Monetary Policy and Financial Stability

Giovanni Dell’Ariccia (IMF and CEPR)

The views expressed in this paper are those of the authors and do not necessarily represent those of the IMF, its Executive Board, or Management.

NBB Conference, October 2016

Talk Plan

Evolution of consensus on role of monetary policy in financial

stability

Some empirical regularities

Cost/benefits of leaning against the wind

Role of financial variables in estimating potential output

Governance challenges when MP has multiple targets

Macro literature:

Financial intermediation seen as macro neutral

Asset prices (including property prices) did matter. They could accentuate the

cycle through financial accelerators (BGG etc.)

But macro model largely ignored their impact on bank risk taking. In equilibrium,

no bank defaults

Banking literature

Focused on excessive risk taking by intermediaries operating under limited

liability and asymmetric information

Defaults/crises in equilibrium

But there was little attention to macro and monetary policy conditions

Before the crisis … A theory gap

Monetary policy to focus on inflation (and output gap): “divine

coincidence”

Asset prices and credit aggregates a concern only through their

impact on GDP and inflation (exceptions RBA, Riksbank, EMs)

Benign neglect approach to boom/busts:

Bubbles difficult to identify

Costs of clean up limited and policy effective

Better clean up than prevent

Bank risk taking important, but job of regulators

Before the crisis …A policy gap

Regulatory policy focused on individual institutions

Limited attention to credit aggregates or asset price dynamics

Ill equipped to deal with booms:

Correlated risk taking

Fire sales and other externalities

Few regulators had necessary tools (exceptions: Spain/Colombia)

Before the crisis …A policy gap

Pre-crisis Consensus: No leaning

against the wind

“I find it difficult to conceive the degree of central bank certainty to justify

the scale of preemptive tightening that would likely be necessary to

neutralize a bubble.” Alan Greenspan, 2002

“First, the Fed cannot reliably identify bubbles in asset prices. Second,

even if it could identify bubbles, monetary policy is far too blunt a tool for

effective use against them.” Ben Bernanke, 2002

“…monetary policy should not respond to asset prices per se, but rather to

changes in the outlook for inflation and aggregate demand resulting from

asset price movements…attempting to "prick" an asset price bubble,

should be avoided.” Rick Mishkin, 2008

Pre-crisis: Macro ok, but risks were

growing

-2

-1

0

1

2

3

4

5

6

7

8

2000 2001 2002 2003 2004 2005 2006 2007

Output Gap(In percent of potential output)

Ireland

Spain

United States

United Kingdom

100

120

140

160

180

200

220

240

260

2000Q1 2001Q3 2003Q1 2004Q3 2006Q1 2007Q3

Residential Real Estate Prices, (2000Q1=100)

100

120

140

160

180

200

220

2000Q1 2001Q3 2003Q1 2004Q3 2006Q1 2007Q3

Credit-to-GDP (2000Q1=100)

0

1

2

3

4

5

6

7

8

2001 2002 2003 2004 2005 2006 2007

Core Inflation(In percent, y/y)

Source: World Economic Outlook (September 2007 vintage for the output gap) and Haver Analytics.

Figure 1. Output Gap, Core Inflation, and Financial Indicators Before the Crisis

Standard policies rapidly hit their limits

Limited effectiveness of less traditional policies

Large fiscal and output costs

Multiple banking crises; especially in countries with their own credit

and real estate booms

Then the crisis came …

8

LVA

EST

LTU

IRL

UKR

JPN

RUS

DNK

HKG

SWE

SVN

GBRNLD

SVK

ESP

BGR

MYS

BOL

THAPHL

AUS

IND

KAZ

PAN

URY

DOM

NPL

VNM

BGD

MOZCHL

MARSUR

IDN

CHN

y = -1.2852x + 12.969R² = 0.14

-50

-25

0

25

50

75

100

-30 -20 -10 0 10 20 30

Ch

an

ge in

cre

dit-

to-G

DP

ra

tio

fro

m 2

00

0 to

20

06

Change in GDP from 2007 to 2009

Credit Growth and Depth of Great Recession

Bubble size shows the

level of credit-to-GDP ratio in 2006.

Crisis: severity in line with

magnitude of credit booms

AKALAR

AZCA

CO

CTDC

DE

FL

GA

HI

IA

ID

IL

IN

KSKY

LA

MA

MD

ME

MI

MN

MO

MS

MT

NC

ND

NE

NH

NJ

NM

NV

NY

OH OK

OR

PA

RI

SC

SDTN

TX

UTVA

VT

WA

WI

WV

WY

y = 1.1159x + 20.457R² = 0.5501

-50

0

50

100

150

200

250

0 20 40 60 80 100 120 140 160

Ch

an

ge in

mo

rtg

ag

e d

elin

qu

ency ra

te, 2

00

7-0

9

House price appreciation, 2000-06

Subprime Boom and Defaults

Bubble size shows the percentage point

change in the ratio of mortgage credit outstanding to household income from 2000 to 2006.

Crisis: severity in line with

magnitude of credit booms

Many stories/theories linking interest rates and risk taking

Some compatible others opposite to each other

Often different implications for different types of agents/intermediaries

Few entail views about “excessiveness” of risks

Empirically: growing evidence that low rates imply greater risk taking.

But magnitudes unclear

The crisis challenged existing consensus

Many argued that monetary policy provided intermediaries with the

wrong incentives (Borio et al., 2008)

Several stories associate low interest rate environment to crisis

Overly loose monetary policy (Taylor, 2009)

Abundant liquidity – search for yield (Rajan, 2005)

Risk-shifting: what matters are transitions (Landier et al., 2011)

Liquidity risk (Acharya and Naqvi, 2010, Freixas et al., 2011)

Adverse selection and strategic effects in credit booms (Allen and Carletti, 2011,

Dell’Ariccia and Marquez, 2006, Ruckes, 2004)

Increase in leverage (Adrian and Shin, 2008, 2009…Dell’Ariccia et al., 2011)

Others focus on how expected macro bailout and risk externalities

seed ground for new crises

Diamond and Rajan (2010), Farhi and Tirole (2009), Acharya and Yorulmazer

(2007)

The risk taking channel: Theory

13

Growing literature linking monetary easing to greater risk taking

Ioannidou, Ongena, and Peydró (2009): Bolivia; Altunbas, Gambacorta, and

Marques-Ibañez (2010), Maddaloni and Peydró (2011): Lending standards

euro area (and US); Jimenez et al. (forthcoming): Spain; Dell’Ariccia, Laeven

and Suarez (2013): US. Lown and Morgan (2006): lending standards (not

significant). Paligorova and Santos (2012), Delis et al. (2012): Differential

spreads on syndicated loans. Buch/Eickmeier/Prieto (2011): aggregate version

of STBL. Adrian and Shin (2011): Leverage

Magnitude of effect less robust

Different papers reach different conclusions

Cross-sectional dimension (which intermediaries are most affected) also in

question

Little sense of whether this risk taking is “excessive”

The risk taking channel: Evidence

Some evidence from the US

2.4

2.6

2.8

33.2

3.4

Ris

k o

f lo

an

s (

detr

en

de

d)

-2 0 2 4 6Real Federal Funds Rate (detrended) (in %)

7.5

8

8.5

9

9.5

C

apital-to

-asset ra

tio

(detr

ended)

(in

%)

-2 0 2 4 6 Real federal funds rate (detrended) (in %)

Dell’Ariccia, Laeven, Suarez, 2016, (JF, forthcoming)

Implications for monetary policy

Is the “divine coincidence” dead?

We already knew short-term trade-off inflation/output

Is there also one between output/inflation eqlb and financial stability?

Financial frictions imply that low/stable inflation is not enough any longer

(assuming systemic risk taking is excessive)

Other tools?

Macroprudential (LTVs, DTIs, dynamic provisioning, cyclical CARs)

But unlikely to work perfectly

Potential need to lean against the wind

Many questions:

What metrics (leverage, asset-prices, credit growth,…)

Rules versus discretion

General overhaul of IT and Taylor rules or case-by-case practical approach?

Today’s views are more diverse

“Monetary policy is poorly suited for dealing with financial stability concerns,

even as a last resort.” John Williams, 2015

“For existing empirical estimates and reasonable assumptions, the marginal

cost of leaning against the wind is much higher than the marginal benefits.

Thus, leaning against the wind is not justified. Lars Svensson, 2015

“Monetary policy faces significant limitations as a tool to promote financial

stability… [But] it may be appropriate to adjust monetary policy to ‘get in the

cracks’ that persist in the macroprudential framework.” Janet Yellen, 2014

“It would make sense not to rule out the possible use of the interest rate for

this purpose, particularly when other tools appear to be lacking.” Stan Fischer,

2015

“In other words, we have been leaning against the wind.” Oystein Olsen, 2015

“Financial stability is too large a task for prudential… frameworks alone.

Monetary policy strategies also need to… lean against the build-up of financial

imbalances even if near-term inflation remains low and stable.” Jaime

Caruana, 2011

To lean or not to lean? A three step

approach

Transmission

How does monetary policy affect financial variables?

What are the effects on financial stability?

Tradeoffs

Is policy tightening for inflation purposes sufficient?

How often do we see a conflict between price and financial stability

objectives?

Welfare analysis

Costs and benefits of leaning against the wind

18

Costs/benefits analysis: Should

monetary policy lean against the wind?

In general, no. Reasonable parameters suggest costs exceed benefits

Other tools (macro- and micro-prudential)

Yet, benefits grow relative to costs when:

Conjuncture: rapid credit growth, low unemployment, high probability of long-lasting

and severe crisis,

Structure: large, interconnected economy (spillovers)

Prudential policies should be the first policy considered More targeted, probably less costly,

Both micro- and macro-prudential can play a role

Putting empirical results together

Costs

Benefits

=

Lower crisis

probability Duration of crisis*

Unemployment

gap in crisis**

Higher short term

unemployment †

x x

(†) Due to 100 bps increase in rates for 1 year.

Illustrative scenarios

Building blocks Average

probability High (peak) probability

High (peak) probability, severe crisis

Lower crisis probability, pp 0.02 0.3 0.3

Duration of crisis, years 4.5-6 4.5-6 6-8

Unemployment gap in crisis, % 5 5 7

Higher unemployment, pp 0.5 0.5 0.5

Benefits 0.008 0.113 0.294

Costs 0.25 0.25 0.25

Ratio (B:C) 0.03 0.45 1.18

A Different Role for Financial

Variables?

Before the GFC:

Real-time estimates of output gaps did not signal substantial overheating

CPI inflation was below target in most advanced economies

After the GFC:

Large upward revisions to output gaps

Greater awareness of the role of housing and credit booms

Use real-time financial data to reduce errors in potential output

estimates

Conflict between mandates looks smaller ex-post than ex-ante

Potential output a bit of a moving target

-6

-4

-2

0

2

4

2000 2007 2012

April 2015 WEO

Real-time WEO

Panel 1. Cross-country average, output gap

(Percent)

-6

-4

-2

0

2

4

2000 2007 2012

April 2015 WEO

Real-time WEO

Panel 2. United States, output gap

(Percent)

100

120

140

160

180

200

2000 2007 2012

Credit

House prices

Panel 3. Cross-country average, credit and house prices

(Real indexes, 2000=100)

100

120

140

160

2000 2007 2012

Credit

House prices

Panel 4. United States, credit and house prices

(Real indexes, 2000=100)

-10

-5

0

5

10

15

2000 2007 2012

April 2015 WEO

Real-time WEO

Panel 1. Greece, output gap

(Percent)

-6

-4

-2

0

2

4

2000 2007 2012

April 2015 WEO

Real-time WEO

Panel 2. Spain, output gap

(Percent)

100

120

140

160

180

200

220

240

260

280

300

320

2000 2007 2012

Credit

House prices

Panel 3. Greece, credit and house prices

(Real indexes, 2000=100)

100

120

140

160

180

200

220

240

260

2000 2007 2012

Credit

House prices

Panel 4. Spain, credit and house prices

(Real indexes, 2000=100)

Potential output a bit of a moving target

Economic and financial overheating

Is macro-prudential policy the answer?

25

Potential issues

Circumvention

Calibration

Difficult political economy

Evidence

Promising results: negative association with incidence of booms

and booms turning bad (even with adverse bias)

But effects often limited (and sometimes temporary)

More success in building up buffers than preventing booms

altogether

Limited use in AEs so far

Relationship with other policies

How many agencies in charge of MoP/MaP?

Two instruments (Policy rate, MaP)/ Two objectives (Inflation/output, Stability)

Each instrument affects both objectives

If perfectly functioning, design does not matter

But, if not, separation improves credibility

Especially if CB’s mandate very clear

Similar to fiscal/monetary policy divorce (think Barro/Gordon)

At potential cost of second-best policy mix

Example, in a recession:

CB cuts rate aggressively to stimulate demand

FA reacts by tightening macro-prudential regulation to reduce risk-taking → CB

eases even more → FA ….

Result: a policy mix with too low interest rates and too tight macro-prudential

measures

Governance issues with financial stability

mandate

Outsourcing price stability to independent CBs was

“easy”:

A clear and measurable objective: low and stable inflation (some

attention to short-term output)

Clearly understood (often mono-dimensional) tools: the policy rate

Accountability led to properly designed incentives for central bankers

Outsourcing financial stability is much more complicated

Governance issues with financial stability

mandate

Paradox of success Unlike monetary pol.: No easily measurable target (is there a too stable

financial sector?)

Unlike prudential supervision: No yardstick

Nobody sees the crisis that did not happen

Politically charged (with or without MaP) Hit most vulnerable

Against increased credit access

Need for rule-based approach. But… Measurability makes delegation challenging

Far from calibration of DSGE standards

Theoretical foundation for CB independence on price stability: Inflation is an inferior tool to deal with fiscal constraints

Time-inconsistency problem

This clearly still desirable

Analogous arguments for financial stability? Governments may be tempted to use regulation to distort incentives for banks to

finance the treasury

Politicians may be reluctant to tighten if this is politically costly

Legitimate concerns Democratic deficit if a central bank is endowed with powers ranging from setting

interest rates to credit allocation and financial regulation

Especially in the context of mandates with measurability issues

Risks to Central Bank Independence?

Evidence: Little difference in inflation

performance with multiple mandates

-5

0

5

10

15

20

25

30

United K

ing

dom

Sw

eden

Chile

Colo

mbia

Guate

mala

Mexic

o

Peru

Kore

a

Hung

ary

Pola

nd

Turk

ey

Icela

nd

New

Zeala

nd

South

Afr

ica

Bra

zil

Uru

guay

Isra

el

Indonesia

Philip

pin

es

Thailand

Ghana

Alb

ania

Czech R

epublic

Slo

vak R

epublic

Rom

ania

inflation d

evia

tion fro

m targ

ets

Perc

enta

ge p

oin

ts

Central Banks not in Charge of Bank Supervisionaverage inflation deviation = 3.16

Central Banks in Charge of Bank Supervisionaverage inflation deviation = 3.49

Figure 7. Inflation Performance and Bank Supervision among Inflation Targeters

Note: Average inflation deviation from the target since the central bank introduced inflation targeting until 2006Q4. Difference in means is not statistically

significant.

Source: Central Banks' websites, Haver Analytics, and staff calculations.

So far, a financial stability mandate has meant to be in charge of

relatively “a-cyclical” bank regulation and supervision

Political pressures can intensify:

Tools with more targeted effects (with clearer winners and losers)

Cyclical use of prudential tools

Communication/credibility challenges

One tool/two targets

Conflicting mandates

Key challenge:

Protecting MP independence (on price stability) if government/public chooses to

exercise greater oversight on new central bank responsibilities

Yet, important concerns remain

Policy summary

Adverse trade-offs in using MP for financial stability purposes

Limited effectiveness of macroprudential measures (intended as cyclical use of prudential rules)

Complicated governance issues

Go back to basics?

32

Bank NPLs in crises

0

10

20

30

40

50

60

70

80

90

Non-OECD Countries OECD Countries

NPL ratio, percent

Role of bank capital/loss absorption

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60

Loss given default = 50% Loss given default = 75%

Risk-weighted bank capital ratio, percent

Share of banking crises avoided

Thank You

35