Working Papers 2026
1-2026 Nittai K. Bergman. A Note on Technological Complexity, Ability, and the Timing of Inequality
Technological progress reshapes not only aggregate productivity but also the distribution of economic outcomes. A large body of work studies how advances in technology affect wage inequality, often emphasizing average skill premia or broad differences between skilled and unskilled workers. Distributional evidence points to complex patterns: Inequality below the median has stabilized or even declined in many settings, while inequality in the upper tail of the income distribution has continued to rise and top income shares have expanded. These divergent movements suggest that inequality evolves unevenly across the income distribution and that different segments respond to technological progress at different times. This paper proposes a simple mechanism that generates such distributional rotation as a natural consequence of technological complexity interacting with heterogeneous worker ability. The core idea is that technological progress increases not only the productivity frontier but also the complexity of production. New technologies typically require greater cognitive, organizational, or problemsolving capacity to be used effectively. Workers differ in these capacities, and crucially, the productivity gains from frontier technologies depend nonlinearly on how well a worker’s ability aligns with the prevailing level of technological complexity.
2-2026 Marlon Azinovic-Yang, Nir Jaimovich, Itay Saporta-Eksten, Yaniv Yedid-Levi. The Anatomy of Amplification: A Frisch-System Approach to Shock Transmission
Elasticities governing consumption and labor choices are central to macroeconomic shock transmission, with existing work emphasizing the intertemporal elasticity of substitution and the Frisch labor-supply elasticity. We highlight the role of a third object: the Frisch consumption-wage elasticity, which captures consumption-labor complementarity. Embedding a nonparametric Frisch demand system in a New Keynesian model, we map these three micro elasticities to aggregate dynamics, clarifying how each elasticity shapes the amplification of monetary, fiscal, markup, and productivity shocks and why consumption-labor complementarity can matter more than Frisch labor-supply elasticity. Quantitatively, in representative-agent and heterogeneous-agent New Keynesian models, we show that consumption-labor complementarity is a dominant amplification channel, often generating responses an order of magnitude larger than those induced by varying the Frisch labor-supply elasticity. These results underscore the value of a framework that allows macroeconomic models to match micro estimates of
consumption-labor complementarity, a key determinant of shock amplification
3-2026 Daniel Fershtman. Alessandro Pavan. Non-Sequential-Non-Cascading Clicking in Online Consumer Search
We consider the problem of a consumer searching online for the best alternative. We show how this problem can be formalized as a variant of Pandora’s boxes problem where the decision maker retains control over the order in which boxes are opened, but where, contrary to the original version, the set of boxes grows over time and is controlled by a third party (the platform). We use the model to (a) endogenize click-through-rates, (b) explain the phenomenon of non-sequential-non-cascading clicking, (c) illustrate why the generalized second-price auction may lead to inefficient assignments even under its ascending-clock implementation, and (d) show why a firm’s profits need not be increasing in the number of ads it displays on a platform even when they are free.
4-2026 Tomer Ifergane. Concentrated Risk: Misallocation and Granular Business Cycles
When firm-level shocks affect aggregate outcomes, distortions reshaping the firm-size distribution affect not only output but also aggregate volatility. The sign depends on the correlation between distortions and productivity: compressing high-productivity firms reduces concentration and volatility, while compressing low-productivity firms raises both. In a U.S. calibration, positively correlated distortions reduce aggregate volatility resulting. From firm-level shocks by 42 percent relative to a zero-distortion benchmark. Normatively, dispersion in marginal products need not imply inefficiency. Rather, in a granular economy, the risk-averse planner optimally diversifies production away from high-productivity firms, incurring an output loss in exchange for reduced aggregate risk.
