Laura Castillo-Martinez

WORKING PAPERS

Price Gaps and Inflation Dynamics with Miguel Bandeira and Shiyuan Wang

April 2026 [pdf]

Non-linear inflation dynamics are a key feature of menu cost models. Recent work has mostly focused on one dimension of this non-linearity: larger shocks have disproportionately larger effects. Yet the impact of a given shock also depends on the state of the economy at the time it hits. This paper studies how the cross-sectional distribution of price gaps—the misalignment between firms' actual and reset prices—shapes inflation dynamics. A central challenge is that this distribution is unobservable. We address this by recasting a random menu cost model in state-space form and developing Bayesian methods to recover quote-level price gaps from observed prices. Applying this approach to UK CPI microdata from 1997 to 2023, we construct a time-varying distribution of price gaps and document substantial variation over time, with the economy often far from its ergodic benchmark. We show that this variation matters: it gives rise to state-dependent transmission of monetary policy and serves as a leading indicator of future inflation.

Capital Slack with Isaac Baley, Miguel Bandeira, Andrés Blanco, Madalena Gaspar and Nicolás Oviedo

February 2026 [pdf]

Investment is lumpy: firms remain inactive for long stretches and then adjust capital in bursts. This lumpiness drives a wedge between the capital firms would like to have and the capital they actually install. We recover this wedge—which we term capital slack —by estimating firms’ latent desired capital from plant-level data. We microfound a nonlinear state-space representation using an Ss investment model with fixed adjustment costs and occasional free adjustment opportunities, and apply filtering and smoothing methods to infer reset capital in real time. Aggregating the recovered reset-capital series yields a capital slack index that leads the business cycle and predicts future movements in aggregate investment.

Sudden Stops, Productivity, and the Exchange Rate

December 2025 [pdf] [online appendix]

Following a sudden stop, productivity often declines, while real exchange rates adjust through a nominal depreciation, lower domestic prices, or both. Cross-country evidence suggests that productivity declines are larger when nominal depreciation dominates the real exchange rate adjustment. Motivated by this pattern, the paper studies how the nature of exchange-rate adjustment shapes productivity dynamics during sudden stops. Using Spanish manufacturing micro-data from two sudden stops under different reg, it shows that, in a currency union, cleansing through exit is stronger than under a floating regime, with aggregate productivity rising despite weaker firm-level performance. A small open-economy DSGE model with firm dynamics, endogenous markups, and nominal rigidities rationalizes these findings. The model identifies three channels through which a sudden stop affects productivity: pro-competitive, cost, and demand. While only the first operates under a floating regime, all three are active in a currency union. Quantitatively, the model explains about 55 percent of the exit-driven contribution to productivity growth in Spain’s 2010–13 episode.

Financial Frictions and Firm Exit with Gideon Bornstein

Awarded NSF Grant (co-PI with Gideon Bornstein)

September 2023 [pdf]

This paper studies the role of financial frictions in firm exit behavior and develops a model to assess the costs of inefficient firm exit. Using European firm-level data, we document a positive relationship between firm-level exit and leverage, controlling for non-financial firm characteristics. We find that this relationship is significantly stronger during recessions. We then construct a firm dynamics model with financial frictions. The model endogenously delivers a stronger relationship between firm exit and leverage during recessions. We show that the correlation between firm exit and a firm's financial condition prior to exit is informative of the degree of financial frictions in the economy. Using a calibrated version of the model, we assess the costs of inefficient firm exit due to financial frictions and study the effectiveness of different government interventions during recessions.

PUBLICATIONS

How Do Central Banks Control Inflation? A Guide For the Perplexed with Ricardo Reis

Journal of Economic Literature, vol. 64, no. 1, March 2026 (pp. 195-245) [published version] [local copy] [online appendix]

Central banks have a primary goal of price stability. They pursue it using tools that include the interest they pay on reserves, the size and the composition of their balance sheet, and the dividends they distribute. We describe the economic theories that justify the central bank’s ability to control inflation and discuss their relative effectiveness in light of the historical record. We present alternative approaches as consistent with each other, as opposed to conflicting ideological camps. While interest-rate setting is often superior, having both a monetarist pillar and fiscal support is essential, and at times pegging the exchange rate or monetizing the debt is inevitable.

WORK IN PROGRESS

Sufficient Statistics for Investment Dynamics: An Empirical Investigation with Isaac Baley, Miguel Bandeira, Andrés Blanco, Madalena Gaspar and Nicolás Oviedo

We test the sufficient statistics formula for aggregate investment dynamics derived from a lumpy adjustment model. Using state-space methods and Bayesian estimation applied to microdata in Chile and Colombia, we show that sufficient statistics successfully predict cumulative investment responses to aggregate productivity shocks.

SS for SS: State-Space Methods for Lumpy Economies with Isaac Baley and Miguel Bandeira

This paper introduces a general statistical framework to study aggregate shock propagation in a dynamic (S,s) economy. The proposed framework enables researchers to combine micro and macro data to estimate all the parameters in this economy by maximum likelihood or Bayesian methods, estimate the cross-sectional distributions latent states over time, estimate impulse responses to aggregate shocks of any function of the cross-sectional distribution of lumpy variables given any initial distribution of state gaps and to forecast any function of the cross-sectional distribution of lumpy variables.