Research

Publications

Abstract: I propose an approach to quantify attention to inflation and show that attention declined after the Great Inflation period. This decline in attention has important implications for monetary policy as it renders managing inflation expectations more difficult and can lead to inflation-attention traps: prolonged periods of a binding lower bound and low inflation due to slowly-adjusting inflation expectations. As attention declines the optimal policy response is to increase the inflation target. The lower bound fundamentally changes the normative implications of declining attention: lower attention raises welfare absent the lower-bound constraint, whereas it decreases welfare when accounting for the lower bound. 

Awards:


Working Papers

Abstract: At the outbreak of the recent inflation surge, the public’s attention to inflation was low but increased quickly once inflation started to rise. In this paper, I quantify when and by how much the public's attention to inflation changes. I estimate an attention threshold at an inflation rate of 4%, and that attention doubles when inflation exceeds this threshold. Adverse supply shocks become twice as inflationary in times of high attention, and the increase in people's attention to inflation in 2021 accounts for more than half of the subsequent supply-driven inflation. I develop a model accounting for the attention threshold and show that shocks that are usually short lived lead to a persistent surge in inflation if they induce an increase in people's attention. The attention threshold further lengthens the last mile of disinflation after an inflation surge, and it leads to an asymmetry in the dynamics of inflation.


Abstract: We develop a heterogeneous agent New Keynesian model in which households' expectations underreact to aggregate news and in which they are unequally exposed to business cycles, both consistent with the data. Unlike existing models, our model simultaneously accounts for the empirical findings that monetary policy affects consumption largely through indirect effects, income risk is countercyclical conditional on monetary policy shocks, and forward guidance is less powerful than contemporaneous monetary policy. While after persistent monetary policy shocks, heterogeneous exposure and households' underreaction counteract each other, supply shocks are unambiguously amplified, offering a theory how supply shocks can trigger large inflation spikes.

Awards:

Abstract: Business cycle models often abstract from persistent household heterogeneity, despite its potentially significant implications for macroeconomic fluctuations and policy. We show empirically that the likelihood of being persistently financially constrained decreases with cognitive skills and increases with overconfidence thereon. Guided by this and other micro evidence, we add persistent heterogeneity in cognitive skills and overconfidence to an otherwise standard HANK model. Overconfidence proves to be the key innovation, driving households to spend instead of precautionary save and producing empirically realistic wealth distributions and hand-to-mouth shares and MPCs across the income distribution. We highlight implications for various fiscal policies.

Abstract: We study how the availability of data in modern economies shapes the propagation of cyclical fluctuations and the effectiveness of monetary policy along the business cycle. We consider a tractable heterogeneous firms framework in which data enters invest- ment decisions by favorably affecting a firm’s productivity distribution and by allowing firms to predict their future productivity realizations. Data accumulates endogenously through a data feedback loop, i.e., firms that produce more accumulate more data. We show that increased availability of data dampens cyclical fluctuations if and only if the data feedback loop is sufficiently weak, that is, when firms accumulate only little data through production. This is because data-rich firms respond less strongly to aggregate productivity shocks when the data feedback loop is weak, while the converse holds true if the data feedback loop is strong. Given that data-rich firms respond more strongly to monetary policy, this result also makes the effectiveness of monetary policy counter- cyclical if the data feedback loop is weak and vice versa. Moreover, the data feedback loop weakens the negative relationship between a firm’s risk sensitivity and its size, which amplifies the effects of aggregate uncertainty shocks. Our work also sheds light on the macroeconomic effects of digital markets regulation such as the EU GDPR.

Abstract: U.S. households’ housing price expectations deviate systematically from full-information rational expectations: (i) expectations are updated on average too sluggishly, (ii) expectations initially underreact but subsequently overreact to housing price changes, and (iii) households are overly optimistic (pessimistic) about housing price growth when the price-to-rent ratio is high (low). We show that weak forms of housing price growth extrapolation allow to simultaneously replicate the behavior of housing prices and these deviations from rational expectations as an equilibrium outcome. Embedding housing price growth extrapolation into a sticky price model with a lower-bound constraint on nominal interest rates, we show that lower natural rates of interest increase the volatility of housing prices and thereby the volatility of the natural rate of interest. This exacerbates the relevance of the lower bound constraint and causes Ramsey optimal inflation to increase strongly with a decline in the natural rate of interest.