Research Papers:
1. Term Structure, Forecast Revision and the Signaling Channel of Monetary Policy, March 2019 (Submitted)
SSRN Working Paper, Online Appendix
Monetary policy shocks affect interest rates at long horizons (10 years or more). Furthermore, the private sector’s real GDP forecasts are revised upward in response to a monetary tightening. These facts challenge the prevailing theories in academic and policy circles. In this paper, I propose a micro-founded model to rationalize those facts, based on the signaling channel of monetary policy. I consider a framework where the central bank has private information about future economic conditions. Agents update their beliefs according to Bayes’ theorem. Policy actions play a signaling role, and may therefore rationalize the above empirical findings.
SSRN Working Paper, Online Appendix
Monetary policy shocks affect interest rates at long horizons (10 years or more). Furthermore, the private sector’s real GDP forecasts are revised upward in response to a monetary tightening. These facts challenge the prevailing theories in academic and policy circles. In this paper, I propose a micro-founded model to rationalize those facts, based on the signaling channel of monetary policy. I consider a framework where the central bank has private information about future economic conditions. Agents update their beliefs according to Bayes’ theorem. Policy actions play a signaling role, and may therefore rationalize the above empirical findings.
2. The Random Walk Beats Professional Forecast: Facts, Puzzles and Explanations, May 2018
For short horizon exchange rate predictability the simple random walk model outperforms professional forecasts. A new puzzle arises: why do professional forecasters not adapt the simple random walk model to provide a more accurate estimate? This paper provides an explanation. In this framework, the forecaster faces model uncertainty and reports the forecast that minimizes the forecast error under the worst-case scenario. Therefore professional forecasts are intentionally suboptimal. Estimation results show that the model matches the empirical puzzle. In addition, the model predicts that the forecaster substantially underreacts to current news, which is consistent with empirical facts provided in this paper. Moreover, the null of "rationality" is rejected using simulated data confirming existing findings even though forecasters in the model perform optimally.
For short horizon exchange rate predictability the simple random walk model outperforms professional forecasts. A new puzzle arises: why do professional forecasters not adapt the simple random walk model to provide a more accurate estimate? This paper provides an explanation. In this framework, the forecaster faces model uncertainty and reports the forecast that minimizes the forecast error under the worst-case scenario. Therefore professional forecasts are intentionally suboptimal. Estimation results show that the model matches the empirical puzzle. In addition, the model predicts that the forecaster substantially underreacts to current news, which is consistent with empirical facts provided in this paper. Moreover, the null of "rationality" is rejected using simulated data confirming existing findings even though forecasters in the model perform optimally.
3. Which Inflation Index to Stabilize? The Role of Market Power, May 2018
Existing empirical evidence suggest cross-sector heterogeneities in both nominal rigidity and market power. This paper studies the optimal choice of inflation index for a central bank to stabilize in a framework that embeds those features. The optimal weight attached to inflation in a sector is increasing in this sector’s: i) price stickiness (stickiness channel) and ii) degree of market competition (competition channel). Moreover, if firms in a more competitive sector adjust their price more frequently as predicted by costly price adjustment models, the competition channel offsets the stickiness channel. The finding challenges the conventional wisdom that the central bank should attach a higher weight to a sector with a higher degree of nominal rigidity, and supports the current practice of central banks around the world (CPI targeting).
Existing empirical evidence suggest cross-sector heterogeneities in both nominal rigidity and market power. This paper studies the optimal choice of inflation index for a central bank to stabilize in a framework that embeds those features. The optimal weight attached to inflation in a sector is increasing in this sector’s: i) price stickiness (stickiness channel) and ii) degree of market competition (competition channel). Moreover, if firms in a more competitive sector adjust their price more frequently as predicted by costly price adjustment models, the competition channel offsets the stickiness channel. The finding challenges the conventional wisdom that the central bank should attach a higher weight to a sector with a higher degree of nominal rigidity, and supports the current practice of central banks around the world (CPI targeting).
4. An Empirical Assessment of the Fed’s Targeting Rule and Loss Function with Shengliang Ou, May 2019
What is the relative weight a central bank attaches to the allocation of the output gap and inflation? The answer to this question is reflected in the optimal targeting rule. Empirical estimation of this measure is challenging due to the simultaneity problem. We overcome this issue by purging out shocks that shift a central bank’s targeting rule. The following results stand out. First, the Fed is willing to decrease the growth rate of the output gap by two and half percentage points for each percentage point increase in (quarterly) inflation. Second, the data rejects optimal discretionary policy in favor of optimal policy under commitment. Third, the Fed’s targeting rule was not much changed during the pre-Volcker and post-Volcker period. And last but not least, we provide an empirical estimate of the deep parameter that characterizes a central bank’s loss function. On average, the weight that the Fed attaches to the volatility of output gap relative to the volatility of (annualized) inflation is roughly 0.2.
What is the relative weight a central bank attaches to the allocation of the output gap and inflation? The answer to this question is reflected in the optimal targeting rule. Empirical estimation of this measure is challenging due to the simultaneity problem. We overcome this issue by purging out shocks that shift a central bank’s targeting rule. The following results stand out. First, the Fed is willing to decrease the growth rate of the output gap by two and half percentage points for each percentage point increase in (quarterly) inflation. Second, the data rejects optimal discretionary policy in favor of optimal policy under commitment. Third, the Fed’s targeting rule was not much changed during the pre-Volcker and post-Volcker period. And last but not least, we provide an empirical estimate of the deep parameter that characterizes a central bank’s loss function. On average, the weight that the Fed attaches to the volatility of output gap relative to the volatility of (annualized) inflation is roughly 0.2.
5. Assessing Macroeconomic Tail Risk with Francesca Loria and Christian Matthes, April 2019 FEDS Working Paper
What drives macroeconomic tail risk? To answer this question, we borrow a definition of macroeconomic risk from Adrian et al. (2019) by studying (left-tail) percentiles of the forecast distribution of GDP growth. We use local projections (Jordà, 2005) to assess how this measure of risk moves in response to economic shocks to the level of technology, monetary policy, and financial conditions. Furthermore, by studying various percentiles jointly, we study how the overall economic outlook—as characterized by the entire forecast distribution of GDP growth—shifts in response to shocks. We find that contractionary shocks disproportionately increase downside risk, independently of what shock we look at.
What drives macroeconomic tail risk? To answer this question, we borrow a definition of macroeconomic risk from Adrian et al. (2019) by studying (left-tail) percentiles of the forecast distribution of GDP growth. We use local projections (Jordà, 2005) to assess how this measure of risk moves in response to economic shocks to the level of technology, monetary policy, and financial conditions. Furthermore, by studying various percentiles jointly, we study how the overall economic outlook—as characterized by the entire forecast distribution of GDP growth—shifts in response to shocks. We find that contractionary shocks disproportionately increase downside risk, independently of what shock we look at.
Selected Works in Progress:
1. The Ambiguous Signaling Channel of Monetary Policy
3. Inflation Stabilization Policy with Heterogeneous Beliefs with Renbin Zhang
5. Optimal Monetary Instruments for an Uncertain World (with Isaac Baley and Renbin Zhang)