macro financial modeling in macro policy dsge models rprd ... · – dependency of firms on direct...

52
Macro Financial RPRD 802 Modeling in Macro Policy DSGE Models P t 2008 Stephen Snudden Post2008 Sept 28, 2016

Upload: others

Post on 16-Mar-2020

10 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

Macro Financial 

RPRD 802

Modeling in Macro Policy DSGE Models 

P t 2008Stephen Snudden Post‐2008 

Sept 28, 2016

Page 2: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

Objectives

• Overview of the types of models used at policy institutions

• History of financial frictions in macroeconomic policy DSGE models post 2008y p y p

• The financial crisis resulted in a shift in mentality: – Financial markets now matter for the economy!a c a a ets o atte o t e eco o y!– Policy models were seen to be severely misspecified

• An insiders scoop:An insiders scoop:– Why were certain models that were tried, discarded?– Which models are currently in use, and how are they lacking?

What properties do central bankers want in their financial sector models?– What properties do central bankers want in their financial sector models?

Page 3: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

What are Macro‐Policy DSGE Models?• All models are a simplification of reality 

– What should be in the core of the model versus kept in a satellite model?• Macro‐Policy DSGE Models:

– are sometimes used for forecasting – are structural and micro‐founded– used extensively for policy and scenario analysis

i i f h k / h i ( h i i ?)– contain important sources of shocks/ mechanisms (what is important?)• Pros:

– GE interactions of theories, quantitative rigorConcentration of resources ability to pull from academic circles– Concentration of resources, ability to pull from academic circles

– A ready to use model for a variety of potential policy questions• Con:

– Difficult to change structure large in sizeDifficult to change structure, large in size– Limited in the amount of detail

Page 4: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

What Makes Macro‐Policy‐DSGE Models Special?Other Types of Models Used in Policy:1. Semi‐Structural (or Non‐Structural) Empirical

– Used for forecasting/ projections, empirical validations– Pros: easy to change better forecasts– Pros: easy to change, better forecasts– Cons: hard to interpret, subject to Lucas critique– Example: PAC models such as MUSE, FRB‐US, VARs, DFMs

2. Specialized Policy ModelsU f l f ifi i– Useful for specific questions

– Pros: Better when acute detail needed– Cons: Resource consuming, few macro. GE effects (no open economy, etc)– Example: Models of the Canadian Banking Sector for FSAPs

3. Toy Theoretical – Useful for developing underlying theory, isolating key channels– Pros: Possibility to work with academic literature– Cons: Resource consuming few macro GE effects limited in policy usefulnessCons: Resource consuming, few macro. GE effects, limited in policy usefulness– Example: Most academic models which propose a theory 

Page 5: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

Examples of Macro Policy‐DSGE Models: 2008

Models Institutions Features Financial Frictions

BoC GEM/ Bank of Canada Global INF+LIQ Exog risk premiums: NFABoC‐GEM/ GEM

Bank of Canada/ IMF

Global, INF+LIQ, commodities,quarterly

Exog. risk premiums: NFA, sovereign debt.

Totem Bank of Canada SOE INF sector details ‐Totem Bank of Canada SOE, INF, sector details, quarterly

GIMF IMF Global, OLG+LIQ, detailed fiscal block annual

Exog. risk premiums: NFA, sovereign debtfiscal block, annual sovereign debt.

QUEST EuropeanCommission

Euro area – ROW, INF‐LIQ, fiscal, quarterly

Exog. risk premium NFACommission fiscal, quarterly

Others: ECB, NAWM; FED, SIGMA

Page 6: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

Questions to Begin

• Why do we care about financial frictions now?– versus the tech bubble– versus the east Asian crisis– versus a entire human history of financial related crises

• What would you devote your limited resources to have?What would you devote your limited resources to have?– type of financial channels to include – other important features: primary commodities, international linkages, labor 

• A need to take stock– look at the whole system/ history 

id if k h k / i h l– identify key shocks/ propagation channels– remain flexible/ forward looking as the system evolves

Page 7: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Modeling the Demand for Credit

Page 8: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. The Financial Accelerator à la BGG

• Been around for a while – well understood?B k d G l (1989)• Bernanke and Gertler (1989)

• Carlstrom and Fuerst (1997)• Bernanke, Gertler, and Gilchrist (1999)• Dib and Christensen (2005)

• Adapted quickly into most macro‐policy models: • IMF, GIMF; BoC, GEM; FED, SIGMA; EC, QUEST; ECB, EAGLE.

8

Page 9: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Risk Spread Sign. Neg. Correl. with Output

Correlation of risk spread(t), output(t+j), HP filtered data, 95% CI

• Notes: Risk spread is measured by the difference between the yield on the p y ylowest rated corporate bond (Baa) and the highest rated corporate bond (Aaa). Bond data were obtained from the St. Louis Fed website.

Christiano et al. 2010

Page 10: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. BGG (1999) Overview

• Capital acquisition financed via net worth and “bank” loans– CSV: asymmetric information about the payoff from capital

• Some firms default in any given period after shock realized– only partially recoverable to “banks”

• Borrowers compensate “banks” for the risk by paying an external finance premium– depends inversely on entrepreneurs’ aggregate net worth

Page 11: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. The Risk Shock in the BGG

Christiano et al. 2010

Page 12: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. CSV Details – desirable or undesirable?

• A set of NONLINEAR optimality conditions

• Entrepreneurs absorb all risksp

• Bank have zero profit condition

• Defaulted capital lost to the bank is distributed back to households • Results in higher household income

• When riskiness increases some firms (right tail) are more profitable.

Page 13: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Impact of Leverage under an Increase in Borrower Riskiness

Andersen and others (2013)

13

Page 14: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Impact of Leverage under an Increase in Borrower Riskiness

Andersen and others (2013)

• The amount of leverage is altered in the U.S. economy – a ratio of corporate debt l ti t fi ’ t th f ith 1 1 75 2 5relative to firms’ net worth of either 1, 1.75, or 2.5. 

• Because entrepreneurs must pay their interest obligations on debt to avoid bankruptcy an increase in leverage increases the cutoff rate for profitability thatbankruptcy, an increase in leverage increases the cutoff rate for profitability that the entrepreneur has to achieve to avoid bankruptcy. 

• Thus the higher leverage is in the steady state the more likely that the• Thus, the higher leverage is in the steady state, the more likely that the entrepreneur will default for a given increase in risk. 

• Thus higher leverage ratios make the user cost of capital more sensitive and• Thus, higher leverage ratios make the user cost of capital more sensitive and business investment more volatile in the presence of other shocks to the economy. 

Page 15: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. The Approximation

• If the model is log linearized, the a parameter (ν) is introduced: the time‐varying elasticity of the external finance premium with respect to the entrepreneurs’

Christiano et al (2010) 

elasticity of the external finance premium with respect to the entrepreneurs  leverage ratio:

• This governs the risk shock.

• All nonlinearities are lostAll nonlinearities are lost.

Page 16: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. The Approximation

Christiano et al (2010) 

Page 17: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. An Increase in U.S. Consumption– U.S. Effects

Beaton and others (2014)

17

Page 18: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Key Mechanisms of Financial AcceleratorBGG (1999)BGG (1999)

• Inverse relationship between the external finance premium and net worth– Lower levels of borrowers net wealth increase the divergence of borrower and 

l d i t tlender interests– Implies greater agency costs– Lenders demand a higher risk premium

• Introduces the “financial accelerator effect”Introduces the  financial accelerator effect– Shocks that raise output tend to be amplified if the shock also raises capital 

values and entrepreneurial income– Helps create a positive correlation between consumption and investmentHelps create a positive correlation between consumption and investment– Increases the persistence of shocks

• Introduces the “Fisher debt deflation effect”– Contracts are in nominal terms: unexpected changes in the price level result in– Contracts are in nominal terms: unexpected changes in the price level result in 

a reallocation of wealth between entrepreneurs and lenders– Shocks that reduce the price level hurt entrepreneurial net worth and depress 

output

Page 19: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

Question: Limitations of the Financial Accelerator Framework?Framework?

19

Page 20: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Key Limitations of BGG (1999)

• Models only the demand side of credit– Loan dynamics are often undesirable for non‐financial shocks– Causes over specification of the risk shock– Deposits are fully flexible

• Ignores alternative sources of risk spread (risk aversion, liquidity)• Applies only to ‘mom and pop grocery stores: 

– bank dependent for outside finance– no access to equity, bond markets, etc

• Retained earnings – for CSV net worth is paid out slowly and exogenously to keep it from accumulating above steady state

• Does not explicitly model banks

20

Page 21: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Financial Flows in Non‐Financial SectorJermann and Quadrini (2012)

21

Page 22: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. How Do Firms Finance?

7

8Nonfinancial corporate business; credit market instruments; liability

Total mortgages

5

6

s

Total mortgages

Other loans and advances

3

4

Trillions

Depository institution loans

Corporate bonds

0

1

2

Municipal securities and loans

22

0

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

Commercial Paper

US Flow of Funds, table L102.

Page 23: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. How Important are Loans for Firms Financing?

800

1000Nonfinancial corporate business; liability

200

400

600

ions

‐400

‐200

0Bill

‐800

‐600

1990

Q1

1990

Q4

1991

Q3

1992

Q2

1993

Q1

1993

Q4

1994

Q3

1995

Q2

1996

Q1

1996

Q4

1997

Q3

1998

Q2

1999

Q1

1999

Q4

2000

Q3

2001

Q2

2002

Q1

2002

Q4

2003

Q3

2004

Q2

2005

Q1

2005

Q4

2006

Q3

2007

Q2

2008

Q1

2008

Q4

2009

Q3

2010

Q2

2011

Q1

2011

Q4

2012

Q3

2013

Q2

2014

Q1

23US Flow of Funds, table F102.

Change in corporate loans Change in corporate bonds

Page 24: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I.  Understanding Firm Financing: The Future

• Need to understand importance of mechanisms for firm financing:– Internal versus external financing (Jermann and Quadrini, 2012)– Dependency of firms on direct loans versus bonds (Adrian et al., 2012)– How does this depend on firm size? (Begenau Salomao, 2015)– Frictions in raising firm capital?g p– Non‐depository‐institutions‐loan financing?

• But…. why so much focus on firms?But…. why so much focus on firms? – What about Households, Government debt, Trade financing, etc.

24

Page 25: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. What Loans are on Bank Balance Sheets?

2%All Commercial Banks

U.S. Loans and Leases in Bank Credit (2012)

8%

20%

20% Credit Cards

Commercial and Industrial Loans20%

8%Residential Real Estate (Closed‐End)

Commercial Real Estate

Home Equity Loans

22%20%

Home Equity Loans

Other loans

Interbank Loans

25Source: FRED database, authors estimates

Page 26: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Demand for Credit Models: Housing

• Iacoviello (2005)– Models the housing demand for credit– Based on collateral constraint frameworks Kiyotaki and Moore (1997)Based on collateral constraint frameworks Kiyotaki and Moore (1997)– Key disturbance: housing demand shock

• Pataracchia et al (2013)– Models housing building on Iacoviello (2005)Models housing  building on Iacoviello (2005).– Adds Endogenous leverage constraints– Incorporated into EC Quest Model

• Less central banks have incorporated housing into their work horse policy models compared to the CSV framework– Focus remains on labor sector banking and financial frictions

26

– Focus remains on labor sector, banking and financial frictions

Page 27: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Demand for Credit Models: Housing

• Key Gains of Modeling Housing– Helps explain the movement in consumption over the business cycle– Able to look at impact of housing as a source of shocksAble to look at impact of housing as a source of shocks– Helps capture the actual holdings on bank balance sheets

• Key Channels Missing from Modeling Housing in Policy Models– Speculative bubbles IO mortgages investment asset (Barlevy and FisherSpeculative bubbles, IO mortgages, investment asset (Barlevy and Fisher, 

2010)– Ex‐post default – idiosyncratic risk (Forlati and Lambertini, 2011)– Alternative mortgage contracts and default risk (Corbae and QuintinAlternative mortgage contracts, and default risk (Corbae and Quintin, 

2010). – Housing Illiquidity and Search friction (Hedlund, 2013)

27

Page 28: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

I. Demand for Credit Models: Take Aways

• What we have seen:– A focus on firm financing from loans via CSV models– Less role for non‐loan credit instrumentsLess role for non loan credit instruments– Less focus on housing – other credit demand– Few models of ex‐post default in housing

• What other loan channels are missing? ‐ Needed?What other loan channels are missing?  Needed? – Student debt– Trade financing 

Corporate real estate– Corporate real estate– International Financial Linkages

• What other credit channels are missing? ‐ Needed? S iti b d t

28

– Securities, bonds, etc– Role for sovereign debt

Page 29: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

II. The Quest for the Supply Side of Credit

Page 30: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

II. Credit Supply Models

1. The financial accelerator framework– Ex‐post informational asymmetry– External finance premiump

2. Collateral constraints framework– Limited contract enforcement environment ted co t act e o ce e t e o e t

3. Costly banking framework– Non‐convex production technologyNon convex production technology

• Most frameworks imply that intermediation costs are positively related to volume of intermediation and are always procyclical over the business cycle – at odds withof intermediation and are always procyclical over the business cycle  at odds with the empirical evidence. Borio et al. (2001)

Page 31: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

II. Where Are the Frictions?

6.00

7.00U.S Interest Rate Spreads

12.00U.S. Interest Rate

Beaton and others (2014)

1 00

2.00

3.00

4.00

5.00

Interbank Spread

6.00

8.00

10.00

Policy Rates 

LIBOR

‐3.00

‐2.00

‐1.00

0.00

1.00 p

Lending Spread

Corperate Spread

0.00

2.00

4.00LIBOR

Prime

Corp. BBB Yields

31Board of Governors of the Federal Reserve System,  table H15

Page 32: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

II. Early Models of the Supply Side of Credit• Markovic (2006) 

– Looks at the role of bank capital, and the effect of bank capital requirements. – The sector is driven by the demand for credit – a higher demand for credit is 

met with higher savings of households.met with higher savings of households.– A double moral hazard framework between banks and depositors and banks 

and entrepreneurs– Still only intermediation – still driven by the demand for credit.M h d M (2008) G tl d Ki t ki (2010)• Meh and Moran (2008) ; Gertler and Kiyotaki (2010)– Bank net worth directly effects lending limits due to collateral constraints.– Banks are at the limit of their bank capital requirement – no buffer.– Lending pinned down by exogenous leverage requirement.rp g p y g g q

A LGov. Bonds

Net worth

D it

A LNet worth

Value of

Central Bank

rd

rl

rp

Interbank lending

Deposits

CB deposits

Loans

Value of Capital

Households

Deposit Banks Firms

Page 33: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

II. Early Models of the Supply Side of Credit• Dib (2010)

– Looks at the role of bank capital, and develops a theory on interbank lending, and quantitative easing. B k h L ti f l di t h l f d it d it l– Banks have Leontief lending technology for deposits and capital –causes problems.

– Shocks to the technology of lending and interbank monitoring– Still only intermediation – still driven by the demand for credit.Still only intermediation  still driven by the demand for credit.

A L A L L

Deposit Banks Lending Banks Firmsrp

A L A LGov. Bonds

Interbank lending

Net worth

Deposits

Net worth

Lending

Gov. Bonds

Interbank lendingCB

A LNet worth

Loans

Value of Capital

Central Bank

Households

rd

ri rl

lendingCB deposits

Households

Page 34: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

II. Current Models of the Supply Side of Credit

• Benes and Kumhof (2011); Van Den Heuvel (2009)– Banks optimally choose a capital buffer against hitting requirements. 

A fi i l l t h i f b k i il t BGG (1999) ll i f– A financial accelerator mechanism for banks similar to BGG (1999) allowing for a s.d. of bank riskiness shock – ex‐post heterogeneity.

– Begins to develop a theory of aggregate credit creation – not just intermediationintermediation.

• Adrian et al. (2012)– In addition to net worth, leverage is endogenous affecting bank lending.

Total credit stable firms shifted from bank loans to bond financing– Total credit stable ‐ firms shifted from bank loans to bond financing.– Total lending by banks instable – reflected in risk premiums.

Page 35: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

III. Policy Models: Current Trends

Page 36: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

III. Financial Frictions Policy Models: post 2008Models

Institutions Other NewFeatures

Financial Frictions

BoC‐GEM/ GEM

Bank of Canada/ IMF

Fiscaldetails

Supply: Deposit and lending banks with interbank marketDemand: Financial accelerator, international loans from banks to foreign firms 

FSGM IMF (2015) Remittances, oil

Risk premiums: Corporate (Output gap), Sovereign, External, Economy Wide, Household

GIMF IMF Labor  Supply: Heterogeneous banks, endogenous leverage search, remittances, oil

pp y g , g gconstraints, capital buffers, ex‐post losses Demand: Financial accelerator

QUEST EuropeanCommission

Creditconstrained HHs

Supply: Heterogeneous banks, endogenous leverage constraints, capital buffers, ex‐post losses Demand: financial accelerator, Iacoviello housing 

Page 37: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

III. What is Happening to Policy DGSE Models?

• Macro‐financial theory continues to be developed– Mainly in academic toy models 

S i tit ti i l di d d l h i (IMF EC)– Some institutions are including more advanced general mechanisms (IMF, EC)

• There has been a move to semi‐structural policy modelsll f fl b l ( )– Allowing for greater flexibility (quantitative accuracy)

– Example: FSGM (IMF), LAN (BoC), G‐MUSE (BoC)

• Detailed financial models remain satellite models– Generally not kept in day‐to‐day workhorse models– Only the key exogenous mechanisms for financial shocks are kept in workhorse

Page 38: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Current Debates in Macro‐Finance

Page 39: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. The Quest for a Decent Supply Side of Credit

A model of the supply side of credit has not yet been satisfactorily developed. Why?• Disagreement on the key mechanisms of the sector

– Which sectors of the financial market matter?– What are the key frictions? 

• Disagreement on the key sources of shockssag ee e t o t e ey sou ces o s oc s– What makes banks/financial firms risky?– Where in the process do shocks originate?

• Disagreement on stylized facts/ calibrations– Few estimations/ empirical studies of new mechanisms

Lack of mapping to observed balance sheets– Lack of mapping to observed balance sheets

Page 40: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Basic Disagreement on the Role of Banks

Are banks sources of shocks, accelerators of shocks, or absorbers of shocks?• Most models are procyclical to aggregate demand movements.

I thi t?• Is this even correct?• Prior to crisis bank lending channel found to play a limited role in cyclical 

fluctuations may even dampen shocks during normal periods– may even dampen shocks during normal periods.

• Gambacorta and Marques‐Ibanez (2011) depends on:– Core capital positions

D d k t f di– Dependency on market funding– Dependency on non‐interest sources of income– Degree of securitisation activity

Page 41: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Changes in Commercial Bank Balance Sheets?

41Adrian et al. (2012)

Page 42: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Leverage Growth and Asset Growth of US Investment BanksInvestment Banks

42Adrian and Shin (2007); Source: SEC

Page 43: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Intermediation versus Credit Creation

Jakab and Kumhof (2015)

Page 44: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Intermediation versus Credit Creation

• Liabilities are created and destroyed anytime credit is extended withdrawn

• Banks alter liability side of balance sheet to expand collateral and assets

i k i h d b l h bl l l f b l h• Risk weighted balance sheet stable relative to actual size of balance sheet

• Leverage is counter cyclical over business cycle because collateral increases in good ti th h l d t htimes even though leverage does not change

Page 45: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Shadow Bank Liabilities versus Traditional Bank LiabilitiesLiabilities

Source: Pozsar et al. (2012)

Page 46: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Bank versus Non‐Bank Financial Instutions

Source: Adrian and Shin (2009)

Page 47: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. Which frictions matter?

• Which sectors of the financial market matter?– Commercial versus financial firms– Shadow banks versus commercial banks– Interbank markets versus final lending– OTC trading versus general assets

• Quantities or prices?The quantities in the balance sheet of banks– The quantities in the balance sheet of banks 

– Maturity mismatch ‐ liquidity risk– Information asymmetries – risk mispricing– Solvency risky

• Spillover channels?– Direct exposure– Bank runs

I f ti t i– Information contagion– Fire sales contagion– Risk aversion

Page 48: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

IV. What is needed?

• Going forward we will need a general theory

Fl ibl h t t f t h k• Flexible enough to capture future shocks

• Captures key macro linkages

• Useful for evaluating policy tools:– Loan‐to‐value– Leverage requirements– Capital requirements– Risk weights– etc?

Page 49: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

V. Conclusions

Page 50: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

V. Conclusions

• Policy makers have focused on firms demand for credit– Simplistic in design– Limited in helpfulness

• The supply side of credit has not yet been satisfactorily developed– Focus on financial intermediation – commercial banks– Disagreement on the source of shocks– Disagreement on key channels

• Going forward we will need a model flexible enough to capture future shocks, their key macro linkages, and use policy instruments

50

Page 51: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

ReferencesAdrian, T., Colla, P., and  Shin, H. S. 2012. “Which Financial Frictions? Parsing the Evidence from the Financial Crisis of 2007‐9” (No. 

w18335). National Bureau of Economic Research.Adrian, T., & Shin, H. S. (2009). The shadow banking system: implications for financial regulation (No. 382). Staff Report, Federal 

Reserve Bank of New York.Andersen, D., B. Hunt, M. Kortelainen, M. Kumhof, D. Laxton, D. Muir, S. Mursula and S. Snudden. 2013. “Global Integrated 

Monetary and Fiscal Model (GIMF) – Model Properties.” IMF Working Paper No. 2013/55y ( ) p g p /Barlevy, G., and J.D. Fisher, 2010.  “Mortgage choices and housing speculation” (No. 2010‐12). Working PaperBeaton, K., R. Lalonde, and S. Snudden. 2014. “The propagation of US shocks to Canada: understanding the role of real financial 

linkages.” Canadian Journal of Economics, 47(2), 466‐493.Benes, J., and M. Kumhof, 2011. "Risky Bank Lending and Optimal Capital Adequacy Regulation," IMF Working Papers 11/130.Bernanke, B. and Gertler, M. 1989. “Agency Costs, Net Worth, and Business Fluctuations.” American Economic Review, 79(1), pp.

14 3114‐31.Bernanke, B., Gertler, M., and Gilchrist, S. 1999. The financial accelerator in a quantitative business cycle framework. Handbook of 

macroeconomics, 1:1341‐1393.Borio, C., Furfine, C., and Lowe, P. 2001. Procyclicality of the financial system and financial stability: issues and policy options. BIS 

papers, 1, 1‐57.Christiano, L., E. Martin, and C. Evans, 2005, “Nominal Rigidities and the Dynamics Effects of a Shock to Monetary Policy,” Journal of 

li i l E 113(1) 1 45political Economy, 113(1), 1‐45.Christiano, L., J. Motto, and R., Rostagno, M., 2010. “Financial Factors in Economic Fluctuations.” In 2010 Meeting Papers (No. 141). 

Society for Economic Dynamics.Dib, A. 2010. “Capital Requirement and Financial Frictions in Banking: Macroeconomic Implications," Bank of Canada, Working 

Papers, 10‐26.Dib, A., and I. Christensen, 2005. “Monetary Policy in an Estimated DSGE Model with a Financial Accelerator” Computing in , , , y y p g

Economics and Finance, 314, Society for Computational Economics.

Page 52: Macro Financial Modeling in Macro Policy DSGE Models RPRD ... · – Dependency of firms on direct loans versus bonds (Adrian et al., 2012) ... – AbleAble toto looklook atat impactimpact

ReferencesIacoviello,  M., 2005. "House Prices, Borrowing Constraints, and Monetary Policy in the Business Cycle," American Economic 

Review, American Economic Association, vol. 95(3), pages 739‐764.Ireland, P. 2004. “Technology shocks in the new Keynesian model” The Review of Economics and Statistics, 923‐936.Jermann U., and V. Quadrini, 2012. “Macroeconomic Effects of Financial Shocks,” American Economic Review, American Economic 

Association vol 102(1) pages 238‐71 FebruaryAssociation, vol. 102(1), pages 238 71, February.Jakab, Z.,  and M. Kumhof, 2015, “Banks are not intermediaries of loanable funds — and why this matters” Bank of England 

Working Paper No. 529Gertler, M. and Kiyotaki, N, 2010. "Financial Intermediation and Credit Policy in Business Cycle Analysis," Handbook of Monetary 

Economics, in: Benjamin M. Friedman and Michael Woodford (ed.), Handbook of Monetary Economics, edition 1, volume 3, chapter 11 pages 547 599 Elsevierchapter 11, pages 547‐599 Elsevier.

Gambacorta, L., & Marques‐Ibanez, D. (2011). The bank lending channel: lessons from the crisis. Economic Policy, 26(66), 135‐182.Kiyotaki, N. and J. Moore, 1997. “Credit Cycles". Journal of Political Economy 105 (2): 211–248.Kumhof, M., D. Laxton, D. Muir and S. Mursula 2010, “The Global Integrated Monetary Fiscal Model (GIMF) ‐ Theoretical Structure”, 

IMF Working Paper Series, WP/10/34.Markovic, B. 2006. “Bank Capital Channels in the Monetary Transmission Mechanism.” Bank of England Working Paper No. 313.Muller, P., Bishop, G., Devnani, S., Lewis, M., & Ladher, R. (2012). Non‐bank financial institutions: assessment of their impact on the 

stability of the financial system (No. 472). Directorate General Economic and Monetary Affairs (DG ECFIN), European Commission.

Pataracchia, B., R. Raciborski, M. Ratto, and W. Roeger. (2013). “Endogenous housing risk in an estimated DSGE model of the EuroPataracchia, B., R. Raciborski, M. Ratto, and W. Roeger. (2013).  Endogenous housing risk in an estimated DSGE model of the Euro Area.” European Commission, Working Paper (No. 505).

Pozsar, Z., Adrian, T., Ashcraft, A., & Boesky, H. (2010). Shadow banking (No. 458). Staff Report, Federal Reserve Bank of New York.Roger, S., and J. Vlcek 2012. "Macrofinancial Modeling At Central Banks," IMF Working Papers 12/21, International Monetary Fund.