|
Home CV Research |
Yuliya RychalovskaPostdoctoral ResearcherThe implications of financial frictions and imperfect knowledge in the estimated DSGE model of the US economy
I study how alternative assumptions about the expectation formation can modify the implications of financial frictions for the real economy. I incorporate financial accelerator mechanism in a version of Smets and Wouters (2007) DSGE model and perform a set of estimation and simulation exercises assuming, on the one hand, complete rationality of expectations and, alternatively, several learning algorithms that differ in terms of the information set used by agents to produce the forecasts. I show that implications of financial accelerator for the busyness cycle may vary depending on the process of modeling the expectations. The results suggest that the learning scheme based on small forecasting functions is able to amplify the effects of financial frictions relative to the model with Rational Expectations. Specifically, I show that the dynamics of real variables under learning is driven to a significant extent by the time variation of agents' beliefs about financial sector variables. During periods when agents perceive asset prices as being relatively more persistent, financial shocks lead to more pronounced macroeconomic outcomes. The amplification effect raises as financial frictions become more severe. At the same time, learning specification in which agents use more information to generate predictions (close to MSV learning) produces very different asset price and investment dynamics. In such a framework, learning cannot significantly alter the real effects of financial frictions implied by the Rational Expectations model.
The implications of sectoral heterogeneity for monetary policy and welfare in a small open economy: a linear quadratic framework In this paper I assess possible risks and challenges of implementing Inflation Targeting (IT) policy in more complicated, but at the same time more realistic, DSGE model economy. I study whether the optimal monetary policy has to account for sector-specific characteristics or can welfare be maximized following simple IT rules in a two-sector, small open economy model. In accordance with empirical evidence, it is assumed that sectors have different characteristics (sector-specific shocks, preference parameters, degree of nominal rigidities and thus may exhibit different inflation persistence). The findings suggest that openness to trade as well as sector-specific features do matter for monetary policy design thus generating important implications for optimal stabilization objectives and social welfare. I find that, in the open economy, it is optimal to stabilize the appropriately weighted inflation index even when prices in both sectors are equally sticky. In particular, the non-tradable sector inflation rate (as well as output) has to be relatively more stabilized under the optimal policy compared to its tradable sector counterpart. The welfare ranking of a number of simple rules indicates that flexible CPI targeting regimes may closely replicate the optimal solution and outperform the policy of domestic inflation stabilization. Moreover, the sensitivity analysis, which is focused on the parameters that determine the optimal policy to a significant degree, demonstrates that the presence of sectoral asymmetries may alter the relative performance of alternative policy rules. In particular, I study the impact of sectoral heterogeneity in the degree of price stickiness, the elasticity of substitution, and the degree of openness on the relative performance of policy rules with sector-specific and aggregate variables (CPI or domestic inflation rates). I find that welfare benefits from targeting "weighted" versus broader inflation index increase as the economy becomes more open and prices in the non-traded sector relatively stickier. In addition, it is welfare improving to weigh appropriately the sectoral inflation rates if the elasticity of substitution between home and foreign goods rises. On the contrary, as goods in the non-traded sector become relatively more elastic, the benefits from targeting the weighted inflation gradually vanish and policy of the domestic inflation stabilization approximates the optimal strategy rather well. Bayesian estimation of DSGE models under adaptive learning: robustness issues, with S.Slobodyan (CERGE-EI) This work is devoted to robustness issues of Bayesian estimation of DSGE models. The paper was motivated by recent tendency to use DSGE models for policy analysis and by large literature that highlights challenges in the modeling approach. In a joint paper with S. Slobodyan, we attempt to achieve the improvement in the model fit as well as to address the issue of possible DSGE models misspecifications by departing from the rational expectation (RE) hypothesis and incorporating more empirically realistic mechanism of public’s expectation formation - adaptive learning (AL). We add to answering the following questions: How restrictive is the RE hypothesis for estimated DSGE models of the Euro area? Can AL generate endogenous persistence at the same time reducing structural rigidities? In other words, we evaluate empirically the relative importance of several types of "frictions" - "mechanical" rigidities like habit formation, Calvo pricing, adjustment costs etc. versus learning. We study the robustness of the estimation results in several dimensions: by varying the model size, information set available to the learning agents, and the way of forming agents' initial beliefs. We find that assuming adaptive expectations results in better model fit than if RE are used, especially when the agents use very little information to form their beliefs. Estimated parameters and the model fit depend significantly on the information set used by the agents, which might explain widely divergent results of previous estimations under adaptive learning. We also find that different ways of forming the initial beliefs influence the dynamics of the model under learning. |