Paula Donaldson

Job Market Paper

Cross-sectional Identification with Heterogeneous Exposure to General Equilibrium Effects

Cross-sectional identification combining heterogeneous exposure to an aggregate shock with time fixed effects has become an important tool to estimate cross-sectional elasticities of macro importance, such as cross-sectional fiscal multipliers. In this paper, I (i) show that identification fails when heterogeneous exposure to the aggregate shock is correlated with exposure to other aggregate variables that respond in general equilibrium, (ii) develop a test for the failure of identification, and (iii) propose a new decomposition method that leverages both cross-sectional and time-series variation to identify cross-sectional elasticities, free of general equilibrium effects. I apply the method to estimate cross-sectional fiscal multipliers in the U.S. and find that accounting for general equilibrium effects operating through monetary policy reduces the estimated two-year multiplier by an order of magnitude. This result challenges the assumption that time fixed effects are sufficient to yield U.S. cross-sectional multipliers invariant to monetary policy responses.

Working Papers

The Financial Channel of Tax Amnesty Policies with Federico Bernini, Ezequiel Garcia Lembergman and Leticia Juarez
[PDF]

In the past two decades, more than 30 countries have implemented tax amnesty policies to encourage the declaration and repatriation of hidden assets, with the goal of increasing government tax revenues. While previous literature has focused primarily on the fiscal impact, this paper studies a new channel: the potential expansion of the financial sector resulting from these policies. We examine the macroeconomic effects of Argentina's 2016 Tax Amnesty, one of the largest programs for disclosing hidden assets, through the financial channel. This amnesty led to an influx of savings into domestic banks, primarily in dollars, equivalent to 1.4% of GDP. We leverage the heterogeneous exposure of banks and firms to this amnesty-induced financial shock to identify bank responses and spillovers to firms in the private sector. We find that more exposed banks significantly increased their lending compared to less exposed ones. Firms connected to banks with higher exposure experienced increased borrowing, along with a boost in imports of intermediate inputs, exports, and employment. Our findings reveal that tax amnesty policies can stimulate economic growth by expanding the financial sector, demonstrating effects beyond their direct fiscal impact. These results are particularly relevant for countries with underdeveloped financial systems, where the potential for growth through improved access to capital is significant.

Work in Progress

Why Are Some Recoveries Weak and Others Strong? with Johannes Wieland

Why were the recoveries from the 1990-1, 2001, and 2007-9 recessions weak relative to other postwar recessions? Leveraging heterogeneous exposure to the national business cycle across U.S. States, we estimate the trajectory of more exposed U.S. States relative to less exposed U.S. States. For the 1990-1, 2001, and 2007-9 recessions we estimate that more exposed States experienced a stronger boom-bust cycle and for the other postwar recessions we estimate a deeper V-shape recession in more exposed States. Our estimates support theories that these recessions are caused by different shocks. In a quantitative model matched to our cross-sectional estimates, boom-bust cycles persistently depress the natural rate of interest $R^{^{*}}$ in the recovery, whereas $R^{^{*}}$ is elevated following V-shaped recessions.
Placed Based Policy for National Welfare with Rowan Shi and Fabian Trottner

We propose a novel rationale for nationally designed place-based policies: imperfect competition in local labor markets. To outline and analyze the implications of this mechanism for welfare and policy, we develop a spatial equilibrium model where firms have market power in both labor and product markets. Market power in local labor markets distorts the allocation of resources within and across regions, implying different industrial efficiencies and returns to scale across space. Solving the problem of a planner, we show that place-based industrial policy can be optimal even absent redistributional motives, because it addresses inefficiencies arising from market power. Moreover, such policy must be designed at the national level, to account for how an integrated goods market transmits local labor market imperfections across regions. To quantify the importance of our theoretical results, we study the effects of the German place-based policies aimed at eliminating disparities between the Western and Eastern states.