Research
Working Papers
The Fed's power over r-star.
Abstract: This paper provides new evidence and theory of the discretional power of the Federal Reserve to determine the expected long-term real risk-free rate, or “r-star”. First, I show that tight windows bracketing FOMC announcements account for the entire trend decline of expected long-run forward real rates since the early 1980s. My evidence improves on a previous discovery by filtering out inflation expectations and risk premia. I then provide a theory that explains these effects of FOMC announcements without having to assume any power of the Fed on the long-run fundamentals affecting r-star (preferences, technological growth, demographics), or any supposed superiority of the Fed’s long-run forecasts, or any deviations from rational expectations. The key is an overlapping-generations setup, which relaxes the tight link between fundamentals and r-star. This makes an exogenous, unexpected, and permanent change in the real short-term rate theoretically admissible. These exogenous changes stand in for a discretional, permanent change in the real intercept of the Taylor rule decided by the Fed and imply changes in the steady-state value of the main macroeconomic variables. These implied long-run macro adjustments are small relative to the observed long-run changes in the last four decades, even for a relatively large annual cut of r-star. This quantitative result supports the empirical plausibility of the theory.
How abundant are reserves? Evidence from the wholesale payment system.
with G. Afonso, D. Duffie and H. Shin.
Abstract: Before the era of large central bank balance sheets, banks relied on incoming payments to fund outgoing payments in order to conserve scarce liquidity. Even in the era of large central bank balance sheets, rather than funding payments with abundant reserve balances, we show that outgoing payments remain highly sensitive to incoming payments. By providing a window on liquidity constraints revealed by payment behavior, our results shed light on thresholds for the adequacy of reserve balances. Our findings are timely, given the ongoing shrinking of central bank balance sheets around the world in response to inflation.
Reassessing the impact of the 1832 Bank Veto with a new dataset.
Abstract: On July 10th, 1832, President Jackson vetoed the renewal of the federal charter of the Second Bank of the United States (SBUS), the US Central Bank, set to expire in 1836. Economic historians have long debated the impact of this political choice on the financial bubble of 1832-36 in the western frontier and subsequent Bank Panic of 1837, with some proposing that the wind down of the SBUS left more leeway for a boom of the local private banks. We contribute to this debate with a new dataset and a more modern empirical technique, improving the analysis of Knodell (2006). The new dataset includes bank balance sheets disaggregated at the State level, not only for the private state banks (as in Knodell), but also for the Central Bank, obtained by digitizing the original reports to the Congress and mapping the branches of the SBUS to their respective States. I then apply a synthetic control method to build a counterfactual for the "treated" States, where the closing of the SBUS was particularly important for deposits and credit. The analysis suggests that Knodell's quantification of the excess credit boom in the Southwest that followed the Bank Veto in roughly correct, but more work is needed to establish a clear causal link. In particular, most of the effect seems to be driven by a spectacular boom in Louisiana.
Other Work
Code of Brookings' Treasury Issuance Model.
Abstract: This project contributes code to replicate and extend Belton et al (2018), Hutchins Center Working Paper #46 "Optimizing the maturity structure of U.S. Treasury debt: A model-based framework". The paper is extended by allowing the model to also consider Treasury Inflation Protected Securities (TIPS) and Floating Rate Notes (FRNs).
Links: GitHub repository and related blog publication.
Asymmetric Fiscal Uncertainty.
Master Thesis, May 2017.
Abstract: What is the effect of skewed macroeconomic uncertainty about fiscal variables? This analysis extends the research by Fernández-Villaverde et al. (2015b) on fiscal stochastic volatility shocks to consider stochastic skewness shocks. Stochastic skewness of US Labor tax rate is recovered from a non-linear State Space Model thanks to a particle filter, and the effects of the shocks to this state process are analyzed in a standard Structural Vector Autoregression including the main US macroeconomic variables. My work differs from the treatment of tail risk in the disaster risk literature, since the new stochastic skewness state can have long tails in both directions. I find that a stochastic skewness shock to the process of Labor taxes can have a sizable effect on GDP, roughly similar to the impact of stochastic volatility shocks, but with some notable differences, including an inverted ranking in the reaction of Investment and Consumption and opposite effects on Prices