News Shocks and Asset Price Volatility in a DSGE Model

Similar documents
News Shocks and Asset Price Volatility in a DSGE Model

New Shocks and Asset Price Volatility in General Equilibrium

Risks for the Long Run: A Potential Resolution of Asset Pricing Puzzles

Exchange Rates and Fundamentals: A General Equilibrium Exploration

Monetary Economics Final Exam

Estimating Macroeconomic Models of Financial Crises: An Endogenous Regime-Switching Approach

Collateralized capital and news-driven cycles. Abstract

1 Explaining Labor Market Volatility

Collateralized capital and News-driven cycles

Business Cycles and Household Formation: The Micro versus the Macro Labor Elasticity

STATE UNIVERSITY OF NEW YORK AT ALBANY Department of Economics. Ph. D. Comprehensive Examination: Macroeconomics Fall, 2010

The Transmission of Monetary Policy through Redistributions and Durable Purchases

Uncertainty Shocks In A Model Of Effective Demand

Labor Economics Field Exam Spring 2011

Heterogeneous Firm, Financial Market Integration and International Risk Sharing

Explaining International Business Cycle Synchronization: Recursive Preferences and the Terms of Trade Channel

On Quality Bias and Inflation Targets: Supplementary Material

Sentiments and Aggregate Fluctuations

Risky Mortgages in a DSGE Model

Oil Price Uncertainty in a Small Open Economy

Inflation Dynamics During the Financial Crisis

TFP Persistence and Monetary Policy. NBS, April 27, / 44

State-Dependent Pricing and the Paradox of Flexibility

Asymmetric Information: Walrasian Equilibria, and Rational Expectations Equilibria

Optimal monetary policy when asset markets are incomplete

Behavioral Theories of the Business Cycle

1 Answers to the Sept 08 macro prelim - Long Questions

1 Dynamic programming

LECTURE NOTES 3 ARIEL M. VIALE

Inflation in the Great Recession and New Keynesian Models

Eco504 Spring 2010 C. Sims FINAL EXAM. β t 1 2 φτ2 t subject to (1)

International Macroeconomics and Finance Session 4-6

Optimal Monetary Policy Rules and House Prices: The Role of Financial Frictions

Comment. The New Keynesian Model and Excess Inflation Volatility

INTERTEMPORAL ASSET ALLOCATION: THEORY

The Transmission of Monetary Policy Operations through Redistributions and Durable Purchases

STATE UNIVERSITY OF NEW YORK AT ALBANY Department of Economics. Ph. D. Comprehensive Examination: Macroeconomics Fall, 2016

One-Period Valuation Theory

Habit Formation in State-Dependent Pricing Models: Implications for the Dynamics of Output and Prices

Sentiments and Aggregate Fluctuations

Asset Pricing and Equity Premium Puzzle. E. Young Lecture Notes Chapter 13

Preference Shocks, Liquidity Shocks, and Price Dynamics

Not All Oil Price Shocks Are Alike: A Neoclassical Perspective

Estimating a Dynamic Oligopolistic Game with Serially Correlated Unobserved Production Costs. SS223B-Empirical IO

Equilibrium Yield Curve, Phillips Correlation, and Monetary Policy

Exercises on the New-Keynesian Model

Consumption and Asset Pricing

Problem Set: Contract Theory

Moral Hazard: Dynamic Models. Preliminary Lecture Notes

Macroeconomics I Chapter 3. Consumption

Credit Shocks and the U.S. Business Cycle. Is This Time Different? Raju Huidrom University of Virginia. Midwest Macro Conference

Prospect Theory and Asset Prices

Financial Economics Field Exam January 2008

ECON 6022B Problem Set 2 Suggested Solutions Fall 2011

Asset Pricing in Production Economies

A Model with Costly-State Verification

Simple Analytics of the Government Expenditure Multiplier

The Risky Steady State and the Interest Rate Lower Bound

Currency Manipulation

GHG Emissions Control and Monetary Policy

A Macroeconomic Framework for Quantifying Systemic Risk. June 2012

WORKING PAPER NO THE ELASTICITY OF THE UNEMPLOYMENT RATE WITH RESPECT TO BENEFITS. Kai Christoffel European Central Bank Frankfurt

... Monetary Policy and a Stock Market Boom-Bust Cycle. Lawrence Christiano, Roberto Motto, Massimo Rostagno

Macroeconomics 2. Lecture 6 - New Keynesian Business Cycles March. Sciences Po

Fiscal and Monetary Policies: Background

DSGE model with collateral constraint: estimation on Czech data

Information Globalization, Risk Sharing and International Trade

Optimal Devaluations

Real Exchange Rates and Primary Commodity Prices

slides chapter 6 Interest Rate Shocks

14.461: Technological Change, Lecture 10 Misallocation and Productivity

Stock Prices and the Stock Market

Adjustment Costs, Agency Costs and Terms of Trade Disturbances in a Small Open Economy

EXAMINING MACROECONOMIC MODELS

The Basic New Keynesian Model

Asset Pricing with Heterogeneous Consumers

Comprehensive Exam. August 19, 2013

Problem Set: Contract Theory

The design of the funding scheme of social security systems and its role in macroeconomic stabilization

A unified framework for optimal taxation with undiversifiable risk

Lecture 2. (1) Permanent Income Hypothesis. (2) Precautionary Savings. Erick Sager. September 21, 2015

Country Spreads as Credit Constraints in Emerging Economy Business Cycles

Uninsured Unemployment Risk and Optimal Monetary Policy

Introduction Some Stylized Facts Model Estimation Counterfactuals Conclusion Equity Market Misvaluation, Financing, and Investment

Chapter 9, section 3 from the 3rd edition: Policy Coordination

Vayanos and Vila, A Preferred-Habitat Model of the Term Stru. the Term Structure of Interest Rates

Idiosyncratic risk, insurance, and aggregate consumption dynamics: a likelihood perspective

Economics 430 Handout on Rational Expectations: Part I. Review of Statistics: Notation and Definitions

John H. Cochrane. April University of Chicago Booth School of Business

On the Optimality of Financial Repression

Problem set 5. Asset pricing. Markus Roth. Chair for Macroeconomics Johannes Gutenberg Universität Mainz. Juli 5, 2010

Information Processing and Limited Liability

Roy Model of Self-Selection: General Case

Reading the Tea Leaves: Model Uncertainty, Robust Foreca. Forecasts, and the Autocorrelation of Analysts Forecast Errors

Signal or noise? Uncertainty and learning whether other traders are informed

Stock Price, Risk-free Rate and Learning

Sampling and sampling distribution

Asset Pricing under Information-processing Constraints

ECON 815. A Basic New Keynesian Model II

Mortgage Amortization and Amplification

Transcription:

News Shocks and Asset Price Volatility in a DSGE Model Akito Matsumoto 1 Pietro Cova 2 Massimiliano Pisani 2 Alessandro Rebucci 3 1 International Monetary Fund 2 Bank of Italy 3 Inter-American Development Bank January 23, 2010 The views expressed in this paper are those of the authors and should not be attributed to the Bank of Italy, the Inter-American Development Bank, the International Monetary Fund, their Executive Boards, or their managements.

Motivation Asset prices seem unrelated to the current fundamentals and are very volatile. Cochrane (1994) attributes our (economists ) lack of understanding of business cycles to shocks unobservable to econometricians but observable to agents News Shocks Recently business cycle models with news shocks gained a lot of attention and achieved some successes. News Shocks may help explain asset price puzzles. e.g. Link between current fundamental and asset price becomes weaker when agents receive news about future. Related Literature Business Cycle with News Shocks: Beaudry and Portier(2004, 2007), Jaimovich and Rebelo(2009), Schmitt-Grohe and Uribe (2008), Christiano, Ilut, Motto, and Rostagno (2007), News Shocks and Asset Prices: Beaudry and Portier(2006), West (1988). Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 2/24

Motivation However, West (1988) shows that in a present discounted value model (with constant discounting)[pvm], the conditional variance of asset prices is less volatile if agents have more information than the history of cash flows given the stochastic process of underlying cash flows. that is, more information would reduce asset price volatility in a partial equilibrium setting. So introduction of news shocks may make asset price puzzle worse on the volatility front though disconnecting the asset prices from current fundamentals. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 3/24

This Paper Proposes a way to introduce news shocks in a present discount value model that keeps asset price volatility high relative to fundamentals. In a PVM, while introducing news always reduces (or keeps) the conditional variance of asset prices, it does not imply that the conditional variance of asset prices relative to underlying cash flow has to be low. 1 Relative to the world without news 2 Relative to the cash flow process One can use the same mechanism of the magnification effects literature, e.g., Frenkel (1976), when one introduces news shocks. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 4/24

This Paper More importantly, this paper shows that in GE, introduction of news may increase asset price volatility relative to the world without news, i.e. the same stochastic processes but agents do not observe news. Does not try to match the asset price volatility in the data. We provide a simple counterexample to show the conjecture more information will always reduce asset price volatility, based on the West s theorem, cannot extend to a GE environment where more information alters agents behavior. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 5/24

Preview of Results In a present discounted value model with given cash flow process, we show that If news about future cash flow is correlated with current surprise then this cash flow process has serial correlation; Thus, this will keep asset price volatility high relative to underlying cash flow in a simple PVM or GE where magnification effect can be operative. We investigate how asset prices volatility is affected by introduction of news in a GE setting. We find that News about future productivity can change agents behavior, e.g., consumption, pricing and so on. The dividend process changes accordingly. Therefore, asset price volatility can be higher depending on the parameter values relative to the world without news. That is, providing more information does not necessarily reduce asset price volatility in GE. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 6/24

Outline Review of West s Theorem More information reduces asset price volatility in present discount value model. Results in PE Correlated news shocks can keep asset price volatility high relative to the cash flow. Results in GE Introduction of news can increase asset price volatility relative to the world without news. Conclusions Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 7/24

Theorem by West (1988) Let F t, be the linear space spanned by the current and past values of a finite number of random variables, with F t, a subset of F t+1 for all t. Let f t, be one of these variables. Let H t, and I t be subsets of F t consisting of the space spanned by current and past values of some subset of the variables in F t including at a minimum current and past values of f t. Let P( ) denote linear projections. PVM: x t (F t ) = If H t I t then, where β j P(f t+j F t ) where 0 < β < 1 j=0 Var(x t (I t ) I t 1 ) Var(x t (H t ) H t 1 ) Var(x t (F t ) F t 1 ) E[x t (F t ) P(x t (F t ) F t 1 )] 2 Matsumoto, (We Cova, use Pisani, this and Rebucci definition throughout News Shocks and Asset the Pricepaper.) January 23, 2010 8/24

Theorem by West (1988) Points: Any additional information, including news about future cash flow, can only reduce (never increase) the asset price volatility. However, it does not necessarily imply that the asset price volatility has to be lower than volatility of cash flow. Nonetheless, including news makes it harder to generate higher volatility in a partial equilibrium setting. Note that f t is given, i.e., cash flow is exogenous. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 9/24

Definition of news What is news? News (new information about future) is information that is useful to predict future value of some variables. In our paper, we label news about future as news shocks new information about today as current shocks. A lot of empirical papers on news are about new information about today, i.e. current shocks. We call z t a news about f t, if Var(f t+k H t, z t ) < Var(f t+k H t ) where H t is a linear space spanned by at a minimum a history of f t (current and past value). Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 10/24

Example: Introducing news shocks in PVM f t = f t 1 + ε t + z t 1 where (ε t, z t ) are jointly i.i.d. normal zero mean processes, with Var(ε t ) = σ 2 1, Var(z t) = σ 2 2 and Cov(ε t, z t ) = ϱσ 1 σ 2 and 1 < ϱ < 1. i.e., correlation between news shocks (z t ) and current shocks(ε t ) to be ϱ and Var( f t ) = σ 2 1 + σ2 2. z t = ϱ σ 2 σ 1 ε t + η t where η t is orthogonal to ε t and Var(η t ) = (1 ϱ 2 )σ 2 2. So, f t = f t 1 + ε t + z t 1 News = f t 1 + ε t + ϱ σ 2 ε t 1 + η t 1 σ 1 Delayed = f t 1 + θ t + ϱ θ θ t 1 Econometrician Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 11/24

Example: Asset Price Implication The asset price at time t conditional on I t is x t (I t ) = x t (H t ) = j=0 j=0 β j E(f t+j I t ) = 1 1 β f t + β 1 β z t β j E(f t+j H t ) = 1 1 β f t + β 1 β ϱσ 2 σ 1 ε t. Note that when agents observe news asset price depends not only on the current fundamental but also on the news shock z t. We might want to reconsider how to evaluate the asset price model empirically. See for instance Engel, Mark, and West (2008) on exchange rates. (vs. Meese and Rogoff) The same logic applies to equity as well. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 12/24

The Volatility of Asset Price ( ) 1 Var(x t (I t ) I t 1 ) = Var 1 β (ε t + βz t ) = (1 β) 2 ( (σ1 2 + 2βϱσ 1 σ 2 + β 2 σ2) 2 ) ( 1 Var(x t (H t ) H t 1 )) = Var 1 β (ε t + η t 1 + βϱ σ ) 2 ε t ) σ 1 = (1 β) 2 ( (σ 2 1 + 2βϱσ 1 σ 2 + [β 2 + (1 β 2 )(1 ϱ 2 )]σ 2 2) ). News reduces volatility of asset prices relative to the world w/o news. However, when news shocks are correlated (i.e. higher ϱ) with current shocks, then the asset price volatility can be high relative to the fundamental, which does not depend on ϱ. Magnification effects. Correlated news shocks disconnect asset prices from current fundamental while keeping the asset price volatility high. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 13/24

General Equilibrium: Main Results News about future productivity changes agents behavior. Therefore, asset price volatility can be higher than in the world without news, where the exogenous productivity process is the same but agents cannot observe news about productivity. That is, providing more information does not necessarily reduce asset price volatility relative to the world without news. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 14/24

Model Setup Households max C t(j),m t(j),l t(j) E t s=t C s (j) 1 ρ 1 ρ + κ 1 1 ε ( Ms (j) P s ) 1 ε κ 2 1 + ψ L s(j) 1+ψ, [ 1 λ/(λ 1) where C t (j) 0 C t(j, i) di] (λ 1)/λ, and s.t. budget constraints. Firms set price one period in advance to maximize expected profit and supply goods as demanded. Y t (i) = A t L t (i). Two exogenous variables. Money M t and productivity A t. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 15/24

Shocks ln A t a t = a t 1 + ν 1,t + ν 2,t 1 }{{} news ln M t m t = m t 1 + µ 1,t + µ 2,t 1 }{{} news We assume no correlation between current shocks and news shocks in the presentation. (see paper for correlated case.) News about future money stock can be regarded as monetary policy announcement. The known reaction to productivity can be also regarded as news (see paper.) Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 16/24

Solution Surprise to the equity price (equity: claim to the firm profit) q t+1 E t q t+1 =(Λ 1 + Λ 2 )ν 1,t+1 + Λ 2 ν 2,t+1 + Λ m [(1 β)ε(µ 1,t+1 ) + β(µ 1,t+1 + µ 2,t+1 )] where ζ Λ 1 (1 β) (ψ + 1) 1 ζ [ ( 1 ρ Λ 2 { 1 + (1 β) ρ [ ( 1 ρ Λ m 1 + (1 β) ρ ζ 1 ζ ζ 1 ζ ρ + ψ ρ )] ρ + ψ ρ (We will ignore money shocks for simplicity.) )] ( ρ 1 1 ) } + (1 ρ) β ψ + 1 ε ρ + ψ Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 17/24

With or Without News Shocks Equity price (claim to the firm profit) surprise With news q t+1 (I t+1 ) E(q t+1 (I t+1 ) I t ) = (Λ 1 + Λ 2 ) ν 1,t+1 +Λ 2 ν 2,t+1 }{{}}{{} current news Without news. When agents do not observe news, current surprise component is ν 1,t+1 + ν 2,t. q t+1 (H t+1 ) E(q t+1 (H t+1 ) H t ) = (Λ 1 + Λ 2 ) (ν 1,t+1 + ν 2,t ) Thus, the difference of conditional variances are Var(q t+1 (I t+1 ) I t ) Var(q t+1 (H t+1 ) H t ) = (Λ 1 + 2Λ 2 )Λ 1 σ 2 ν2. Sign of Λ 1 + 2Λ 2 is the key. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 18/24

With or Without News Shocks 2 Recall ζ Λ 1 (1 β) (ψ + 1) 1 ζ [ ( 1 ρ Λ 2 { 1 + (1 β) ρ ζ 1 ζ If ε = 1 and ψ = 0, then Λ 2 = (1 ρ)β 1 ρ. Λ 1 + 2Λ 2 = 2β 1 ρ ρ )] ( ρ + ψ ρ 1 1 ) } + (1 ρ) β ψ + 1 ρ ε ρ + ψ ζ + (1 β) 1 ζ So the equity price can be more volatile than without news when ρ (1 1 β ζ 2β 1 ζ ) 1. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 19/24

How big is the effect of news? 1 Asset Price Volatility standard deviation of projection error 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 with news without news 0 0 1 2 3 4 5 ρ :the cofficient of relative risk aversion ε = 1, ψ = 0, ζ = 2/3, β = 0.95 Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 20/24

How big is the effect of news? 1.4 log difference of asset price standard deciation 1.2 1 Higher Volatility w News 0.8 0.6 0.4 0.2 0 0.2 0.4 0 1 2 3 4 5 ρ ε = 1, ψ = 0, ζ = 2/3, β = 0.95 Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 21/24

Intuition Why extra information to agents can increase asset price volatility? When agents know about future, they behave differently. current shocks have two offsetting effects on equity return. With nominal rigidity, positive current shocks on productivity have redistribution effects as in Engel and Matsumoto (2009): increase current profit. But in the future, higher productivity reduce nominal profit because the price goes down as it is cheaper to produce: reduce future profit. news shocks have only second effects. Note that dividend process is not invariant to information set unlike the assumption in West s theorem. That s why one cannot extend the conjecture derived from West (1988) Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 22/24

Other Results in the Paper In the paper, We setup two country world to study exchange rates relative equity returns We find exchange rates volatility cannot be increased by the introduction of news with preset price as exchange rates looks like a simple present discount value model of relative money supply. But magnification effects can be used to induce higher volatility relative to fundamentals as shown in PE. the volatility of relative equity returns can increase with the introduction of news relative to the world without news. Similar to what we have shown. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 23/24

Conclusion In this paper, We argue that introduction of news shocks helps to understand disconnect between asset prices and current fundamentals. We show that correlated news shocks in a PVM generate higher asset price volatility relative to fundamentals. We show that introduction of news shocks does not necessarily reduce asset price volatility in a simple DSGE model. For future Realistic DSGE modeling. (e.g. leverage, more frictions) Relate news to unobservable stochastic discount factor. Matsumoto, Cova, Pisani, and Rebucci News Shocks and Asset Price January 23, 2010 24/24