Understanding the Aggregate Effects of Credit Frictions and Uncertainty

Abstract

We examine the interaction of uncertainty and credit frictions in a New Keynesian framework. To do so, uncertainty is modeled as time-varying stochastic volatility-the product of monetary policy uncertainty, financial risk (micro-uncertainty), and macrouncertainty. The model is solved using a pruned third-order approximation and estimated by the Simulated Method of Moments. We find that: 1) Micro-uncertainty aggravates the information asymmetry between lenders and borrowers, worsens credit conditions, and has first-order effects on real economic activity. 2) When credit conditions are poor, as indicated by elevated credit spreads, additional micro-uncertainty shocks produce even larger real effects. 3) Poor credit conditions notably affect the transmission mechanism of monetary policy amplifying the real effects of monetary shocks while mitigating the economic boost from TFP shocks. 4) While macro-uncertainty and policy uncertainty exert relatively little direct impact on aggregate economic activity, policy uncertainty accounts for around 40% of the business cycle volatility by affecting the size of monetary policy shocks in the presence of nominal rigidities.

A Generalized Time Iteration Method for Solving Dynamic Optimization Problems with Occasionally Binding Constraints

“…of poliy funtion itertion @psA sed tehniquesF 4 e ompre the speed nd ury gins from qs to those of the stndrd improvement lgorithm disussed in rowrd @IWTHAF he stndrd ps is high threshold to pss in terms of oth speed nd uryF st mkes use of eh new omputed poliy funtion fter onsidering the vlue of using tht poliy foreverD usully tking smller numer of steps for onvergene thn the stndrd ps @ee vjungquist nd rgent @PHIPA nd ntos nd ust @PHHQAAF hile we otin omprle results on uryD signi(nt di'erenes rise in the implementtion of qs nd psD lthough oth methods re sed on (xedEpoint itertionF sn prtiulrD ps tends to su'er from the urse of dimensionlityD wherey qs provides notle speed dvntgesF elsoD unlike iuler equtionE sed tehniques @suh s in den rn nd wret @IWWHAAD onvergene is hieved under qs without n eduted guess for the poliy funtion of future pitl holdingsF 5 yur min illustrtion of qs is sed on n irreversile investment model with lorEleisure hoieD where the dditionl stti hoie vrile produes more omplex solutions nd the ene(ts of the fster qs solution lgorithm re more pprentF 6 yther useful pplitions inlude the potentil use of this solution method in workhorse heterogeneousEgent mroeonomi models tht llow for inomplete mrketsF es more generl pplitionD we inlude twoEountry model in the spirit of eiygri @IWWRA nd with progressive lor inome txesF his model extends the one in uukçuo §lu @PHIUA y inluding lorEleisure hoieF he speed nd ury results of qs n e seen further in this exeriseF en erlier pplition in reltively simple fewleyEtype eonomy ws studied y querrieri nd vorenzoni @PHIUAF yur (nl omment pertins to the omprtive performne of the qs in reltion to grroll @PHHTA9s psE sed methodsF hen pplying his pprohD grroll @PHHTA inorportes liquidity onstrints in the sene of lorEleisure hoieF frills nd pernndezEillverde @PHHUA generlize grroll @PHHTA9s endogenous grid point method in n fg model with lorEleisure hoie @leit with no inequlity onstrintsA nd ompre the performne of the generlized iqw ginst the stndrd psD doumenting its dvntges in terms of speedF elthough the ps is nturl enhmrk in their seD it is not very di0ult enhmrk to outperform in terms of speed s the urse of dimensionlity is n even more serious prolem with psF he urrent workD thereforeD lso ontriutes to the literture y doumenting the dvntges of time itertionD prtiulrly in terms of speedD in the presene of multiple hoie vriles nd osionlly inding onstrints"n issue whih hs not een ddressedF…”

Understanding the Aggregate Effects of Credit Frictions and Uncertainty: Supplementary Materials and Additional Results

“…We also present an overview of the data used to estimate the model as well as a detailed explanation of the codes developed for the estimation and for the quantitative simulation of the pruned third-order approximation of the model's solution. Finally, this paper also contains an additional discussion of the main results inferred from the estimated model as well as a comprehensive set of experiments not reported in Balke et al (2017) but conducted to establish the robustness of our …ndings.…”

“…This paper complements the work of Balke et al (2017). We describe our approach to model mean-preserving stochastic volatility in detail, provide a derivation of the optimal nominal risky debt contract between entrepreneurs (borrowers) and …nancial intermediaries (lenders), and include a brief description of the equilibrium conditions and steady state equations that characterize the solution of the Balke et al (2017) model. We also present an overview of the data used to estimate the model as well as a detailed explanation of the codes developed for the estimation and for the quantitative simulation of the pruned third-order approximation of the model's solution.…”

“…The framework of Balke et al (2017) employs a medium-scale Dynamic New Keynesian model with credit frictions (asymmetric information and costly monitoring as in Bernanke et al (1999)) and uncertainty (stochastic volatility as in Fernández-Villaverde et al (2010)). In Balke et al (2017), we study speci…cally how …nancial frictions and micro-uncertainty a¤ect our understanding of the monetary policy transmission mechanism through the lens of the Dynamic New Keynesian model. We also investigate the interaction between credit frictions and macro/policy and micro-uncertainty explicitly.…”

“…The inverse relationship between the endogenous external …nance premium and the nominal interest rate found in Balke et al (2017) suggests that low nominal short-term interest rates tend to be associated with periods of high spreads (due to high …nancial risk or micro-uncertainty). Hence, our research shows that periods of high …nancial risk (or microuncertainty) like those experienced during the Great Recession of 2008 2009 tend to be associated with high endogenous credit spreads and low interest rates.…”