BOUNIE David

We analyze competition between data intermediaries collecting information on consumers, which they sell to firms for price discrimination purposes. We show that competition between data intermediaries benefits consumers by increasing competition between firms, and by reducing the amount of consumer data collected. We argue that merger policy guidelines should investigate the effect of the data strategies of large intermediaries on competition and consumer surplus in related markets.

This article investigates the strategies of a data broker selling information to one or to two competing firms. The data broker combines segments of the consumer demand that allow firms to third-degree price discriminate consumers. We show that the data broker (1) sells information on consumers with the highest willingness to pay. (2) keeps consumers with low willingness to pay unidentified. The data broker strategically chooses to withhold information on consumer demand to soften competition between firms. These results hold under first degree price discrimination, which is a limit case when information is perfect.

No summary available.

No summary available.

No summary available.

No summary available.

No summary available.

This paper investigates interregional retail trade linkages, and changes in the gravity effects between cities and regions arising from online commerce, as opposed to traditional point-of-sale commerce. We build original interregional retail trade measures from nearly 20 billion domestic consumer online and in-store transactions made through bank cards, in France 2018-19. We are able to study the mobility of individual bank card holders throughout France, their on-site purchases, and also the locations from which their online purchases were made. We find evidence that online consumer expenditure tends to be more heavily concentrated in the already-large regional economies. This result suggests that the increasing movement toward online purchasing may tend to increase the concentration of overall economic activity, and may have important implications for regional economic development.

Although decision-support algorithms are now part of everyday life and are used in all sectors, especially finance, they are still criticized for their lack of transparency and their biased results. Researchers have studied this issue and make recommendations to reverse the situation.

No summary available.

No summary available.

This paper investigates a number of general phenomena connected with consumer behaviour in response to a severe economic shock, using billions of French card transactions measured before and during the COVID-19 epidemic. We examine changes in consumer mobility, anticipatory behaviour in response to announced restrictions, and the contrasts between the responses of online and traditional point-of-sale (off-line) consumption expenditures to the shock. We track hourly, daily and weekly responses as well as estimating an aggregate fixed-period impact effect via a difference-indifference estimator. The results, particularly at the sectoral level, suggest that recourse to the online shopping option diminished somewhat the overall impact of the shock on consumption expenditure, thereby increasing resiliency of the economy.

No summary available.

No summary available.

Identifying the effect of COVID-19 pandemic on individuals and economies is a challenging task, as official economic data have, as they inevitably do, lagged events. Using billions of anonymised French bank card transactions from before and during the COVID-19 epidemic, this column examines changes in consumer mobility, anticipatory behaviour in response to announced restrictions, and the contrasts between the responses of online and traditional point-of-sale consumption expenditures to the shock. Based on responses tracked at hourly, daily, and weekly frequencies, it finds that consumers' use of the online shopping option has mitigated the overall impact of the shock.

This paper investigates interregional retail trade linkages, and changes in the gravity effects between cities and regions arising from online commerce, as opposed to traditional point-of-sale commerce. We build original interregional retail trade measures from nearly 20 billion domestic consumer online and in-store transactions made through bank cards, in France 2018-19. We are able to study the mobility of individual bank card holders throughout France, their on-site purchases, and also the locations from which their online purchases were made. We find evidence that online consumer expenditure tends to be more heavily concentrated in the already-large regional economies. This result suggests that the increasing movement toward online purchasing may tend to increase the concentration of overall economic activity, and may have important implications for regional economic development.

We present a framework for defining the "right" level of explainability based on technical, legal and economic considerations. Our approach involves three logical steps: First, define the main con-textual factors, such as who is the audience of the explanation, the operational context, the level of harm that the system could cause, and the legal/regulatory framework. This step will help characterize the operational and legal needs for explanation, and the corresponding social benefits. Second, examine the technical tools available, including post-hoc approaches (input perturbation, saliency maps.) and hybrid AI approaches. Third, as function of the first two steps, choose the right levels of global and local explanation outputs, taking into the account the costs involved. We identify seven kinds of costs and emphasize that explanations are socially useful only when total social benefits exceed costs.

The recent enthusiasm for artificial intelligence (AI) is due principally to advances in deep learning. Deep learning methods are remarkably accurate, but also opaque, which limits their potential use in safety-critical applications. To achieve trust and accountability, designers and operators of machine learning algorithms must be able to explain the inner workings, the results and the causes of failures of algorithms to users, regulators, and citizens. The originality of this paper is to combine technical, legal and economic aspects of explainability to develop a framework for defining the "right" level of explain-ability in a given context. We propose three logical steps: First, define the main contextual factors, such as who the audience of the explanation is, the operational context, the level of harm that the system could cause, and the legal/regulatory framework. This step will help characterize the operational and legal needs for explanation, and the corresponding social benefits. Second, examine the technical tools available, including post hoc approaches (input perturbation, saliency maps.) and hybrid AI approaches. Third, as function of the first two steps, choose the right levels of global and local explanation outputs, taking into the account the costs involved. We identify seven kinds of costs and emphasize that explanations are socially useful only when total social benefits exceed costs.

No summary available.

This thesis studies firms that specialize in selling personal information and how their strategies affect competition. In three chapters, we build theoretical models representing the interactions between data brokers specializing in the collection and sale of information, firms buying personal information about their customers, and consumers. We first study the strategies of data brokers in selling information for price targeting of consumers. We show that data brokers do not sell all the information they have, and that a portion of consumers remain unidentified. We then study the strategies of data brokers when they compete in price and quantity of data sold. We show that competition affects consumers in two ways. The more competition there is between brokers, the less information they collect, but the less strategic information retention they engage in. Finally, we consider the reaction of consumers when they can hide from data brokers, and show that the fact that consumers have the ability to hide will induce the broker to sell more consumer data, creating a data externality. These studies help answer important questions about regulating the collection and sale of information, while taking into account a potential active role for consumers in these practices.

We investigate the strategies of a data intermediary selling customized consumer information to firms for price discrimination purpose. We analyze how the mechanism through which the data intermediary sells information influences how much consumer data he will collect and sell to firms, and how it impacts consumer surplus. We consider three selling mechanisms tailored to sell customized consumer information: take it or leave it offers, sequential bargaining, and simultaneous offers. We show that the more data the intermediary collects, the lower consumer surplus. Consumer data collection is minimized, and consumer surplus maximized under the take it or leave it mechanism, which is the least profitable mechanism for the intermediary. We argue that selling mechanisms can be used as a regulatory tool by data protection agencies and competition authorities to limit consumer information collection and increase consumer surplus.

No summary available.

We investigate the strategies of a data intermediary selling customized consumer information to firms for price discrimination purpose. We analyze how the mechanism through which the data intermediary sells information influences how much consumer data he will collect and sell to firms, and how it impacts consumer surplus. We consider three selling mechanisms tailored to sell customized consumer information: take it or leave it offers, sequential bargaining, and simultaneous offers. We show that the more data the intermediary collects, the lower consumer surplus. Consumer data collection is minimized, and consumer surplus maximized under the take it or leave it mechanism, which is the least profitable mechanism for the intermediary. We argue that selling mechanisms can be used as a regulatory tool by data protection agencies and competition authorities to limit consumer information collection and increase consumer surplus.

No summary available.

Algorithms are becoming more and more a part of our daily lives, as in the case of decision support algorithms (recommendation or scoring algorithms), or autonomous algorithms embedded in intelligent machines (autonomous vehicles). Deployed in many sectors and industries for their efficiency, their results are more and more discussed and contested. In particular, they are accused of being black boxes and of leading to discriminatory practices related to gender or ethnicity. The objective of this article is to describe the biases linked to algorithms and to outline ways to remedy them. In particular, we are interested in the results of algorithms in relation to equity objectives, and in their consequences in terms of discrimination. Three questions motivate this article: By what mechanisms can algorithmic biases occur? Can they be avoided? And, finally, can they be corrected or limited? In a first part, we describe how a statistical learning algorithm works. In a second part, we are interested in the origin of these biases, which can be of a cognitive, statistical or economic nature. In a third part, we present some promising statistical or algorithmic approaches to correct biases. We conclude the article by discussing the main societal issues raised by statistical learning algorithms such as interpretability, explainability, transparency, and accountability.

Disruptive innovations in digital payments are happening in a large number of countries around the world. While consumers may access to a wide variety of payment technologies, a natural question arises: does accepting a new payment technology allow merchants to increase their business sales? Using matching and difference-indifference techniques on a unique sample of about 275,580 merchants in France, we find that accepting contactless payments in 2018 increases on average the card-sales amount by 17 percent (and by 20 percent the card-sales count) compared to merchants who do not accept con-tactless payments. We also find evidence that accepting contactless payments exerts a positive spillover of about 3 percent in the amount of contact card sales, and is also more profitable for small merchants and new entrepreneurs.

No summary available.

No summary available.

The thesis presents an economic analysis of three new topics related to the economics of online advertising. It uses both empirical and theoretical methods. Chapter 1 presents the background of the thesis as well as the main results developed in the following chapters. In chapter 2, the thesis shows through a theoretical model how transparency in the modes of advertising purchases modifies the market equilibrium. In a context where advertising is considered a nuisance by users, the chapter shows that the introduction of such transparency-enhancing technologies can affect welfare in different ways. Chapter 3 discusses the economic relevance of profiling technologies that allow websites to adapt their level of advertising to the user's sensitivity. The chapter builds a theoretical model and draws the main implications arising from the use of these technologies. In Chapter 4, the thesis highlights how a privacy policy that allows users to opt out of ad targeting can have economic implications in the online advertising market. The chapter presents a new computational methodology and estimates the impact of such a privacy policy on ad prices. The results differ from previous work and show that the impact depends strongly on the characteristics of the website as well as the attractiveness of other sales channels. Finally, Chapter 5 concludes by highlighting how the key points developed throughout the chapters have significant economic weight.

This paper investigates the strategies of a data broker in selling information to one or to two competing firms that can price-discriminate consumers. The data broker can strategically choose any segment of the consumer demand (information structure) to sell to firms that implement third-degree price-discrimination. We show that the equilibrium profits of the data broker are maximized when (1) information identifies the consumers with the highest willingness to pay. (2) consumers with a low willingness to pay remain unidentified. (3) the data broker sells two symmetrical information structures. The data broker therefore strategically sells partial information on consumers in order to soften competition between firms. Extending the baseline model, we prove that these results hold under first-degree price-discrimination.

No summary available.

No summary available.

This paper investigates the strategies of a data broker in selling information to one or to two competing firms that can price-discriminate consumers. The data broker can strategically choose any segment of the consumer demand (information structure) to sell to firms that implement third-degree price-discrimination. We show that the equilibrium profits of the data broker are maximized when (1) information identifies the consumers with the highest willingness to pay. (2) consumers with a low willingness to pay remain unidentified. (3) the data broker sells two symmetrical information structures. The data broker therefore strategically sells partial information on consumers in order to soften competition between firms. Extending the baseline model, we prove that these results hold under first-degree price-discrimination.

Bulletin de la Banque de France 220/2-NOVEMBER-DECEMBER 2018 The use of cash in France: priority to low-value transactions This article uses French data from a survey on the use of cash in the euro area, published at the end of 2017 on behalf of the European Central Bank. Cash turns out to be the preferred means of payment in eurozone outlets: it accounts for three out of four payments and more than half of the value of purchases. However, the results by country are heterogeneous, with France standing out for its concentrated use of small-value purchases and its high propensity for cashless payments. Among the determinants of cash usage, the socio-demographic characteristics of consumers (gender, age, income, occupation) are not very significant. On the other hand, the characteristics of the transactions play a dominant role: in particular, the share of cash payments decreases when the value of the transaction increases, and increases when the transaction is made in a local store. JEL codes E4, E41 68% the share of cash payments made at the point of sale in France (in volume) 73.5% the share of cash purchases made by French people aged 55-64 90% the share of payments with a value of less than 5 euros that are settled in cash.

This thesis studies the economic consequences of consumers controlling the amount of personal data they are willing to share with firms. This topic is of vital importance to the digital economy as many firms collect and use consumer information afin order to increase their profits. However, these firms may face increasing difficulties in generating profits from personal data. First, more and more consumers are concerned about their online privacy. Second, a growing number of tools are available to control the collection of personal data. We show in the thesis that the use of such tools have a positive effect on consumers' willingness to share their personal information, thus enabling more privacy-friendly data collection. These findings also apply to the topic of financial data. Encouraging the use of non-bank payment methods (and thus allowing consumers to hide information from banks) could be beneficial to online commerce, leading consumers to make more purchases. Infin a context where consumers are concerned about their privacy, banks could benefit from making their screening practices less intrusive, as it would make their credit policies more effective.

No summary available.

No summary available.

No summary available.

No summary available.

The two-sided market theory holds that consumer adoption and merchant acceptance of payment cards are interdependent. However, empirical evidence on such network externalities is scarce, especially for the merchant side. This paper addresses this issue by examining merchant card acceptance in France. We exploit shopping diary data to construct a novel and fine-grained measure of French consumers’ payment preferences and match these with data from a nation-wide merchant survey. Controlling for (among other factors) cost, degree of competition, and customer characteristics, we find that the higher the probability that the average basket of a merchant is paid for by card in shops in the same sector and region, the higher the probability that the merchant will accept cards. In other words, we find that consumer preferences drive merchant card acceptance, which underpins the existence of network externalities on the merchant side of the payment card market.

No summary available.

No summary available.

No summary available.

Despite various payment innovations, today, cash is still heavily used to pay for low-value purchases. This paper develops a simulation model to test whether standard implications of the theory on cash management and payment choices can explain the use of payment instruments by transaction size. In particular, using diary survey data from Canada, France, Germany and the Netherlands, we test the assumption that cash is still the most efficient payment instrument, and the idea that people hold cash for precautionary reasons when facing uncertainty about their future purchases. The results of the simulations show that these two factors are significant determinants of the high shares of low-value cash payments in Canada, France and Germany. Yet, they are not so crucial in the Netherlands, which exhibits a significant share of low-value card transactions. We discuss how the differences in payment markets across countries may explain the performance of the model.

We measure consumers' use of cash by harmonizing payment diary surveys from seven countries: Australia, Austria, Canada, France, Germany, the Netherlands, and the United States (conducted 2009 through 2012). Our paper finds important cross-country differences, for example, the level of cash usage differs across countries. Cash has not disappeared as a payment instrument, especially for low-value transactions. We also find that the use of cash is strongly correlated with transaction size, demographics, and point-of-sale characteristics, such as merchant card acceptance and venue.

No summary available.

No summary available.

Recent rapid growth in electronic book sales has raised a critical question for publishers and bookstores: do e-books cannibalize or increase print sales? In this article, we compare the best-selling titles sold on www.Amazon.com in print or electronic Kindle formats during the period from November 2007 to July 2010. Using econometric methods, we find that the cannibalization of print sales by e-books is more likely to occur for superstar titles written by successful authors. However, we find that a new segment of successful electronic titles that are not best-selling in print format emerge. these books would probably have been unpopular without the new Kindle store and therefore this new distribution channel has expanded the market. We refer to these titles as digital outsiders. The latter are characterized not only by lower prices but also by older release dates. They also include titles that are only released in electronic format. We then argue that electronic books increase the market viability of old print releases. Finally we identify a category that we call print-preferred books that are top sellers in print but not as e-books for reasons of color, graphics, or the need to navigate non-linearly, a style to which the current generation of e-book readers are not well adapted.

No summary available.

No summary available.

No summary available.

No summary available.

This thesis is presented in the form of a 134-page volume composed of 5 research articles. The thesis is an empirical contribution to the literature that has developed since the early 2000s on the upheavals introduced by the Internet on the trade of cultural goods. In particular, it examines a set of questions concerning the complementarity or substitution of physical and virtual distribution channels, the "Long Tail" effect and the dispersion of prices on the Internet. The interest of the thesis is to bring empirical evidence to these debates, thanks in particular to the creation of databases obtained by automated capture of observable data on the Internet. The statistical and econometric results of this work detail the specificities of the best-selling books, CDs and DVDs according to the distribution channels (Amazon, Amazon Marketplace, physical stores) but also according to their format (paper books/digital books). On the other hand, the results show a low variability of prices by Amazon Marketplace sellers as well as a low impact of traditional measures of reputation (seller ratings) compared to the size of the sellers' catalogs, questioning the use of ratings as a proxy for reputation.

Empirical studies on the social cost of payments agree that cash is the most costly payment instrument for society. Central and commercial banks have implemented various strategies to reduce both the holding and use of cash in payments. These include a category of so-called non-price strategies such as improving the currency division system (CDS) and changing the technology for making cash. However, the actual effectiveness of these strategies is subject to debate. This thesis contributes to this field of study through the determination of optimal currency divisions and the cost-effectiveness of MDS. It provides a better understanding of the implications of both payment behavior, manufacturing technology, and MDS structure on the distribution of cash transactions in the economy and the social cost of cash. First, we relax some restrictive assumptions in comparing the efficiency of MDS based on an empirical distribution of cash transactions in France. Second, we question a widely held view that efficient MDSs reduce the social cost of payments and also establish that payment solutions that are efficient in terms of the least effort principle may be more costly for the central bank. We propose a novel framework for measuring the benefit of adopting a new cash technology and, finally, we compare the social cost of two SDMs by simulating a mature version of the search models.

No summary available.