理论论文

Which countries drive global inequality? Which states drive US inequality? We characterize the class of inequality measures that can answer these questions, permitting both an additive decomposition by subgroup and a multiplicative decomposition into within- and between-group components. Its central measure is simply the expected ratio of incomes between two randomly selected individuals. We show that global inequality fell since 1990, driven by declining disparities between countries. Inequality is primarily driven by large middle-income countries (India, China) and poor countries (Democratic Republic of Congo, Ethiopia). In the US, rising inequality is explained by widening gaps within states, while between-state inequality remains limited.

Many people are impatient. We introduce domain-specific discounting to a contracting model and test a strategy for improving incentives when people are impatient over effort: implement time-bundled contracts that make the payment for future effort increase in current effort. Using a randomized evaluation of an exercise incentive program among diabetics in India, we find that time-bundled contracts generate equivalent effort to time-separable contracts at 15 percent lower cost. Moreover, they perform 30 percent better among individuals with above-median effort impatience. Pooled across contracts, incentives increase steps by 20 percent and improve blood sugar relative to a control group. (JEL D86, D91, I12, I18, O12)

This paper develops a structural framework for analyzing top income taxation and applies it to six decades of U.S. tax history and to eight other advanced economies. The paper derives explicit expressions for top-bracket tax revenue and total excess burden across the full range of tax rates, while preserving local consistency with the sufficient-statistics approach of Saez (2001). Applied to the United States, the analysis implies that the current excess burden of the top federal tax bracket is $101 billion and that raising the top marginal tax rate to its revenue-maximizing level of 72 percent would increase annual revenue by $111 billion, given a taxable income elasticity of 0.25. The current tax rate is lower than what is consistent with a purely utility-based notion of social welfare. By contrast, Nordic countries and the United States in earlier decades appear to have operated at tax rates exceeding revenue-maximizing levels.

As shortages of essential resources become more frequent, high prices alone may fail to curb demand. We study power limits for residential electricity, which restrict household consumption rather than setting it to zero as in rolling blackouts. We show theoretically and empirically that power limits can deliver electricity savings equivalent to sizable blackouts while affecting fewer households. The mechanism works through selection of high users, creating distributional trade-offs that depend on whether limits are uniform or proportional to contracted power. This highlights the potential of smart meters to improve shortage management while preserving access to essential electricity services.

Should public policies address inequality due to heterogeneous life expectancy? Intuitively, taking a short life as a disadvantage, such policies should favor those with high mortality. Yet, pension systems implicitly redistribute from low-life-expectancy to high-life-expectancy people. Moreover, this direction of redistribution is optimal from the perspective of the standard utilitarian welfare criterion. We study mortality-related redistribution in a more flexible setting. We start by establishing a formal framework for the analysis by clearly distinguishing between redistribution along the mortality and income dimensions. We then characterize the conditions under which it is optimal to redistribute toward high-mortality people. One such condition is that the welfare criterion must feature an aversion to lifetime inequality that exceeds the aversion to consumption inequality.

This paper studies the optimal combination of government grants that provide direct public funding to nonprofits and tax incentives that reduce the cost of charitable giving. When nonprofits face high fundraising costs to attract donors, tax incentives cannot fully replace grants, which remain part of the optimal funding policy. When fundraising costs are sufficiently low, a government that can tailor tax incentives across causes can finance some causes entirely through donations. However, when cause-specific tax differentiation is unavailable, grants remain necessary to accommodate differences in funding needs across causes. In a calibrated example, imposing a uniform tax treatment of donations across charitable causes raises the optimal share of nonprofit funding covered by grants from 0 to 60%.

We study technology investment and public good provision in a setting where two parties with differing valuations of the public good alternate in power. In each period, the incumbent chooses public good provision and invests in technology that lowers the cost of future public good provision. We characterize Markov perfect equilibria and show how polarization and irreversibility of investment jointly shape long-run outcomes. When polarization is low, power turnover may increase expected technology investment and public good provision, and steady-state outcomes are unaffected by the degree of reversibility. When polarization is high, however, irreversibility becomes important: the party favoring the public good invests more aggressively, and irreversible investment induces higher public good provision by both parties. As polarization rises, expected technology stock and public good provision initially decline but jump upward once the irreversibility constraint binds.

A designer employs a dynamic protocol to elicit private information. Protocols produce a set of contextual privacy violations—information learned that may be superfluous given the context. A protocol is maximally contextually private if there is no protocol that produces a proper subset of the violations it produces, while implementing the choice rule. Contextual privacy violations arise when a choice rule makes some agents collectively, but not individually, pivotal. In auctions, designing for contextual privacy requires choosing an initial question posed to each agent and the order for querying agents. Ascending-join protocols are maximally contextually private for k-item Vickrey auctions. (JEL D44, D82)

A principal hires an agent to learn about the cost of a project (experimentation) and then to execute it (production). The agent is privately informed about the probability that the cost is low, with the high-type agent more optimistic than the low type. Learning is modeled as a series of experiments, where the agent privately chooses his effort in each experiment. Due to the joint presence of adverse selection and moral hazard, even the low type can misrepresent his type, shirk during experimentation and earn a rent. In the optimal contract, the principal uses the length and outcome of experimentation as well as the timing of payments to induce effort and screen the agent. The principal may find it optimal to reward the low type after failure in experimentation. We provide sufficient conditions for both success and failure to be rewarded. We show that over-experimentation can be optimal as it reduces the asymmetric information after a series of failed experiments. We also consider whether it may be optimal to separate experimentation and production between two different agents. When the same agent works on both tasks, the principal can leverage the adverse selection rent to mitigate moral hazard. Therefore, integrating experimentation and production is optimal when adverse selection is severe.

We develop a simple and intuitive Pigouvian perspective on optimal trade policy. Our approach unifies a wide range of rationales for taxing trade, from the classical optimal tariff argument to contemporary debates about global carbon emissions and geopolitics. We also clarify when trade policy intervention is warranted and when alternative domestic instruments should be used instead.

This paper proposes a framework for assessing whether misspecified decision-makers would be willing to pay for information that can potentially make them less misspecified. We introduce a prior-free approach, based on “constrained” maximal regret, to derive an upper bound on the subjective assessment of potential gains from acquiring a more accurate model. The constraint stems from the information currently available to the decision-maker. We apply our approach to three prominent models of misspecified beliefs: coarse expectations, causal misperceptions, and sampling equilibria. (JEL D83, D84, D91)

People participate in protests even though it is costly, benefits are not exclusive to participants, and each individual is unlikely to be pivotal. Building on these characteristics, I propose a sequential model of protest formation where opportunities to participate arrive stochastically, and the protest succeeds if participation surpasses a threshold. I show a unique equilibrium exists and characterize its dynamics. If the unique equilibrium involves participation, the time between participation decisions shortens as the protest progresses—a pattern observed empirically. A player with an opportunity becomes sequentially pivotal if, after abstaining, the protest will likely end before another opportunity arrives. (JEL D72, D84, D91)

We study misperceptions of private-signal correlation in an incomplete-information Cournot duopoly game. Exaggerating the correlation between players' demand signals is beneficial when agents hold flexible beliefs about price elasticity but is harmful when their beliefs are dogmatically correct. For agents with flexible beliefs who learn elasticity by observing prices, correlation misperceptions indirectly distort behavior through elasticity misinference. If agents know the true elasticity, this learning channel is eliminated. Correlation misperceptions have opposite direct and indirect effects on behavior, so the presence of elasticity inference can reverse an error's viability. (JEL C72, D21, D43, D81, D82, D83)

We study a data provider's problem of optimally designing and selling marketing lists that enable a retailer with private types to access consumers with heterogeneous characteristics. When virtual values satisfy a particular form of convexity (named boundary dominance), under certain regularity conditions, the optimal mechanism comprises fully revealing marketing lists, irrespective of whether the retailer's types are vertically or horizontally ordered. Moreover, in the vertical case, the optimal mechanism allows a simple “fixed-price” implementation. In the horizontal case, it exhibits no-exclusion and rotation properties. Conversely, if boundary dominance is violated, the optimal mechanism may not be fully revealing. (JEL D82, L81, M31)

This paper studies bid rigging in auctions with bidder preselection. We develop a theoretical model to analyze the optimal behavior of partial bid-rigging cartels and show how commonly used two-stage auction formats, in which the first stage is used to preselect bidders, may be exploited. Based on our theoretical results, we construct a collusion marker reflecting the optimal cartel strategy and, using administrative data on public procurement auctions in Slovakia, we identify bidders suspected of collusion. We show that bidder preselection allows cartels to exclude competitive rivals and thereby increase procurement costs above what would be possible without preselection. (JEL D43, D44, H57, L12)

We show that the formalization of representativeness (Kahneman and Tversky, 1972, Cognitive Psychology, 3, 430–454) developed by Gennaioli and Shleifer (2010, The Quarterly Journal of Economics, 125, 1399–1433) features an intrinsic connection between uncertainty and overreaction. In the time series domain, we develop this connection in a smooth version of diagnostic expectations (DE), where overreaction varies smoothly with uncertainty. Intuitively, under smoothness, lower uncertainty leaves less room for representativeness to distort beliefs. Smooth DE provides a joint, parsimonious micro-foundation for key features of survey data: overreaction to news, stronger overreaction at longer horizons, overconfidence in subjective uncertainty, and a new stylized fact documented here—overreaction intensifies with higher uncertainty. An analytical Real Business Cycle model with Smooth DE accounts for survey overreaction and overconfidence, as well as three salient business cycle properties: asymmetry, countercyclical macro volatility, and countercyclical micro volatility.

This paper investigates how to persuade multiple interacting receivers under arbitrary equilibrium selection and when receivers’ preferences may exhibit psychological traits. Our main result characterizes the conditions under which the sender can categorize information using partitions and then disclose partition-related information without altering the equilibrium outcome. This finding broadens the traditional direct approach based on the revelation principle, which categorizes messages solely by equilibrium actions. Consequently, our results support a generalized direct approach, which we apply to study information disclosure in fostering fairness within groups where members may demonstrate altruistic behaviors.

We extend global games á la Carlsson and van Damme (1993) to environments where the risk-dominant equilibrium is selected even if there is no dominance solvable game in the underlying class of games. Strict dominance can emerge in the global game from strategic uncertainty due to discrete payoff changes in underlying games, and we provide sufficient conditions on payoff changes that warrant iterated dominance of the risk-dominant equilibrium. Thus, strategic uncertainty creates strictly dominant actions as well as fostering iterated dominance, in contrast to global games hitherto where strategic uncertainty does only the latter. Discrete payoff changes tend to arise, in particular, in situations where a public good can be provided with varying degrees of coordination depending on the state, so that coordinating actions can be strategic substitutes and free-riding incentives are present. We illustrate our findings in a stylized public good provision model.

This paper studies a model with wage bargaining, random on-the-job search and renegotiation. Wages are determined by a bargaining process in which the firm and the worker alternate in making offers and there is a probability that this process breaks down. In the model, a given wage contract ends at a Poisson rate, and then the firm and the worker bargain over a new wage. In this setting, a higher wage decreases the firm's markup, but this effect is partly offset by lowering turnover, which increases match surplus. This increase in match surplus enables the worker to capture a higher share of the surplus. This positive effect of a higher wage on match surplus diminishes when they renegotiate more frequently. Eventually, as the contract length tends to zero, the equilibrium of the model converges to the equilibrium payoffs discussed by Pissarides (1994). My model thereby justifies using the Nash bargaining solution with perfectly transferable values in models with on-the-job search. In contrast, when the Poisson rate goes to zero, the equilibrium in the model, which I show to be unique, converges to one of the equilibria found by Shimer (2006).

We consider many-to-many matching with contracts where each agent can sign multiple contracts with the same agent. In other words, the unitarity condition is not assumed in this model. We offer a weaker substitutability condition called substitutability across doctors, which is a generalization of unilateral substitutability. We show that if every hospital’s choice function satisfies substitutability across doctors and every doctor’s choice function is responsive, then a stable outcome exists. We establish this result by generalizing the cumulative offer process. Moreover, we propose an even weaker substitutability condition that incorporates observability of doctors’ contract offers during the algorithm. The same algorithm guarantees the existence of a stable outcome even when substitutability across doctors is replaced with the weaker condition. We additionally present a specific class of choice functions satisfying the weaker condition that can be applied to a temporary staffing market, particularly in which temp agencies may assign multiple temporary workers to each client.