Research

Asymmetries in the Interest Rate Dynamics (2018)

In this paper I document that during 1984-2007 decreases of the interest rate in the US are less frequent but larger in absolute terms that interest rate rises. I then analyze if the observed asymmetry in the interest rate dynamics can be explained by asymmetric reaction function of the central bank. I introduce a flexible formulation of the Taylor rule which allows for a state-dependent asymmetry in its coefficients into a New Keynesian DSGE model. I then estimate the second-order approximation of the solution of the DSGE with the help of particle filter. The results indicate significant asymmetries in the reaction function of the US central bank. Reaction to inflation is stronger when inflation is higher than the target and reaction to the output growth is stronger when the output growth rate is below the target. I further show that the identified nonlinearities alone cannot fully explain why interest rate adjustments are asymmetric. Interest rate dynamics in the model fails to pass a second-order accuracy test that was suggested by Aruoba, Bocola, Schorfheide (2017).JEL Codes: C11, E32, E43

Nonlinear Inflation Forecasting with Recurrent Neural Networks (2018) with N. Andresen

This paper demonstrates that machine learning techniques can be used to eﬀectively forecast macroeconomic time series. We show that recurrent artiﬁcial neural networks outperform linear autoregressive (AR) and random walk (RW) models in forecasting monthly US CPI inﬂation. Performance of a simple fully-connected neural network (NN) is on a par with the linear AR model when the later is trained with regularization and superior to the RW while a long short-term memory (LSTM) recurrent neural network has a higher forecasting power than both the RW and AR models. At all horizons root mean squared forecast error (RMSFE) of the LSTM is approximately one third of the RMSFE of the RW. We also show that rolling-window real-time forecasts of the LSTM are signiﬁcantly more accurate to the AR and NN forecasts. Additionally we conduct a sensitivity analysis with respect to hyper-parameters and provide a qualitative interpretation of what the networks learn by applying a novel layer-wise relevance propagation technique. JEL Codes: C45, C53, E37

Working Papers

A Monetary Model of Blockchain (2018)

IRTG 1792 Discussion Paper 2018-008, 2018

This paper builds a model of the economy with a single generally acepted blockchain-based currency. In the spirit of the search and matching literature I use a matching function to model the operation of the blockchain. The formulation of the money demand is taken from a workhorse of monetary economics - Lagos and Wright (2005). I show that in a blockchain-based monetary system money demand features a precautionary motive which is absent in the standard Lagos-Wright model. Due to this precautionary money demand the monetary equilibrium can be stable for some calibrations. I also used the developed model to study how the equilibrium return on money is dependent on the blockchain parameters such as mining costs and rewards.

A Note on Cryptocurrencies and Currency Competition (2018)

IRTG 1792 Discussion Paper

In this paper I focus on two such features. First, cryptocurrencies operate according to a protocol - a blockchain - and are, therefore, free from the time-inconsistency problem. Second, the operation of the blockchain costs real resources. I use the Lagos-Wright search theoretic monetary model augmented with privately issued currencies as in Fernandez-Villaverde and Sanches (2016) and extend it by linear costs of private currency circulation. I show that in contrast to Fernandez-Villaverde and Sanches (2016) cryptocurrency competition 1) does not deliver price stability and 2) puts downward pressure on the inflation in the public currency only when the costs private currency circulation (mining costs) are sufficiently low.

Social Security Contributions and the Business Cycle (2017) with S. Voigts and M. Burda

C.E.P.R. Discussion Paper https://ideas.repec.org/p/cpr/ceprdp/12096.html

This paper examines magnitudes and business cycle dynamics of social security contributions (SSC). In most OECD countries studied, we document a negative covariation of payroll tax burdens with GDP and GDP growth at business cycle and lower frequencies. We assess the overall magnitude of the distortion following Barro and Redlick (2011). For most countries, average marginal SSC tax rates exceed average rates, but the latter tracks the former tightly. Changes in average payroll tax burdens are mostly accounted for by changes in tax schedules rather than shifts in the earnings distribution over time. For many countries, SSC rates behave like estimated values of the “labor wedge” (Chari et al. 2007, Brinca et al., 2016).

Labor Market Frictions and Monetary Policy Design (2016)

CRC 649 Discussion Paper

This paper estimates a New Keynesian DSGE model with search frictions and monetary rules augmented with different labor market indicators. In accordance with a theoretical literature I find that a central bank reacts to a labor market tightness, employment or unemployment. Posterior odds tests speak in favor of models with augmented Taylor rules versus a model with a model with a standard rule. The augmented rules were also shown to be more efficient in terms of welfare.