Shoshana Vasserman

New technologies have enabled firms to elicit granular behavioral data from consumers in exchange for lower prices and better experiences. This data can mitigate asymmetric information and moral hazard, but it may also increase firms' market power if kept proprietary. We study a voluntary monitoring program by a major U.S. auto insurer, in which drivers accept short-term tracking in exchange for potential discounts on future premiums. Using a proprietary dataset matched with competitor price menus, we document that safer drivers self-select into monitoring, and those who opt-in become 30% safer while monitored. Using an equilibrium model of consumer choice and firm pricing for insurance and monitoring, we find that the monitoring program generates large profit and welfare gains. However, large demand frictions hurt monitoring adoption, forcing the firm to offer large discounts to induce opt-in while tempering the profitability of heavy ex-post rent extraction. A counterfactual data portability policy would further reduce the firm's incentive to elicit monitoring data, leading to less monitoring and lower consumer welfare in equilibrium.

Economists routinely make functional form assumptions on demand curves to derive welfare conclusions. How sensitive are these conclusions to such assumptions? In this paper, we develop robustness measures that quantify the extent to which the true demand curve must deviate from common functional form assumptions in order to overturn a welfare conclusion. We parametrize this variability in terms of the gradient and curvature of the demand curve. By leveraging tools from information design, we show that our measures are easy to compute. Our measures are also flexible and easy to use, as we illustrate through several empirical applications.

Most U.S. government spending on highways and bridges is done through "scaling" procurement auctions, in which private construction firms submit unit price bids for each piece of material required to complete a project. Using data on bridge maintenance projects undertaken by the Massachusetts Department of Transportation (MassDOT), we present evidence that firm bidding behavior in this context is consistent with optimal skewing under risk aversion: firms limit their risk exposure by placing lower unit bids on items with greater uncertainty. We estimate bidders' risk aversion, the risk in each auction, and the distribution of bidders' private costs. Simulating equilibrium item-level bids under counterfactual settings, we estimate the fraction of project spending that is due to risk and evaluate auction mechanisms under consideration by policymakers. We find that scaling auctions provide substantial savings relative to lump sum auctions and show how our framework can be used to evaluate alternative auction designs.

Firms have ever increasing access to consumer data, which they use to personalize their advertising and to price discriminate. This raises privacy concerns. Policymakers have argued in response that consumers should be given control over their data, able to choose what to share and when. Since firms learn about a consumer's preferences both from what they do and do not disclose, the equilibrium implications of consumer control are unclear. We study whether such measures improve consumer welfare in monopolistic and in competitive markets. We find that consumer control can improve consumer welfare relative to both perfect price discrimination and uniform pricing. First, consumers can use disclosure to amplify competitive forces. Second, consumers can disclose information to induce even a monopolist to lower prices. Whether consumer control improves welfare depends on the disclosure technology and market competitiveness. Simple disclosure technologies suffice in competitive markets. When facing a monopolist, a consumer needs partial disclosure possibilities to obtain any welfare gains.

Major changes to the operation of local newsrooms — ownership restructuring, layoffs, and a reorientation away from print advertising — have become commonplace in the last decades. However, there have been few systematic attempts to characterize the impact of these changes on the types of reporting that local newsrooms produce. In this paper, we propose a method to measure the investigative content of news articles based on article text and influence on subsequent articles. We use our method to examine over-time and cross-sectional patterns in news production by local newspapers in the United States over the past decade. We find surprising stability in the quantity of investigative articles produced over most of the time period examined, but a notable decline in the last two years of the decade, corresponding to a recent wave of newsroom layoffs.

Empirical evidence shows that equally informed, experienced negotiators may refuse to settle because they fundamentally disagree on each one's probability of success. We study the dynamics of agreement to settle in pretrial negotiations when the negotiating parties are both optimistic and new information may arrive at any point.We characterize the conditions under which the negotiators do or do not reach an agreement at every period of negotiation and discuss the implications for policy design such as timing periods of discovery and jury selection, and whether or not to allow the winning party to recover the legal costs incurred from the losing party.

How well can an informed central planner like Waze do at routing drivers on paths with uncertain wait times using an incentive compatible protocol? We find that the mediation ratio is at most 8/7 in the case of two links with affine cost functions, and remains strictly smaller than the price of anarchy of 4/3 for any fixed m. However, it approaches the price of anarchy as m grows. For general (monotone) cost functions, the mediation ratio is at most m, a significant improvement over the unbounded price of anarchy.

We study the impacts of New York City's Central Business District (CBD) Tolling Program, the first cordon-based congestion pricing scheme in the United States. Using a generalized synthetic controls approach that compares outcomes in NYC to contemporaneous outcomes in other cities, we find that the policy increased speeds on CBD roads by 11%, with little-to-no effect on air quality, transactions at shops and restaurants, or overall foot traffic in the CBD. Speeds also increased outside the CBD, especially on roads commonly traversed by drivers traveling to/from the CBD. These spillovers lead to faster trips throughout the metro area, including for many unpriced trips. We develop a simple model to bound the effects on driver welfare. If drivers have a Value of Travel Time (VOTT) of $40/hour, then we estimate that driver welfare increased by at least $14.3 million/week, before any revenue recycling or environmental benefits. Passenger vehicles headed to the CBD are only better off if their VOTT exceeds $153/hour, but the modest speed improvements for the many unpriced trips reduce the overall 'break-even' VOTT of these drivers to at most $21/hour. Finally, we show how characteristics of local travel patterns and road networks can inform the potential impacts of introducing cordon-based congestion pricing in other cities.

Using a novel dataset covering the complete history of individual-level web traffic and digital subscriptions from a major metropolitan newspaper in the United States from 2020 to 2023, we investigate consumers' willingness to pay for different kinds of content, including that believed to have the greatest civic externalities. Our identification relies on the quasi-random arrival of paywall events which force consumers to subscribe if they wish to continue reading. Using this variation, we estimate a model of consumer demand and construct the optimal content portfolio for the paper under different counterfactual revenue models: a fully subscription-based model and a fully ad-supported model. Our results suggest that news consumers are willing to pay for investigative and local content, and that measures of demand based only on time-use substantially underestimate the value of investigative news.

This paper analyzes the effects of the largest field experiment to date that incentivizes drivers to limit driving during peak hours and congested areas via usage-based congestion pricing. The experiment monitored the driving behavior of 10,000 Israeli drivers who were recruited over the course of 2020. During the first six months of a driver's participation in the experiment, their driving behavior is monitored and recorded; afterwards, drivers receive an annual budget and are charged for each kilometer driven during historically congested times in congested areas. Whatever remains in each driver's budget is paid to them a year later. We use comprehensive data on driving behavior of participating drivers over the course of 2020 and 2021 to evaluate how usage-based congestion pricing affects driving behavior. The staggering of driver recruitment facilitates identifying treatment effects via a difference-in-differences approach. We find that drivers decrease their congested driving behavior by 10-20\% across a myriad of outcomes designed to detect both extensive margin (i.e. whether to take a trip) and intensive margin (i.e. when to take a trip) responses. We also find that there is significant treatment effect heterogeneity across drivers that can be predicted by pre-treatment driving behavior. The most affected drivers tend to be those who contributed more to congestion and who appear to have more flexibility in their driving choices and easier access to public transit, but they are not disproportionately socioeconomically advantaged or disadvantaged. We then estimate a model of demand for taking "habitual" trips by car at different times given different travel durations and costs. Using this model and an estimated relationship between traffic densities and speeds on major highways, we consider the equilibrium impacts of congestion pricing on network-wide speeds and welfare.

Many matching markets are suspected to suffer from inefficient levels of congestion. We show this is a real concern in an online labor market and present results of two market-wide experiments designed to reduce congestion. The first intervention introduced a "soft" cap on the number of applications that could be received for a job opening and the number of days applications were accepted. Despite reducing the number of applications per opening, the intervention did not reduce the hiring probability or reported match quality. A second, more complex intervention that attempted to price externalities directly failed. We find that application fees introduced by the platform reduced hire rates and competition among candidates, suggesting that these fees may have been miscalibrated or higher than socially efficient.

The United States spends twice as much per person on pharmaceuticals as European countries, in large part because prices are higher in the US. This has led policymakers in the US to consider legislation for price controls. This paper assesses the effects of a hypothetical US reference pricing policy that would cap prices in US markets by those offered in Canada. We estimate a structural model of demand and supply for pharmaceuticals in the US and Canada, in which Canadian prices are set through a negotiation process between pharmaceutical companies and the Canadian government. We then simulate the impacts of the counterfactual international reference pricing rule, allowing firms to internalize the cross-country impacts of their prices both when setting prices in the US and when negotiating prices in Canada. We find that such a policy results in a slight decrease in US prices and a substantial increase in Canadian prices. The magnitude of these effects depends on the particular structure of the policy. Overall, we find modest consumer welfare gains in the US, but substantial consumer welfare losses in Canada. Moreover, we find that pharmaceutical profits increase in net, suggesting that reference pricing of this form would constitute a net transfer from consumers to firms.

Data collection and personalization are ubiquitous today, raising concerns of privacy and price discrimination. In response to these concerns, both regulatory authorities and the market emphasize the value of consumer control. In the public sphere, the European Union's GDPR and California's Consumer Privacy Act require websites to obtain consent before collecting browsing information and restrict the duration that consumer data can be retained. In the private sphere, Apple, Google, and other firms have rolled out product features that allow consumers to opt out of personalized tracking. Although consumer control features prominently in discussions of privacy policies, relatively little is known about how to model consumer control and its effects on market outcomes. If consumers control the information possessed by sellers, is price discrimination beneficial? Or should it nevertheless be prohibited? Does it suffice for the consumer to be able to opt out from sharing information or is control at a more granular level needed? How does this relate to market competitiveness? Clearly, a strategic framework is necessary to assess the implications of privacy policies. Our work (Ali, Lewis and Vasserman, 2022) offers such a framework. We view consumer control through the lens of voluntary disclosure. The consumer has certain verifiable characteristics—her age, income, or her data—that are correlated with her preferences. Rather than sellers having this information at the outset, the consumer chooses what to disclose to the market. From this perspective, opting in to a firm's tracking policy is tantamount to disclosing data predictive of her preferences; opting out, by contrast, corresponds to choosing not to disclose information. These are but two choices; one may envision contexts that endow the consumer with control at a more granular level—e.g., the ability to disclose a student ID or a senior citizen card—without having to opt in entirely. In equilibrium, firms do not take disclosed information at face value; instead, they draw inferences both from what is said and what is left unsaid. We use this framework to answer the questions above. In monopolistic markets, consumers do not benefit from personalized pricing if the only choice they have is to opt in or out from sharing information; more fine-tuned control is necessary for them to benefit from personalized pricing. By contrast, if the market is competitive, control even in the form of simple opt-in / opt-out policies are enough to assure consumer gains. Disclosure amplifies competition. Contrary to the view that firms should not price discriminate, our findings suggest that consumers may benefit from price discrimination if they control the flow of information. We describe these findings in greater below and also pose new questions for which our approach may be useful.

Auctions are inherently risky: bidders face uncertainty about their prospects of winning and payments, while sellers are unsure about revenue and chances of a successful sale. Auction rules influence the allocation of risk among agents and the behavior of risk-averse bidders, leading to a breakdown of payoff and revenue equivalence and a heightened significance of auction design decisions by sellers. In this paper, we review the literature on risk aversion in auctions, with an emphasis on what can be learned about auction design from theoretical modeling and empirical studies. We survey theoretical results relating to the behavior of risk-averse agents in auctions, the comparison of standard auction formats in the presence of risk aversion and implications for auction design. We discuss standard and more recent approaches to identifying risk preferences in empirical studies and evidence for the significance of risk aversion in auction applications. Finally, we identify areas where existing evidence is relatively scant and ask what questions empirical research might ask given the theory and where further theoretical research may be beneficial given existing empirical results.

We develop and implement a heterogeneous-agents network-based empirical model to analyze alternative policies during a pandemic outbreak. We combine several data sources, including information on individuals' mobility and encounters across metropolitan areas, information on health records for millions of individuals, and information on the possibility to be productive while working from home. This rich combination of data sources allows us to build a framework in which the severity of a disease outbreak varies across locations and industries, and across individuals who differ by age, occupation, and preexisting health conditions. We use this framework to analyze the impact of different social distancing policies in the context of the COVID-19 outbreaks across US metropolitan areas. Our results highlight how outcomes vary across areas in relation to the underlying heterogeneity in population density, social network structures, population health, and employment characteristics. We find that policies by which individuals who can work from home continue to do so, or in which schools and firms alternate schedules across different groups of students and employees, can be effective in limiting the health and healthcare costs of the pandemic outbreak while also reducing employment losses.