ניירות דיון 2025
WP 1-2025
Risky College Enrollment, Dropout, and Student Debt Forgiveness
Michael Kaganovich and Itzhak Zilcha
Abstract
The paper analyzes the effects of two kinds of college education subsidies: unconditional tuition discounts and targeted forgiveness of student loans on student college enrollment and completion or dropout decisions. We focus on students’ imperfect knowledge of their academic ability at the time of matriculation and its updating in the course of study as key factors in their responses to funding policies. We find that while unconditional tuition subsidies incentivize both matriculation and continued study even upon the revelation of low ability hence low returns to college, a policy combining such subsidy with partial forgiveness of student debt conditional on dropping out has a doubly efficient effect of risk mitigation: it maintains incentives to matriculate but discourages continued study when low future returns are revealed. It is, moreover, superior in terms of mitigating the “bad debt” held by students, that unrecouped by returns to college. Budget neutral conversion of a part of unconditional tuition subsidy to targeted debt forgiveness reduces the aggregate bad debt held by students.
WP 2-2025
Bureaucracy in quest of feasibility
Herve Cres a, Itzhak Gilboa, Nicolas Vieille
Abstract
A bureaucracy has to determine the values of many decision variables while satisfying a set of constraints. The bureaucracy is not assumed to have any objective function beyond achieving a feasible solution, which can be viewed as ‘‘satisficing’’ à la Simon (1955). We assume that the variables are integer-valued and the constraints are linear. We show that simple and (arguably) natural versions of the problem are already NPHard. We therefore look at decentralized decisions, where each office controls but one decision variable and can determine its value as a function of its past values. However, an attempt to consult more than a single past case can lead to Condorcet-style consistency problems. We prove an Arrovian result, showing that, under certain conditions, feasibility is guaranteed only if all offices mimic their decisions in the same past case. This result can be viewed as explaining a status quo bias.
WP 3-2025
Negative Control Falsification Tests for Instrumental Variable Designs
Oren Danieli, Daniel Nevo, Itai Walk, Bar Weinstein, Dan Zeltzer
Abstract
The validity of instrumental variable (IV) designs is typically tested using two types of falsification tests. We characterize these tests as conditional independence tests between negative control variables—proxies for unobserved variables posing a threat to the identification—and the IV or the outcome. We describe the conditions that variables must satisfy in order to serve as negative controls. We show that these falsification tests examine not only independence and the exclusion restriction, but also functional form assumptions. Our analysis reveals that conventional applications of these tests may flag problems even in valid IV designs. We offer implementation guidance to address these issues.
WP 4-2025
News Media as Suppliers of Narratives (and Information)
Kfir Eliaz and Ran Spiegler
Abstract
We present a model in which news media shape beliefs by providing information (signals about an exogenous state) and narratives (models of what determines outcomes). To amplify consumers' engagement, the media maximize their anticipatory utility. We characterize the optimal monopolistic media strategy under various classes of separable consumer
preferences, and demonstrate the synergy between false narratives and biased information. Consumer heterogeneity gives rise to a novel menu- design problem due to an equilibrium data externality among consumers. The optimal menu features multiple narratives and creates po-
larized beliefs and choices. These e¤ects also arise in a competitive media market model.
WP 5-2025
Central Bank Digital Currency: When Price and Bank Stability (Don't) Collide
Daniel Bird and David Weiss
Abstract
In a recent influential paper, Schilling, Fernández-Villaverde and Uhlig (2024) caution that the introduction of a central bank digital currency gives rise to a central bank trilemma in a nominal version of the quintessential Diamond and Dybvig (1983) model of bank-runs. We show that there is a natural policy tool that can be introduced into their environment to solve this trilemma.
WP 6-2025
Industry Firm Dynamics and Financing Conditions
Nittai K. Bergman, Rajkamal Iyer, Paymon Khorrani, William Mullins
Abstract
This paper analyzes how changes in financing conditions affect industry dynamics. We show that
industry leaders, defined as firms with the largest market share in a given industry, perform better than industry followers during periods of tight financing conditions. When financing conditions are tight, industry leaders outperform industry followers and experience higher abnormal returns, gain market share, invest more, raise more long-term debt, and have higher profitability rates. These effects are primarily concentrated in industries in which the disparity between leaders and followers, as measured by market capitalization, is large. Given these heterogeneous intra-industry effects, our results indicate that tight financing conditions widen inequality between industry leaders and followers, particularly when pre-existing industry inequality is large.
WP 7-2025
Likes
Tuval Danenberg, Zvika Neeman, Daniel Spiro
Abstract
A principal faces n agents. She privately favors a subset of these agents but wants as many agents as possible to believe they are favored. To communicate, the principal uses public “likes.” We characterize the “liking strategies” that can be sustained in a robust equilibrium and show that they must involve a fixed and constant number of likes. Additionally, we describe conditions for when monotone liking strategies can and cannot be sustained as an equilibrium, regardless of robustness. We apply the model to workplace promotion promises, grade inflation, political campaigns, and liking on social media.
WP 8-2025
Entrepreneurship and the Racial Wealth Gap
Daniel Albuquerque, Tomer Ifergane
Abstract
Entrepreneurship promotes wealth accumulation. However, Black households face significant barriers to entrepreneurship, operating fewer and smaller businesses. We formalize a general equilibrium model of entrepreneurship choice and wealth accumulation in which Black households experience adverse distortions as entrepreneurs and as workers. Disciplined by microdata, our model matches well the observed racial wealth gap and the correlation between wealth and entrepreneurship. We find that distortions faced by Black entrepreneurs are the key factor for understanding the racial wealth gap across the wealth distribution. Our analysis also indicates that addressing racial disparities in the U.S. can substantially increase output.
WP 9-2025
Beyond Test Scores: Does Public Information on School Satisfaction and Violence Levels Affect Parental School Choice?
Genia Rachkovski
Abstract
Amid the proliferation of school choice policies, substantial disparities persist in the quality of schools selected by high and low socioeconomic-status (SES) families. Can we decrease this gap by providing parents with better information, and if so, what information is effective at inducing parents to select better schools? To address these questions, I leverage a unique natural experiment in Israel, where a Supreme Court ruling mandated the public release of comprehensive school-level information. Employing a discrete choice model, an event study design, and a difference-in-difference approach, I examine the impact of this information disclosure on student school choice. I find that in regions with multiple school options, following the information disclosure, parents increasingly favor schools with better attributes. Notably, this shift is primarily attributed to factors such as violence levels and students’ school satisfaction ratings, rather than test score information. Importantly, the results show that the effect is driven by the increased response of lower SES households to the non-score-related attributes, narrowing the preexisting selection gap from their high SES counterparts. Crucially, I establish a robust association between school violence levels and school value-added measures, and find that lower SES households were more likely to select schools with higher value added following the information disclosure. Furthermore, I find that schools respond to the increased accountability by improving in the non academic publicized attributes. Finally, I identify information-sharing networks, based on shared ethnic and cultural ties, influencing school choices. My findings demonstrate the importance of providing a diverse set of school attribute information to enhance equity, academic outcomes and social mobility.
WP 10-2025
Macroprudential Regulations of Mortgage Contracts
Matilda Kilström, Kathrin Schlafmann, Ofer Setty, Roine Vestman
Abstract
We build an incomplete markets model to analyze the costs and benefts of borrower-based macroprudential policies. We calibrateit to match Sweden’s economic environment in the early 2000’s and analyze the effects of two regulations: a stricter down-payment re-quirement, from 10% to 15%, and a cap on debt-to-income (DTI) of 500%. We fnd that a stricter downpayment requirement has lim-ited positive benefts in terms of reducing consumption sensitivityto temporary shocks to income or the mortgage risk premium whileit reduces the homeownership rate of the middle-aged and has neg-ative welfare effects for 78% of the households. A cap on DTIs at 500% has small effects—on house prices as well as households’ wel-fare and resilience—because it is a binding constraint for only 8.5%of 25-year-olds but for no 35-year-old.
WP 11-2025
Frictions in News Consumption: Evidence from Social Media
Luca Braghieri, Ro’ee Levy, and Hannah Trachtman
Abstract
We document pervasive behavioral and information frictions in news consumption on social media, and test interventions designed to address them through a large field experiment on Facebook. We find that a simple, platform-integrated interface prompting active choice significantly and persistently reduced the partisan slant of users’ Facebook news portfolios, while coupling this interface with information substantially improved portfolio reliability. Our interventions also increased alignment between users’ stated preferences and their Facebook portfolios and influenced off-platform news consumption. The results underscore the promise of simple, behaviorally-informed, and scalable interventions to enhance news consumption on social media platforms.
