returns to behavioral targeting but reduces the returns to geo-targeting. Thus, managers ...

REVISION: Competitive Price Targeting
Date Posted:Mon, 23 Nov 2015 20:49:42 -0600
We conduct a large-scale field experiment to study competitive price discrimination in a duopoly market with two rival movie theaters. The firms use mobile targeting to offer different prices based on location and past consumer activity. A novel feature of our experiment is that we test a range of relative ticket prices from both firms to trace out their respective best-response functions and to assess equilibrium outcomes. We find an empirically large return on investment when a single firm unilaterally targets its prices based on the geographic location or historical visit behavior of a mobile customer. However, these returns can be mitigated by competitive interactions whereby both firms simultaneously engage in targeting. In practice, firms typically test only their own prices and do not consider the competitive response of a rival. In our study of movie theaters, competition enhances the returns to behavioral targeting but reduces the returns to geo-targeting. Thus, managers ...

REVISION: Self-Signaling and Pro-Social Behavior: A Cause Marketing Experiment
Date Posted:Tue, 10 Nov 2015 07:03:10 -0600
We test a self-signaling theory using two large-scale, randomized controlled field experiments. Smartphone users are randomly sampled to receive promotional offers for movie tickets via SMS technology. Subjects are exposed to different pre-determined levels of price discounts and charitable donations tied to the ticket purchase. The main effects of price discounts and charitable donations increase ticket demand. However, the combination of both discounts and donations can decrease ticket demand. In a post-purchase survey, the same subjects self-report lower ratings of “feeling good about themselves” as the motivation for buying a ticket when discounts and donations are both large. These findings are consistent with a self-signaling theory, whereby the discount crowds out the consumer's “warm-glow” feeling from the charitable donation. Alternative behavioral explanations are ruled out. A structural model of demand with self-signaling is fit to the data using a constrained optimization ...

REVISION: Self-Signaling and Pro-Social Behavior: A Cause Marketing Mobile Field Experiment
Date Posted:Mon, 02 Nov 2015 03:02:58 -0600
We test a self-signaling theory using two large-scale, randomized controlled field experiments. Smartphone users are randomly sampled to receive promotional offers for movie tickets via SMS technology. Subjects are exposed to different pre-determined levels of price discounts and charitable donations tied to the ticket purchase. The main effects of price discounts and charitable donations increase ticket demand. However, the combination of both discounts and donations can decrease ticket demand. In a post-purchase survey, the same subjects self-report lower ratings of “feeling good about themselves” as the motivation for buying a ticket when discounts and donations are both large. These findings are consistent with a self-signaling theory, whereby the discount crowds out the consumer's “warm-glow” feeling from the charitable donation. Alternative behavioral explanations are ruled out. A structural model of demand with self-signaling is fit to the data using a constrained optimization ...

REVISION: Income and Wealth Effects on Private-Label Demand: Evidence from the Great Recession
Date Posted:Sun, 11 Oct 2015 08:28:29 -0500
We measure the causal effects of income and wealth on the demand for private-label products. Prior research suggests that these effects are large and, in particular, that private-label demand rises during recessions. Our empirical analysis is based on a comprehensive household-level transactions database matched with price information from store-level scanner data and wealth data based on local house value indices. The Great Recession provides a key source of the variation in our data, with a large and geographically diverse impact on household incomes over time. We estimate income and wealth effects using “within” variation of income and wealth at the household level. Our estimates can be interpreted as income and wealth effects consistent with a consumer demand model based on utility maximization. We establish a precisely measured negative effect of income on private-label shares. The effect of wealth is negative but not precisely measured. However, the estimated effect sizes are ...

REVISION: Self-Signaling and Pro-Social Behavior: A Cause Marketing Mobile Field Experiment
Date Posted:Mon, 27 Jul 2015 11:17:16 -0500
We test a self-signaling theory using two large-scale, randomized controlled field experiments. Smartphone users are randomly sampled to receive promotional offers for movie tickets via SMS technology. Subjects are exposed to different pre-determined levels of price discounts and charitable donations tied to the ticket purchase. The main effects of price discounts and charitable donations increase ticket demand. However, the combination of both discounts and donations can decrease ticket demand. In a post-purchase survey, the same subjects self-report lower ratings of “feeling good about themselves” as the motivation for buying a ticket when discounts and donations are both large. These findings are consistent with a self-signaling theory, whereby the discount crowds out the consumer's “warm-glow” feeling from the charitable donation. Alternative behavioral explanations are ruled out. A structural model of demand with self-signaling is fit to the data using a constrained optimization ...

REVISION: Self-Signaling and Pro-Social Behavior: A Cause Marketing Mobile Field Experiment
Date Posted:Sun, 26 Jul 2015 13:28:25 -0500
We test a self-signaling theory using two large-scale, randomized controlled field experiments. Smartphone users are randomly sampled to receive promotional offers for movie tickets via SMS technology. Subjects are exposed to different pre-determined levels of price discounts and charitable donations tied to the ticket purchase. The main effects of price discounts and charitable donations increase ticket demand. However, the combination of both discounts and donations can decrease ticket demand. In a post-purchase survey, the same subjects self-report lower ratings of “feeling good about themselves” as the motivation for buying a ticket when discounts and donations are both large. These findings are consistent with a self-signaling theory, whereby the discount crowds out the consumer's “warm-glow” feeling from the charitable donation. Alternative behavioral explanations are ruled out. A structural model of demand with self-signaling is fit to the data using a constrained optimization ...

New: Income and Wealth Effects on Private-Label Demand: Evidence from the Great Recession
Date Posted:Fri, 17 Jul 2015 07:58:12 -0500
We measure the causal effects of income and wealth on the demand for private-label products. Prior research suggests that these effects are large and, in particular, that private-label demand rises during recessions. Our empirical analysis is based on a comprehensive household-level transactions database matched with price information from store-level scanner data and wealth data based on local house value indices. The Great Recession provides a key source of the variation in our data, with a large and geographically diverse impact on household incomes over time. We estimate income and wealth effects using “within” variation of income and wealth at the household level. Our estimates can be interpreted as income and wealth effects consistent with a consumer demand model based on utility maximization. We establish a precisely measured negative effect of income on private-label shares. The effect of wealth is negative but not precisely measured. However, the estimated effect sizes are ...

REVISION: The Joint Identification of Utility and Discount Functions from Stated Choice Data: An Application to Durable Goods Adoption
Date Posted:Sat, 26 Jul 2014 02:44:19 -0500
We present a survey design that generalizes static conjoint experiments to elicit inter-temporal adoption decisions for durable goods. We show that consumers’ utility and discount functions in a dynamic discrete choice model are jointly identified using data generated by this specific design. In contrast, based on revealed preference data, the utility and discount functions are generally not jointly identified even if consumers’ expectations are known. The separation of current-period preferences from discounting is necessary to forecast the diffusion of a durable good under alternative marketing strategies. We illustrate the approach using two surveys eliciting Blu-ray player adoption decisions. Both model-free evidence and the estimates based on a dynamic discrete choice model indicate that consumers make forward-looking adoption decisions. In both surveys the average discount rate is 43 percent, corresponding to a substantially higher degree of impatience than the rate implied by ...

New: Do Pharmacists Buy Bayer? Sophisticated Shoppers and the Brand Premium
Date Posted:Tue, 01 Jul 2014 10:29:04 -0500
We estimate the effect of information on consumers’ willingness to pay for branded goods in physically homogeneous consumer packaged goods categories. In a case study of headache remedies, we find that college education, working in a healthcare occupation, and other proxies for product knowledge predict more purchases of private labels relative to brands. Pharmacists devote almost 90 percent of headache remedy purchases to private labels, against 71 percent for the average consumer. The effect of knowledge is similar across a broad set of health products, and in a set of relatively homogeneous food products, but smaller for food and drink products overall. We conclude that a significant share of the willingness to pay for brands in these categories would disappear in a world where consumers were fully informed.

REVISION: The Joint Identification of Utility and Discount Functions from Stated Choice Data: An Application to Durable Goods Adoption
Date Posted:Mon, 14 Oct 2013 04:00:20 -0500
We present a survey design that generalizes static conjoint experiments to elicit inter-temporal adoption decisions for durable goods. We show that consumers’ utility and discount functions in a dynamic discrete choice model are jointly identified using data generated by this specific design. In contrast, based on revealed preference data, the utility and discount functions are generally not jointly identified even if consumers’ expectations are known. The separation of current-period preferences from discounting is necessary to forecast the diffusion of a durable good under alternative marketing strategies. We illustrate the approach using two surveys eliciting Blu-ray player adoption decisions. Both model-free evidence and the estimates based on a dynamic discrete choice model indicate that consumers make forward-looking adoption decisions. In both surveys the average discount rate is 43 percent, corresponding to a substantially higher degree of impatience than the rate implied by ...

REVISION: The Joint Identification of Utility and Discount Functions from Stated Choice Data: An Application t
Date Posted:Mon, 19 Nov 2012 17:59:45 -0600
We present a survey design that generalizes static conjoint experiments to elicit inter-temporal adoption decisions for durable goods. We show that consumers’ utility and discount functions in a dynamic discrete choice model are jointly identified using data generated by this specific design. In contrast, based on revealed preference data, the utility and discount functions are generally not jointly identified even if consumers’ expectations are known. The separation of current-period ...

REVISION: The Joint Identification of Utility and Discount Functions from Stated Choice Data: An Application t
Date Posted:Sat, 29 Sep 2012 01:47:40 -0500
We present a survey design that generalizes static conjoint experiments to elicit inter-temporal adoption decisions for durable goods. We show that consumers’ utility and discount functions in a dynamic discrete choice model are jointly identified using data generated by this specific design. In contrast, based on revealed preference data, the utility and discount functions are generally not jointly identified even if consumers’ expectations are known. The separation of current-period ...

REVISION: The Joint Identification of Utility and Discount Functions from Stated Choice Data: An Application t
Date Posted:Sun, 09 Sep 2012 17:20:41 -0500
We present a survey design that generalizes static conjoint experiments to elicit inter-temporal adoption decisions for durable goods. We show that consumers’ utility and discount functions in a dynamic discrete choice model are jointly identified using data generated by this specific design. In contrast, based on revealed preference data, the utility and discount functions are generally not jointly identified even if consumers’ expectations are known. The separation of current-period ...

REVISION: The Joint Identification of Utility and Discount Functions from Stated Choice Data: An Application t
Date Posted:Sun, 09 Sep 2012 05:15:24 -0500
We present a survey design that generalizes static conjoint experiments to elicit inter-temporal adoption decisions for durable goods. We show that consumers’ utility and discount functions in a dynamic discrete choice model are jointly identified using data generated by this specific design. In contrast, based on revealed preference data, the utility and discount functions are generally not jointly identified even if consumers’ expectations are known. The separation of current-period ...

REVISION: Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients De
Date Posted:Sat, 31 Dec 2011 00:24:19 -0600
The widely-used estimator of Berry, Levinsohn and Pakes (1995) produces estimates of consumer preferences from a discrete-choice demand model with random coefficients, market-level demand shocks and endogenous prices. We derive numerical theory results characterizing the properties of the nested fixed point algorithm used to evaluate the objective function of BLP's estimator. We discuss problems with typical implementations, including cases that can lead to incorrect parameter estimates. As a ...

REVISION: Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients De
Date Posted:Mon, 24 Oct 2011 17:33:24 -0500
The widely-used estimator of Berry, Levinsohn and Pakes (1995) produces estimates of consumer preferences from a discrete-choice demand model with random coefficients, market-level demand shocks and endogenous prices. We derive numerical theory results characterizing the properties of the nested fixed point algorithm used to evaluate the objective function of BLP's estimator. We discuss problems with typical implementations, including cases that can lead to incorrect parameter estimates. As a ...

REVISION: Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients De
Date Posted:Sun, 09 Oct 2011 08:24:25 -0500
The widely-used estimator of Berry, Levinsohn and Pakes (1995) produces estimates of consumer preferences from a discrete-choice demand model with random coefficients, market-level demand shocks and endogenous prices. We derive numerical theory results characterizing the properties of the nested fixed point algorithm used to evaluate the objective function of BLP's estimator. We discuss problems with typical implementations, including cases that can lead to incorrect parameter estimates. As a ...

REVISION: Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients De
Date Posted:Mon, 07 Mar 2011 09:53:06 -0600
The widely-used estimator of Berry, Levinsohn and Pakes (1995) produces estimates of consumer preferences from a discrete-choice demand model with random coefficients, market-level demand shocks and endogenous prices. We derive numerical theory results characterizing the properties of the nested fixed point algorithm used to evaluate the objective function of BLP's estimator. We discuss problems with typical implementations, including cases that can lead to incorrect parameter estimates. As a ...

REVISION: State Dependence and Alternative Explanations for Consumer Inertia
Date Posted:Wed, 14 Apr 2010 05:07:52 -0500
For many consumer packaged goods products, researchers have documented inertia in brand choice, a form of persistence whereby consumers have a higher probability of choosing a product that they have purchased in the past. Using data on margarine and refrigerated orange juice purchases, we show that the finding of inertia is robust to flexible controls for preference heterogeneity and not due to autocorrelated taste shocks. Thus, the inertia is at least partly due to structural, not spurious ...

REVISION: Do DVRs Influence Sales?
Date Posted:Fri, 29 Jan 2010 19:08:34 -0600
We analyze a multimillion dollar, three-year field study sponsored by five firms to assess whether DVRs impact consumers’ shopping behavior for advertised and private label goods. A large sample of households received an offer for a free DVR and service and close to 20% accepted. We observe each household's shopping history for 48 consumer packaged goods categories during the 13 months prior and the 26 months following the DVR offer. We fail to reject the null of no DVR treatment effect on ...

REVISION: Do DVRs Influence Sales?
Date Posted:Wed, 28 Oct 2009 12:04:53 -0500
We analyze a multimillion dollar, three-year field study sponsored by five firms to assess whether DVRs impact consumers’ shopping behavior for advertised and private label goods. A large sample of households received an offer for a free DVR and service and close to 20% accepted. We observe each household's shopping history for 48 consumer packaged goods categories during the 13 months prior and the 26 months following the DVR offer. We fail to reject the null of no DVR treatment effect on ...

New: Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients De...
Date Posted:Thu, 11 Jun 2009 23:42:16 -0500
The widely-used estimator of Berry, Levinsohn and Pakes (1995) produces estimates of consumer preferences from a discrete-choice demand model with random coefficients, market-level demand shocks and endogenous prices. We derive numerical theory results characterizing the properties of the nested fixed point algorithm used to evaluate the objective function of BLP's estimator. We discuss problems with typical implementations, including cases that can lead to incorrect parameter estimates. As a ...

REVISION: Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients De
Date Posted:Mon, 18 May 2009 09:45:15 -0500
The widely-used estimator of Berry, Levinsohn and Pakes (1995) produces estimates of consumer preferences from a discrete-choice demand model with random coefficients, market-level demand shocks and endogenous prices. We derive numerical theory results characterizing the properties of the nested fixed point algorithm used to evaluate the objective function of BLP's estimator. We discuss problems with typical implementations, including cases that can lead to incorrect parameter estimates. As a ...

New: State Dependence and Alternative Explanations for Consumer Inertia
Date Posted:Wed, 29 Apr 2009 03:40:44 -0500
For many consumer packaged goods products, researchers have documented a form of state dependence whereby consumers become loyal to products they have consumed in the past. That is, consumers behave as though there is a utility premium from continuing to purchase the same product as they have purchased in the past or, equivalently, there is a psychological cost to switching products. However, it has not been established that this form of state dependence can be identified in the presence of ...

REVISION: State Dependence and Alternative Explanations for Consumer Inertia
Date Posted:Wed, 25 Mar 2009 05:56:04 -0500
For many consumer packaged goods products, researchers have documented a form of state dependence in which consumers become "loyal" to products they have consumed in the past. That is, consumers behave as though there is a utility premium from continuing to purchase the same product as they have purchased in the past or, equivalently, there is a psychological cost to switching products. However, it as not been established that this form of state dependence can be identified in the presense of ...

REVISION: State Dependence and Alternative Explanations for Consumer Inertia
Date Posted:Wed, 18 Mar 2009 17:15:04 -0500
For many consumer packaged goods products, researchers have documented a form of state dependence in which consumers become "loyal" to products they have consumed in the past. That is, consumers behave as though there is a utility premium from continuing to purchase the same product as they have purchased in the past or, equivalently, there is a psychological cost to switching products. However, it as not been established that this form of state dependence can be identified in the presense of ...

REVISION: Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients De
Date Posted:Wed, 25 Feb 2009 18:17:49 -0600
The widely-used estimator of Berry, Levinsohn and Pakes (1995) produces consistent, instrumental variables estimates of consumer preferences from a discrete-choice demand model with random coefficients, market-level demand shocks and endogenous regressors (prices). The nested-fixed point algorithm typically used for estimation is computationally intensive, largely because a system of market-share equations must be repeatedly numerically inverted. We provide numerical theory results that ...

New: Brand History, Geography, and the Persistence of Brand Shares
Date Posted:Wed, 11 Feb 2009 19:49:35 -0600
We study persistence in the geographic variation in market shares of branded goods in consumer packaged goods industries across 50 U.S. city-markets. We match scanner data on local market shares and survey data on local quality perceptions for the largest brands in 34 consumer packaged goods industries. These data are then matched with historic information on the year and US city-market in which each brand was first launched. We find that these consumer brands have persistently higher market ...

REVISION: Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients De
Date Posted:Thu, 05 Feb 2009 09:24:08 -0600
The widely-used estimator of Berry, Levinsohn and Pakes (1995) produces consistent, instrumental variables estimates of consumer preferences from a discrete-choice demand model with random coefficients, market-level demand shocks and endogenous regressors (prices). The nested-fixed point algorithm typically used for estimation is computationally intensive, largely because a system of market-share equations must be repeatedly numerically inverted. We provide numerical theory results that ...

New: Do Switching Costs Make Markets Less Competitive?
Date Posted:Mon, 19 Jan 2009 17:19:13 -0600
The conventional wisdom in economic theory holds that switching costs make markets less competitive. This paper challenges this claim. We find that steady-state equilibrium prices may fall as switching costs are introduced into a simple model of dynamic price competition that allows for differentiated products and imperfect lock-in. To assess whether this finding is of empirical relevance, we consider a more general model with heterogeneous consumers. We calibrate this model with data from ...

New: Brand History, Geography, and the Persistence of Brand Shares
Date Posted:Sun, 23 Nov 2008 17:23:32 -0600
We study persistence in the geographic variation in market shares of branded goods in consumer packaged goods industries across 50 U.S. city-markets. We match scanner data on local market shares and survey data on local quality perceptions for the largest brands in 34 consumer packaged goods industries. These data are then matched with historic information on the year and US city-market in which each brand was first launched. We find that these consumer brands have persistently higher market ...

New: Brand History, Geography, and the Persistence of Brand Shares
Date Posted:Mon, 29 Sep 2008 08:13:39 -0500
We study persistence in the geographic variation in market shares of branded goods in consumer packaged goods industries across 50 U.S. city-markets. We match scanner data on local market shares and survey data on local quality perceptions for the largest brands in 34 consumer packaged goods industries. These data are then matched with historic information on the year and US city-market in which each brand was first launched. We find that these consumer brands have persistently higher market ...

REVISION: Accounting for Primary and Secondary Demand Effects with Aggregate Data
Date Posted:Wed, 11 Jun 2008 10:38:41 -0500
Discrete choice models of aggregate demand, such as the random coefficients logit, can handle large differentiated products categories parsimoniously while still providing flexible substitution patterns. However, the discrete choice assumption may not be appropriate for many categories in which we expect consumers may purchase more than one unit of the selected item. We derive the aggregate demand system corresponding to a discrete/continuous household-level model of demand. We also propose a ...

New: Prominence Effect in Shanghai Apartment Prices
Date Posted:Fri, 23 May 2008 06:59:11 -0500
A field study conducted in Shanghai identified a robust inconsistency between real estate developers' desired sales pattern (selling all apartments in a building at similar rates) and the actual sales pattern (selling good apartments faster). The authors explained this inconsistency with Tversky, Sattath, and Slovic (1988)'s prominence principle, according to which buyers, who were in a choice mode, weighed the desirability of floors more heavily than developers, who were in a matching mode ...

Estimating an SKU-level Brand Choice Model Combining Household Panel Data and Store Data
Date Posted:Tue, 22 Apr 2008 09:04:30 -0500
The extant literature using household scanner data to estimate consumer choice models has identified two key sources of bias in estimated mean responses to marketing variables. Omitted heterogeneity may bias mean responses towards zero. At the same time, omitted time-varying characteristics of alternatives that influence consumer choices may also bias mean responses towards zero if these characteristics are correlated with observed factors such as price - the endogeneity bias. Both these ...

Product Differentiation and Mergers in the Carbonated Soft Drink Industry
Date Posted:Tue, 22 Apr 2008 07:28:28 -0500
I simulate the competitive impact of several soft drink mergers from the 1980s on equilibrium prices and quantities. An unusual feature of soft drink demand is that, at the individual purchase level, households regularly select a variety of soft drink products. Specifically, on a given trip households may select multiple soft drink products and multiple units of each. A concern is that using a standard discrete choice model that assumes single unit purchases may understate the price elasticity ...

Empirical Analysis of Indirect Network Effects in the Market for Personal Digital Assistants
Date Posted:Tue, 22 Apr 2008 07:13:49 -0500
We present a framework to measure empirically the size of indirect network effects in high-technology markets with competing incompatible technology standards. These indirect network effects arise due to inter-dependence in demand for hardware and compatible software. By modeling the joint determination of hardware sales and software availability in the market, we are able to describe the nature of demand inter-dependence and to measure the size of the indirect network effects. We apply the ...

New: Do Switching Costs Make Markets Less Competitive?
Date Posted:Tue, 01 Apr 2008 09:45:56 -0500
The conventional wisdom in economic theory holds that switching costs make markets less competitive. This paper challenges this claim. We find that steady-state equilibrium prices may fall as switching costs are introduced into a simple model of dynamic price competition that allows for differentiated products and imperfect lock-in. To assess whether this finding is of empirical relevance, we consider a more general model with heterogeneous consumers. We calibrate this model with data from ...

New: Cross-Brand Pass-Through in Supermarket Pricing
Date Posted:Wed, 30 Jan 2008 01:29:31 -0600
We investigate the sensitivity of cross-brand pass-through estimates to two types of pooling: across stores, and across regular price and promotional price weeks. Using the category data from Besanko, Dubé and Gupta (2005), hereafter BDG, we find consistent support across all 11 categories for the predictive power of the wholesale prices of substitute products for retail shelf prices. A Bayesian procedure is used to address the small sample issues that arise in the absence of pooling. Even ...

REVISION: Tipping and Concentration in Markets with Indirect Network Effects
Date Posted:Thu, 24 Jan 2008 18:35:19 -0600
This paper develops a framework to measure 'tipping' - the increase in a firm's market share dominance caused by indirect network effects. Our measure compares the expected concentration in a market to the hypothetical expected concentration that would arise in the absence of indirect network effects. In practice, this measure requires a model that can predict the counter-factual market concentration under different parameter values capturing the strength of indirect network effects. We build ...

REVISION: Tipping and Concentration in Markets with Indirect Network Effects
Date Posted:Mon, 21 Jan 2008 13:30:20 -0600
This paper develops a framework to measure 'tipping' - the increase in a firm's market share dominance caused by indirect network effects. Our measure compares the expected concentration in a market to the hypothetical expected concentration that would arise in the absence of indirect network effects. In practice, this measure requires a model that can predict the counter-factual market concentration under different parameter values capturing the strength of indirect network effects. We build ...

New: Do Switching Costs Make Markets Less Competitive?
Date Posted:Sun, 12 Aug 2007 18:30:07 -0500
The conventional wisdom in economic theory holds that switching costs make markets less competitive. This paper challenges this claim. We find that steady-state equilibrium prices may fall as switching costs are introduced into a simple model of dynamic price competition that allows for differentiated products and imperfect lock-in. To assess whether this finding is of empirical relevance, we consider a more general model with heterogeneous consumers. We calibrate this model with data from ...

REVISION: Category Pricing with State Dependent Utility
Date Posted:Sat, 09 Dec 2006 17:46:25 -0600
There is a substantial literature that documents the presence of state dependent utility with packaged goods data. Typically, a form of brand loyalty is detected whereby there is a higher probability of purchasing the same brand as has been purchased in the recent past. The economic significance of the measured loyalty remains an open question. We consider the category pricing problem in the presence of loyalty and demonstrate that a retailer has an incentive to invest in building brand ...

REVISION: Accounting for Primary and Secondary Demand Effects with Aggregate Data
Date Posted:Fri, 17 Nov 2006 08:13:27 -0600
Discrete choice models of aggregate demand, such as the random coefficients logit, can handle large differentiated products categories parsimoniously while still providing flexible substitution patterns. However, the discrete choice assumption may not be appropriate for many categories in which we expect consumers may purchase more than one unit of the selected item. We derive the aggregate demand system corresponding to a discrete/continuous household-level model of demand. We also propose a ...

New: Do Switching Costs Make Markets Less Competitive?
Date Posted:Wed, 27 Sep 2006 08:45:20 -0500
The conventional wisdom in economic theory holds that switching costs make markets less competitive. This paper challenges this claim. We find that steady-state equilibrium prices may fall as switching costs are introduced into a simple model of dynamic price competition that allows for differentiated products and imperfect lock-in. To assess whether this finding is of empirical relevance, we consider a more general model with heterogeneous consumers. We calibrate this model with data from ...

New: A Behavioral Analysis of Shanghai Real Estate Prices
Date Posted:Tue, 29 Aug 2006 10:17:49 -0500
In a field study, we identified an intriguing inconsistency between real estate developers' desired sales pattern (selling all apartments at the same rate) and the actual sales pattern (selling good apartments faster). We explained this inconsistency with Tversky, Sattath and Slovic (1988)'s prominence principle, according to which buyers, who were in a choice mode, weighed the desirability of floors more heavily than developers, who were in a matching mode when setting prices. We corroborated ...

New: Do Switching Costs Make Markets Less Competitive?
Date Posted:Tue, 13 Jun 2006 16:22:17 -0500
The conventional wisdom in economic theory holds that switching costs make markets less competitive. This paper challenges this claim. We find that steady-state equilibrium prices may fall as switching costs are introduced into a simple model of dynamic price competition that allows for differentiated products and imperfect lock-in. To assess whether this finding is of empirical relevance, we consider a more general model with heterogeneous consumers. We calibrate this model with data from ...

REVISION: Category Pricing with State Dependent Utility
Date Posted:Tue, 06 Jun 2006 11:27:39 -0500
There is a substantial literature that documents the presence of state dependent utility with packaged goods data. Typically, a form of brand loyalty is detected whereby there is a higher probability of purchasing the same brand as has been purchased in the recent past. The economic significance of the measured loyalty remains an open question. We consider the category pricing problem in the presence of loyalty and demonstrate that a retailer has an incentive to invest in building brand ...

REVISION: Category Pricing with State Dependent Utility
Date Posted:Fri, 07 Apr 2006 11:37:56 -0500
There is a substantial literature that documents the presence of state dependent utility with packaged goods data. Typically, a form of brand loyalty is detected whereby there is a higher probability of purchasing the same brand as has been purchased in the recent past. The economic significance of the measured loyalty remains an open question. We consider the category pricing problem in the presence of loyalty and demonstrate that a retailer has an incentive to invest in building brand ...

REVISION: Category Pricing with State Dependent Utility
Date Posted:Sat, 14 Jan 2006 17:20:13 -0600
There is a substantial literature that documents the presence of state dependent utility with packaged goods data. Typically, a form of brand loyalty is detected whereby there is a higher probability of purchasing the same brand as has been purchased in the recent past. The economic significance of the measured loyalty remains an open question. We consider the category pricing problem in the presence of loyalty and demonstrate that a retailer has an incentive to invest in building brand ...

Banner Advertising as a Customer Retention Tool in Customer Relationship Management
Date Posted:Thu, 05 Jan 2006 15:55:59 -0600
One of the major advances of the digital economy is the facilitation of building and managing individual customer relationships - a process usually referred to as "customer relationship management" or CRM. For a typical web site selling frequently-purchased consumer items, the most important stage of CRM is customer retention. This is because the long-term viability of a website is based on its ability to retain a significant customer base. In this study, we focus on a hitherto unexplored ...

An Empirical Model of Advertising Dynamics
Date Posted:Mon, 31 Jan 2005 02:46:09 -0600
We develop a model of dynamic advertising and apply it to the problem of optimal advertising scheduling through time. In many industries we observe advertising pulsing, whereby firms systematically switch advertising on and off at a high-frequency. The previous literature has explained such patterns through an S-shaped sales response to advertising, and long-run effects of advertising on demand (advertising carry-over). We extend a discrete choice based demand system to allow for a threshold ...

Endogenous Sunk Costs and the Geographic Distribution of Brand Shares in Consumer Package Goods Indu
Date Posted:Thu, 20 Jan 2005 10:58:01 -0600
This paper describes industrial market structure in consumer package goods (CPG) industries using a unique database spanning 31 industries and the 50 largest US metropolitan markets. A general set of stylized facts is documented pertaining mainly to the geographic patterns in brand shares. A connection between the patterns and a model of endogenous sunk costs in advertising is established by testing several predictions of the theory. We establish that concentration is bounded below in ...

Empirical Analysis of Indirect Network Effects in the Market for Personal Digital Assistants
Date Posted:Mon, 15 Mar 2004 02:42:16 -0600
We present a framework to measure empirically the size of indirect network effects in high-technology markets with competing incompatible technology standards. These indirect network effects arise due to inter-dependence in demand for hardware and compatible software. By modeling the joint determination of hardware sales and software availability in the market, we are able to describe the nature of demand inter-dependence and to measure the size of the indirect network effects. We apply the ...