Modeling Financial Imperfections JEM027 - Monetary Economics

(1)

Modeling Financial Imperfections

JEM027 - Monetary Economics

Simona Malovan ´a

simona.malovana@gmail.com

Institute of Economic Studies, Charles University, Prague December 14, 2015

Partly based on presentation of T. Holub from 2013

(2)

Motivation

The recent financial and economic crisis has increased the interest in studying financial market imperfections

Interest rate spreads increased, the availability of credit has declined, asset prices dropped, etc.

Modern macroeconomics had not neglected these topics

But they were typically abstracted from in canonical general equilibrium models (and core models used for the MP)

Recently an effort to change this in order to be able to model the credit transmission channel and the negative feedback loop“ between financial sector and real economy (incl. research at the CNB)

(3)

Perfect & imperfect financial markets

Perfect FM Imperfect FM

Information Symmetric and costless Asymmetric and costly (moral hazard, adverse selection) Competition Perfect (no significant impact on

market price)

Imperfect Transaction

cost

No Yes (consequences – brokers,

middlemen and transaction taxes)

Interest rate &

lending conditions

Just one known with certainty

& all lend/borrow at the same terms

Many & credit rationing, collateral constraints, debt ceilings etc.

Transmission Interest rate and exchange rate channels

+ asset price channel, credit channel, risk-taking channel ...

”The deteriorating of credit market conditions is not simply a passive reflection of a declining real economy but is itself a major factor depressing the

(4)

Models with financial imperfections

Micro-foundations of the imperfections

I Imperfect enforcement of contracts

I Asymmetric information (moral hazard, adverse selection)

Macro-models with ad hoc assumptions about imperfections, studying their implications

I Static models (e.g. Krugman, 1999 - 3rd generation model)

I Dynamic models (e.g. DSGE models)

(5)

Imperfect enforcement & sovereign risk

Figure:10Y government bond yields

(6)

Adverse selection & moral hazard

Adverse selection

I hiddeninformationproblem: one side of the market has information about its own quality that other side does not have

I Ex.: quality of goods (lemons problem, Akerlof, 1970), insurance Moral hazard

I hiddenactionproblem: one side of the market has information about its actions/effort that other side does not have

I Krugman:”...any situation in which one person makes the decision about how much risk to take, while someone else bears the cost if things go badly.”

I Ex.: bail-outs / TBTF

(7)

DSGE models

DSGE models may be in central banks used to:

(1) help to identify sources of fluctuations

(2) forecast and predict the effect of policy changes (3) perform counterfactual experiments

(4) communicate monetary policy decisions (5) support/complement stress testing

(6) explore the feedback effects between real economy and financial sector

Would this be possible in model without financial frictions?

(8)

DSGE models with financial frictions

External finance premium

I The price of credit depends on net worth

I Bernanke, Gertler (1989); Bernanke, Gertler, Gilchrist (1999); Christiano, et al. (2003); Beneˇs, Kumhof (2011)

Collateral constraints

I Volume limits on borrowing

I Hart, Moore (1994); Kiyotaki, Moore (1997); Kocherlakota (2000); Iacoviello (2005); Iacoviello, Neri (2010)

Models with banks / financial sector

I Bank lending channel as well as balance sheet channel

I Existence of multiple interest rates

I Goodfriend, McCallum (2007); C ´urdia, Woodford (2009); Christiano, Trabandt, Walentin (2007); Gerali et al. (2009; 2010)

(9)

DSGE models with financial frictions

No model or stream of research has won the race yet.

Each stream focuses on some aspects of financial imperfections.

Each has its strengths and weaknesses and/or robustness issues.

Difficult to put all frictions into one model (would be too complicated - black box).

Research thus develops along these competing lines.

Models so far used for simulation purposes rather than for forecasting and MP decision-making.

(10)

Transmission channels

Conventional MP transmission - MP tools affect aggregate demand:

↓i →↑ AD

Credit channel - MP tools affects the availability of credit:

↓i →↑ credit supply/demand

Credit channel usually enhances the conventional channel

I Important credit market - greater impact of MP

I Emerging markets - credits are more cyclical - interest rate should react less

Risk-taking channel - in the LR MP easing encourages to higher risk-taking:

↓i →↑ risk-taking

(11)

Credit channel

CB can affect the amount of credit that banks issue to firms and consumers for purchases, which in turn affects the real economy.

Not an alternative to conventional“ transmission channels (IR, ER), it amplifies and enhances existing transmission mechanism

Monetary policy changes the external finance premium Works through two channels

I The balance sheet channel

I The bank lending channel

(12)

Balance sheet & bank lending channel

Balance sheet channel

I Monetary policy affects the net worth of the borrower (liquid assets minus liquid liabilities)

I Decline in net worth worsens the external finance premium

I Firms that have a higher level of liquidity are less affected than those that have lower levels

Bank lending channel

I Monetary policy affects the supply of bank credit

I Closing bank credit increases the external finance premium

I Firms dependent on bank financing are constrained by the implicit higher cost of credit

⇒Implication: the availability of credit has short-run real output effect

(13)

Credit channel

(14)

DSGE models with financial frictions

(15)

External finance premium

Bernanke, Gertler and Gilchrist (1999) - BGG→financial accelerator mechanism

Currently the most used approach

Friction is placed on a non-financial side of the economy (firms) Firms need to borrow funds (external finance) to be able to engage in investment opportunities

External finance premium (EFP)- difference between firms’ cost of raising capital internally and externally

EFP depends inversely on the net worth of the firm

An endogenous development in balance sheet positions of borrowers can significantly amplify (accelerate) shocks

(16)

Bernanke, Gertler, Gilchrist (1999)

Figure:Br ázdik, Hlav áˇcek, Marˇs ál (2011)

(17)

DSGE models with financial frictions

(18)

Collateral constraint

EFP:limit in terms of price of funds (costs) vs.CC:volume limit on borrowing

Collateral - protection of lender against a default of borrower

The collateral value determines the amount of the loan the borrower can get

CC helps to explain observedasymmetryof the business cycle fluctuations (which EFP is not able):

I High collateral value - CC is slack - low sensitivity of borrowing and spending to changes in collateral value

I Low collateral value - CC is tight - high sensitivity of borrowing and spending to changes in collateral value

(19)

Kiyotaki and Moore (1997)

(20)

Iacoviello and Neri (2010)

CC drives an asymmetry between house prices and economic activity→ central mechanism to explain the collapse of the Great Recession

(21)

CC & EFP comparison

Brzoza-Brzezina, Kolasa, Makarski (2011)

Both models add volatility to NK framework, bringing it closer to data

Both models dampen the impact of productivity shocks and amplify the impact of monetary policy shocks on investment and output

Both models are superior to the simple NK framework without frictions

Business cycle properties of the EFP framework (BGG) is

more in line with empirical evidence than CC (Kiyotaki and

Moore, 1997)

(22)

CC & EFP comparison

MP shock

Figure:Brzoza-Brzezina, Kolasa, Makarski (2011)

(23)

DSGE models with financial frictions

(24)

Models with banks

CC & EFP: focused on the credit demand vs. banking sector: focused on the credit supply

Motivated by the aim of explaining specific banks’

behavior/role of banks in the financial crisis

Allow modeling of interest rate setting behavior & the

heterogeneity in the adjustment of bank rates to changing

conditions in money market interest rates

(25)

Christiano, Trabandt, Walentin (2011)

Figure:Ryˇs ´anek, Tonner, Vaˇs´ıˇcek (2011)

(26)

Christiano, Trabandt, Walentin (2011)

Figure:Christiano, Trabandt, Walentin (2011)

Financial shock is pivotal for explaining fluctuations in investment and GDP (Swedish data, 2006–2010)

It contributed substantially to the economic downturn

(27)

Adverse Financial Shock

CTW estimated on Czech data

(28)

Financial frictions and CB´s policy

Monetary policy

I Financial frictions can improve thefitof DSGE models and itsforecasting performance,

I the calibrationof the relevant coefficients and elasticities,

I understanding and quantification of thetransmission mechanismsof monetary policy to the real economy.

Macroprudential policy

I Theory - identifying, monitoring and assessing risks to financial stability, contribute to the resilience of the financial system

I Practice - the application of a set of instruments (e.g. capital charges) to reduce the vulnerability of the financial system

I DSGE models with frictions useful tocreate stress testing scenarios(can improve the fit of the satellite models),

I improve modeling of thegradual build-up of risksduring a ”boom and bust” financial cycle

(29)

Stress testing

Macroprudential tool for assessing the resilience of the banking sector to potential adverse shocks

Two scenarios – baseline and alternative

I Baseline – the most likely scenario of future macro development, the official central bank forecast (CB’s official prediction model, e.g. DSGE)

I Alternative - based on the identification of risks to the economy in the near future based on the CB’s official prediction model

Advantages of using DSGE for stress-testing

I DSGE models usually include many sources of volatility (shocks), allowing alternative trajectories to be modeled and

I ..rely on calibrated parameters, which can be altered to the match the requirements of the stress scenario

(30)

Stress testing & satellite models

(31)

Stress testing & satellite models

Satellite models with financial frictions are useful for

I

creating more realistic stress-testing scenarios,

I

better estimation of the satellite models,

I

including the feedback effects between the real economy and financial variables (e.g. the link between investment and loans to the private sector)

Models with financial sectors could replace satellite models

(32)

Summary

The crisis has increased the focus on financial imperfections Imperfect enforcement of contracts; imperfect info→moral hazard, adverse selection, etc.

The imperfections have important effects on allocation of capital (i.e. they reduce efficiency)

Balance sheets, cash-flow and asset prices important for the level of investment and consumption

Shocks to financial wealth may thus have important real consequences, with a negative feedback loop

Credit channel of monetary policy transmission, DSGE models with financial frictions

(33)

Modeling Financial Imperfections JEM027 - Monetary Economics