Rethinking Regulation
Learning to Be Fair: A Consequentialist Approach to Equitable Decision Making
Alex Chohlas-Wood et al.
Management Science, forthcoming
Abstract:
In an attempt to make algorithms fair, the machine learning literature has largely focused on equalizing decisions, outcomes, or error rates across race or gender groups. To illustrate, consider a hypothetical government rideshare program that provides transportation assistance to low-income people with upcoming court dates. Following this literature, one might allocate rides to those with the highest estimated treatment effect per dollar while constraining spending to be equal across race groups. That approach, however, ignores the downstream consequences of such constraints and, as a result, can induce unexpected harm. For instance, if one demographic group lives farther from court, enforcing equal spending would necessarily mean fewer total rides provided and potentially more people penalized for missing court. Here we present an alternative framework for designing equitable algorithms that foregrounds the consequences of decisions. In our approach, one first elicits stakeholder preferences over the space of possible decisions and the resulting outcomes -- such as preferences for balancing spending parity against court appearance rates. We then optimize over the space of decision policies, making trade-offs in a way that maximizes the elicited utility. To do so, we develop an algorithm for efficiently learning these optimal policies from data for a large family of expressive utility functions. In particular, we use a contextual bandit algorithm to explore the space of policies while solving a convex optimization problem at each step to estimate the best policy based on the available information. This consequentialist paradigm facilitates a more holistic approach to equitable decision making.
The Effects of Advertising Disclosure Regulations on Social Media: Evidence From Instagram
Daniel Ershov & Matthew Mitchell
RAND Journal of Economics, forthcoming
Abstract:
We study the effects of advertising disclosure regulations in social media markets. Using data from a large sample of Instagram influencers in Germany and Spain and a difference-in-differences approach, we empirically evaluate the effects of German strengthening of disclosure regulations on post content and follower engagement. We measure whether posts include suggested disclosure terms and use text-based approaches (keywords, machine learning) to assess whether a post is sponsored. We show substantial adoption of disclosure but also a 12% increase in sponsored content and an increase in the share of undisclosed-sponsored content consumers are exposed to. We also find reductions in engagement, suggesting that followers were likely negatively affected.
The Bronx is Burning: Urban Disinvestment Effects of the Fair Access to Insurance Requirements
Ingrid Gould Ellen et al.
Federal Reserve Working Paper, December 2024
Abstract:
In response to private insurers' postwar withdrawal from urban neighborhoods, roughly half of U.S. states developed programs in the late 1960s that offered residual property insurance to property owners denied in the private market. These plans, known as Fair Access to Insurance Requirements (FAIR) plans after 1968, inadvertently encouraged moral hazard through underwriting restrictions, risk pooling, and generous payouts. We use a triple-difference design to estimate FAIR's impact, comparing: (1) pre-and post-FAIR participation periods, (2) neighborhoods likely offered FAIR plans versus those not, and (3) similar contrasts in non-participating states. FAIR plans led to significant housing disinvestment and declines in central neighborhood population and income in the late 1960s and 1970s.
Minimum Wages and Workplace Injuries
Michael Davies, Jisung Park & Anna Stansbury
MIT Working Paper, December 2024
Abstract:
Do minimum wage changes affect workplace health and safety? Using the universe of workers' compensation claims in California over 2000-2019, we estimate whether minimum wage shocks affect the rate of workplace injuries. Our identification exploits both geographic variation in state-and city-level minimum wages and local occupation-level variation in exposure to minimum wage changes. We find that a 10% increase in the minimum wage increases the injury rate by 11% in an occupation-metro area labor market which is fully exposed to the minimum wage increase. Our results imply an elasticity of the workplace injury rate to minimum-wage-induced wage changes of 1.4. We find particularly large effects on injuries relating to cumulative physical strain, suggesting that employers respond to minimum wage increases by intensifying the pace of work, which in turn increases injury risk.
Unintended workplace safety consequences of minimum wages
Qing Liu et al.
Journal of Public Economics, November 2024
Abstract:
We investigate the unintended impact of minimum wage increases on workplace safety. Using establishment-level data from the United States and a cohort-based stacked difference-in-differences design, we find that large increases in minimum wages have significant adverse effects on workplace safety. Our findings indicate that, on average, a large minimum wage increase results in a 4.6 percent increase in the total case rate. Event study estimates show that this adverse effect persists in the medium run. Furthermore, we find a more salient effect for firms more likely to be financially constrained or subject to a higher labor market rigidity in firing workers. We provide suggestive evidence that small minimum wage increases might reduce injury rates, highlighting the potential heterogeneity in the impact of minimum wage changes. We do not find evidence that capital-labor substitution could be behind the findings.
Asymmetric Information Sharing in Oligopoly: A Natural Experiment in Retail Gasoline
David Byrne et al.
Journal of Political Economy, forthcoming
Abstract:
Using a natural experiment from a retail gasoline antitrust case, we study how asymmetric information sharing affects oligopoly pricing. Empirically, price competition softens when, following case settlement, information sharing shifts from symmetric to asymmetric, with one firm losing access to high-frequency, granular rival price data. We provide theory and empirics illustrating how strategic ignorance creates price commitment, leading to higher price-cost margins. Using a structural model, we find substantial profit-enhancing effects of asymmetric information sharing. These results provide a cautionary tale for antitrust agencies regarding the potential unintended consequences of limiting price information sharing among firms.
Sources of Market Power in Web Search: Evidence from a Field Experiment
Hunt Allcott et al.
NBER Working Paper, January 2025
Abstract:
We evaluate the economic forces that contribute to Google’s large market share in web search. We develop a model of search engine demand in which consumer choices are influenced by switching costs, quality beliefs, and inattention, and estimate it using a field experiment with US desktop internet users. We find that (i) requiring Google users to make an active choice among search engines increases Bing’s market share by only 1.1 percentage points, implying that switching costs play a limited role; (ii) Google users who accept our payment to try Bing for two weeks update positively about its relative quality, with 33 percent preferring to continue using it; and (iii) after changing the default from Google to Bing, many users do not switch back, consistent with persistent inattention. In our model, correcting beliefs and removing choice frictions would increase Bing’s market share by 15 percentage points and increase consumer surplus by $6 per consumer-year. Policies that expose users to alternative search engines lower Google’s market share more than those requiring active choice. We then use Microsoft search logs to assess the impact of additional data on search result relevance. The results suggest that sharing Google’s click-and-query data with Microsoft may have a limited effect on market shares.
Information sharing across MLS platforms and housing prices: Evidence from a temporary suspension of an agreement
Ping Cheng et al.
Real Estate Economics, forthcoming
Abstract:
This study examines a unique case of a suspension of a data-sharing agreement between two multiple listing services (MLSs) in South Florida, and investigates the impact of such a (negative) shock on the trading outcomes of the MLS platforms. Using property listing data from the two MLSs, we find that sale prices were lower during the data suspension period for transactions involving properties that were in areas where the listing broker may rely on brokers from the other MLS to bring potential buyers. The negative price impact remains stable and consistent under various robustness checks. Furthermore, the results also reveal heterogeneous pricing and liquidity effects due to the data-sharing suspension-related shock.
Harassment or neglect? How market dynamics and rent control shape landlord behaviour in Los Angeles
Sean Angst et al.
Urban Studies, forthcoming
Abstract:
This paper examines whether and how housing market dynamics shape landlords’ profit-seeking behaviours, focusing on harassment and property neglect. Leveraging household survey data, we assess whether differences between market and contract rents, rent control and gentrification influence landlord behaviour. Findings reveal that one-quarter of respondents reported inadequate maintenance from landlords within the past two years, and more than one-fifth reported at least one form of harassment. However, the incidence of these issues varied across contexts. Tenants in rent-controlled buildings and gentrifying census tracts were 14.8 and 9.4 percentage points more likely than peers not in those situations to experience harassment, respectively. Moreover, rent-controlled tenants were more likely to experience illegal eviction practices while those in gentrifying tracts were more likely to experience threats and assault. In contrast, paying lower rents relative to market estimates alone was not associated with a greater likelihood of refusal to provide maintenance and a lower likelihood of harassment. These results suggest that landlords respond in illegal ways when frictions in the market make it difficult to simply increase rents in response to strengthening market conditions.
Pricing Inequality
Simon Mongey & Michael Waugh
NBER Working Paper, January 2025
Abstract:
This paper studies household inequality and product market power in dynamic, general equilibrium. In our model, households’ price elasticities of demand endogenously vary with wealth. Heterogeneous firms set their price as oligopolistic competitors given the endogenous distribution of demand. A firm’s market power varies with the distribution of demand as households with different elasticities sort into high- and low-price varieties. Under standard preferences, larger firms’ products are more appealing, sell at higher prices, to more households, and a relatively richer customer base, face less elastic demand, and set higher markups. Quantitatively (a) our model rationalizes a wide set of recent empirical studies in the cross-section of households and firms, (b) we find household heterogeneity to be a dominant source of markup variation across firms, and (c) a one-time fiscal transfer of one percent of GDP to households leads to a 0.3 percentage point increase in the aggregate markup.