Personal Blogs

Overview

Speaker: Antonella Trigari

Presentation of a work-in-progress paper on how countercyclical fiscal policy and aggregate demand interact with sectoral labor reallocation. The study documents U.S. sectoral reallocation patterns, estimates policy effects on reallocation and unemployment, and develops a multi-sector heterogeneous-agent New Keynesian framework with search and matching frictions and input-output linkages to analyze mechanisms and policy transmission.

Participants

• Presenter: Antonella Trigari (Bocconi University)
• Panelists: Anna Rogantini-Picco (ECB), Krisztina Molnar (Norwegian School of Economics), Morgane Richard (Sciences Po)
• Host affiliation mentioned: BIS

Motivation and Research Goals

• Shocks have become increasingly asymmetric and sectoral (pandemic, trade, energy, climate, artificial intelligence), generating uneven labor displacement and making labor reallocation central for adjustment.
• Policy responses often remain untargeted across sectors; debate centers on transfers to households vs. subsidies to firms.
• Goals:
  • Document U.S. sectoral labor reallocation and sectoral heterogeneity over the last two decades, introducing new gross reallocation indices.
  • Estimate effects of countercyclical fiscal policy—e.g., Unemployment Insurance (UI) extensions and firm subsidies like PPP (Paycheck Protection Fund)—on reallocation and aggregate unemployment, both directly and via reallocation.
  • Develop and calibrate a multi-sector Heterogeneous Agent New Keynesian (HANK) model with search and matching frictions and an input-output (IO) network to study policy transmission and connect model counterfactuals to empirical estimates.

Key Empirical Evidence and Measures

• Data and approach: CPS (Current Population Survey) longitudinal data used to measure “realized reallocation” via cross-sector new hires and last pre-unemployment sector.
• Realized reallocation magnitude: On average, 48% of job finders move to a sector different from their last employment sector.
• Search reallocation (unobserved search sectors): New method combining data and a recursive stock equation for sectoral searchers, using estimated sectoral job-finding rates, to infer switching across search sectors.
  • Finding: 32% of job seekers switch the sector in which they search.
• Cyclicality: Both realized and search-based reallocation indices are cyclical and decline during major recessions.
• Sectoral heterogeneity in risk and insurance:
  • Scatter of sectoral unemployment rates (by last employment and search-based measures) versus a consumption ratio (consumption of unemployed relative to employed) shows substantial heterogeneity.
  • Negative correlation: Sectors with higher unemployment risk tend to offer lower consumption insurance.

Mechanisms: Aggregate Demand Amplifications and Reallocation

• HANK feedback loop (aggregate demand–unemployment risk):
  • Compositional channel: As workers move across sectors with heterogeneous insurance against unemployment risk, aggregate demand effects shift with sectoral composition.
  • Within-sector channel: Reallocation alters sectoral conditions (e.g., sectors losing workers may see higher job-finding rates for remaining workers; receiving sectors may see higher risk).
  • Shock increases unemployment risk → stronger precautionary saving → lower consumption/aggregate demand → with nominal frictions, firms post fewer vacancies → unemployment risk rises further.
  • Interaction with reallocation:
• IO amplification (HANK–IO):
  • Sectoral demand shocks cascade through input-output linkages, reducing demand in connected sectors and amplifying initial shocks.
  • Also features compositional and within-sector effects.

Model: Structure and Novel Elements

• Baseline: Tractable HANK-SAM setup with incomplete markets and endogenous idiosyncratic unemployment risk.
• Additions:
  • Multiple ex-ante heterogeneous sectors with input-output linkages.
  • Sectoral reallocation via a family construct implying cross-reallocation across sectors.
  • Randomness in sectoral search choices (as in trade literature) to match observed gross flows where moves occur even when sector values are similar.
• Labor market: Search and matching frictions; vacancy posting responds to demand conditions.
• Firms: New Keynesian Dixit–Stiglitz structure with an additional wholesale layer to separate price-setting decisions.

Policy Focus and Examples

• Countercyclical fiscal policies examined include:
  • UI extensions (e.g., during the Great Recession or COVID).
  • Firm subsidies such as PPP (Paycheck Protection Fund), noted during the Great Recession.

Positioning in Related Literature

• Contrasts with the Lilien (1982) perspective, where desired reallocation across sectors, slowed by frictions, causes unemployment.
• Focus here is on realized reallocation—actual worker movements—mediating the transmission and effects of shocks and policies.

Planned Components and Application

• Presentation plan includes evidence, a model sketch, and “an application to an automata.”
• Claimed contribution: First model combining heterogeneous agents, endogenous unemployment risk, search and matching, input-output linkages, and endogenous sectoral reallocation in a tractable setup.

Next Steps / Ongoing Work

• Continue estimating the impact of countercyclical fiscal policies on sectoral reallocation and aggregate unemployment, both directly and via reallocation channels.
• Calibrate the model to U.S. data.
• Use model counterfactuals to connect predictions to empirical estimates.
• Refine and apply new indices of gross labor reallocation to additional episodes and shocks.

References

• Lilien, D. M. (1982). Sectoral Shifts and Cyclical Unemployment. Journal of Political Economy, 90(4), 777–793. http://www.jstor.org/stable/1831352