Financial Contracting and the Specialization of Assets

Download Report

Transcript Financial Contracting and the Specialization of Assets

Property Prices and Bank Risk Taking
Giovanni Dell’Ariccia (IMF and CEPR)
Norges Bank Macroprudential Regulation Workshop, Oslo, November 29-30, 2012
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.
Before the crisis …A policy gap

Monetary policy to focus on inflation and output gap

Asset prices a concern only through their impact on GDP
and inflation (exceptions RBA, Riksbank, some 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 theory gap


Macro literature:

Financial intermediation seen as macro neutral

Asset prices (including property prices) did matter. They could
accentuate the cycle through financial accelerator

But macro models 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

There are defaults/crises in equilibrium

But there is little attention to macro and monetary policy
Before crisis … Macro looked OK
United States
Euro area
Average of other economies1
Core CPI Inflation
4.0
Output Gap2
2
3.5
1
3.0
2.5
0
2.0
-1
1.5
1.0
-2
0.5
0.0
2000
1 Japan
02
04
06
08:
Q4
2000
omitted.
of output gap using rolling Hodrick-Prescott filter.
2 Estimate
02
04
06
08:
Q4
-3
But house prices were rising rapidly
80
Change in real house prices
(2001:Q4-2006:Q3)
Real house price falls from recent
peak
60
40
20
0
-20
ESP
NZL
IRL
FRA
GBR
DEN
SWE
CAN
AUS
FIN
NOR
ITA
USA
GRC
NLD
CHE
PRT
AUT
DEU
JPN
-40
Then the crisis came …

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 real estate booms
7
House boom/busts and great recession
Figure 2. House Price Run-Up and Severity of Crisis
Cumulative decline in GDP f rom start to end of recession
10
IND
0
AUS
CHN
NZL
CAN
FRA
GRC
CHE CYP
PRT
AUT
USA
KOR
NLD
CZE HRV
HUN
DNK SWE
BGR
FIN
SVN
-10
ZAF
ESP
GBR
NOR
ITA
POL
y = -0.0416x - 4.1152
R² = 0.1496
IRL
ISL
UKR
EST
-20
Bubble size shows the change in bank
credit from 2000 to 2006.
LTU
LVA
-30
-20
0
20
40
Source: Claessens et al (2010).
60
80
100
120
140
160
Change in house prices from 2000 to 2006
180
200
220
240
A closer look at real-estate booms
and bank risk taking behavior

Most large banking crises preceded by some form
of property price boom





Property cycles can have macro consequences,
even without open banking crises


Scandinavia 1990s
Asia 1997
Japan 1990
More recently: US, Spain, Ireland, Iceland, Latvia,…
Borrower debt overhang
But things are worse when credit booms (and lax
standards) are involved
Text Table 1. Booms, Crises, Macroeconomic Performance
followed by
financial crisis
followed by
poor
performance
followed by
financial crisis
or poor
performance
followed by
financial crisis
and poor
performance
Number of
countries
Real estate
Credit
53%
67%
77%
78%
87%
93%
43%
52%
30
27
Real estate but not credit
29%
71%
71%
29%
7
Credit but not real estate
100%
75%
100%
75%
4
Both
61%
78%
91%
48%
23
Neither
27%
18%
45%
0%
11
Boom
Notes: The sample consists of 40 countries. The numbers, except in the last column, show the percent of the cases
in which a crisis or poor macroeconomic performance happened after a boom was observed (out of the total number
of cases where the boom occurred). The last column shows the number of countries in which a boom occurred. A
real estate boom exists if the annual real house price appreciation rate during 2000-2006 is above the ad-hoc
threshold of 1.5 percent or the annual real house price appreciation rate in the upward phase of the housing cycle
prior to the crisis exceeds the country-specific historical annual appreciation rate. A credit boom exists if the growth
rate of bank credit to the private sector in percent of GDP is more than the arbitrary cut-off of 20 percent or it
exceeds the rate implied by a country-specific, backward-looking, cubic time trend by more than one standard
deviation. A financial crisis is a systemic banking crisis as identified in Laeven and Valencia (2010). Poor
performance is defined as more than 1 percentage point decline in the real GDP growth rate in 2008-09 compared to
the 2003-07 average.
Change in credit-to-GDP ratio from 2000 to 2006
Credit Growth and Depth of Great
Recession
100
y = -1.2852x + 12.969
R² = 0.14
IRL
75
ESP
LVA
50
DNK
EST
LTU
GBR
NLD
BGR
SVN
UKR
25
RUS
KAZ VNM
AUS
SWE
IND
0
-25
Bubble size shows the
level of credit-to-GDP
ratio in 2006.
SVK
HKG
JPN
THAPHL
BOL
SUR
NPLBGD
MAR
IDN
CHL
CHN
MOZ
DOM
PAN
URY
MYS
-50
-30
-20
-10
0
10
Change in GDP from 2007 to 2009
20
30
Real-estate cycles and bank behavior

Credit constraints – Leverage cycles

Adverse selection and strategic interaction

Bubbles

Govt. guarantees - Risk externalities
Financial Accelerators – Leverage Cycles

Collateralized credit as solution to agency problems
(Kiyotaki/Moore, 1997)

Cycle emerges: asset prices  credit aggregates 
investment/demand  asset prices

Effect magnified if logic applied to intermediaries
(Kiyotaki/Gertler, 2009, Iacoviello, 2011)

Further widening if leverage is cyclical (Adrian/Shin,
2009/Geanakoplos 2010)

Regulation may also contribute (Herring/Wachter, 1999)

But most models do not deal with risk taking
Magnified macro fluctuations
Duration of recession
(quarters)
Time to return to trend
(quarters)
5
4
6
Recessions without
Recessions with
Recessions with Severe
*
*
5
Recoveries without
Recoveries with
Recoveries with Strong
4
3
3
2
*
2
1
*
1
0
0
Credit Crunch
House Price Bust
Source: Claessens/Kose/Terrones, 2008
Credit Boom
House Price Boom
Adverse selection and strategic effects

Rising house prices reduce incentives to screen
borrowers


 Winner curse reduced in good times:





Even bad borrowers can refinance/sell property
My competitors screen less
More untested applicant borrowers
 Better distribution of applicants
 Less incentives to screen
“Conservative” lending punished


Investor pressure on managers (compensation schemes)
Borrowers shop for lax standards
Easy mortgages during U.S. boom
Decline in Lending Standards
(share of no-downpayment and limited documentation loans in
originations; combined loan-to-value in percent)
80
86
No downpayment & limited documentation
No downpayment
Limited documentation
Combined LTV
84
70
82
60
80
50
78
40
76
30
74
20
72
10
70
0
68
2001
2002
2003
Source: Dell’Ariccia, Igan, and Laeven 2009
2004
2005
2006
2007
Change in mortgage delinquency rate, 2007-09
Subprime Boom and Defaults
250
AZ
NV
y = 1.1159x + 20.457
R² = 0.5501
CA
200
VT
HI
150
FL
MD
OR
ID WA
100
IL
WI MN
GA
NC
SC
AL
IACO TN
MO
AR
ND
KY
KS SD
OH
OK MS
NE
WV
LA
MI IN
50
UT
MA
CTNY
MT
NM
NH ME WY
DE VA NJ
RI
DC
Bubble size shows the percentage point
change in the ratio of mortgage credit
outstanding to household income from
2000 to 2006.
AKPA
0
0
20
40
60
80
100
120
TX
-50
House price appreciation, 2000-06
140
160
Bubbles

Normal times: prices reflect fundamentals

Bubble: speculative motive allows for deviation
from fundamentals (Allen/Carletti, 2011)

Banks may fund speculators:




Govt. guarantees
Risk shifting (limited liability)
Can’t separate speculators from “legitimate” consumers
Increasing recourse to instruments with
correlated risks


U.S.: teaser-rate/interest-only loans
East Europe: FX denominated loans
Interest-only loans and boom
Source: Barlevy and Fisher (2011)
Credit and housing booms in East Europe
Figure A1. Selected CEE Countries: Private Sector Credit and Housing
Prices, 2003–081
Change in real housing prices, 2003–08
250
250
Lithuania
y = 3.0989x - 16.334
R² = 0.3625
200
200
150
Ukraine
Bulgaria
Poland
Russia
100
150
Estonia
100
Slovak Republic
50
Czech Republic
Croatia
Serbia
Slovenia
50
0
0
10
20
Hungary 30
40
50
-50
60
0
Change in private sector credit-to-GDP ratio, 2003–08
Sources: IMF International Financial Statistics; and country statistical offices.
the boom in the Baltic states ended in 2007, data for the Baltics refer to 2002–07.
1As
FX lending and credit boom
Emerging Europe: Total Private Sector Credit by Currency, 2008
(Stock in percent of GDP)
120
120
National currency
Foreign currency
Foreign currency indexed
100
80
100
80
60
60
40
40
20
20
Sources: National authorities; and IMF, International Financial Statistics.
Belarus
Turkey
Albania
Moldova
Romania
Serbia
Russia
Macedonia
Slovak Republic
Poland
Czech Republic
Bosnia & Herzegovina
Lithuania
Croatia
Hungary
Ukraine
Bulgaria
Latvia
0
Estonia
0
Strategic complementarities

Government guarantees




Risk taking externalities




Do not want to die alone (Farhi/Tirole, 2012)
Greenspan put
FX in Eastern Europe
Poor incentives structure if systemic banks take
excessive risk
Correlated risk taking
Self fulfilling equilibria
Ex-post …

Macro bailouts did occur
A new policy consensus?

If benign neglect is dead, then what?





Objectives?




Asset price booms difficult to spot
But other policy decisions also taken under uncertainty
Booms involving leveraged agents more dangerous
 Real estate case
Prevent unsustainable booms altogether
Alter lender/borrower behavior
Increase resilience to busts
No silver bullet


Broader measures: hard to circumvent but more costly
23
Targeted tools: limited costs but challenged by loopholes
Monetary policy


Natural place to start

Credit highly correlated with monetary aggregates

Increase cost of borrowing, decrease loan demand, lower
collateral values

Risk-taking channel
Potential issues

Conflict of objectives

Impact on balance sheets

Capital inflows (especially in SOEs)

Switch to riskier lending (FX, IO loans)
Credit Growth and Core Inflation
Figure 8. Credit Growth and Monetary Policy
(Selected countries that had a boom in the run -up and a crisis in 2007-08)
United Kingdom 2007
Ireland 2008
250
4
250
4
200
200
3
3
150
150
2
2
100
100
1
50
1
50
Core inf lation
Credit (right axis)
Core inf lation
Credit (right axis)
0
0
0
T-5
T-4
T-3
T-2
T-1
0
T-5
T
T-4
T-3
T-2
T-1
T
Greece 2008
Spain 2008
250
250
4
4
200
200
3
3
150
150
2
2
100
100
1
1
Core inf lation
Credit (right axis)
0
0
T-5
T-4
T-3
T-2
T-1
50
50
T
Core inf lation
Credit (right axis)
0
0
T-5
T-4
T-3
T-2
T-1
T
Sources: IMF International Financial Statistics, World Economic Outlook; staff calculations.
Notes: Credit is indexed with a base value of 100 five years prior to the crisis.
9
Evidence of risk shifting
Source: Landier et al. 2011
Fiscal policy

Prudent stance can:




Time lags make it an impractical tool



Reduce overheating
Buffer for bailout/stimulus during a potential bust
Reduce incentives for leverage (deductibility, FAT)
Some measures hard to use countercyclically
“Tax planning”, circumvention, calibration
Little evidence of effectiveness in stopping
booms… …but fiscal room critical in busts
Macro-Prudential Tools

Most ‘experiments’ in emerging markets,
particularly Asia

Common tools:



Maximum LTV/DTI limits
Differentiated risk weights on high-LTV loans
Dynamic provisioning

Discretion rather than rule-based

Mixed evidence on effectiveness
Hong Kong: Limited Effectiveness of
LTV Limits
160
New loans approved
Prices
170
150
150
140
130
110
90
70
2009 - Mar 2009 - May
August 2010:
LTV for properties over HK$12 million
lowered to 60 percent, applications for
mortgage insurance exceeding 90% LTV and
50% DTI suspended, maximum loan size for
mortgage insurance eligibility if LTV>90%.
October 2009:
Maximum LTV for properties over
HK$20 million lowered to 60
percent, maximum loan size for
mortgage insurance eligibility
reduced and non-owner-occupied
properties disqualified.
130
120
110
2009 - Jul
2009 - Sep 2009 - Nov 2010 - Jan
2010 - Mar 2010 - May
2010 - Jul
Korea: Effective LTV Limits, but
Difficult Calibration?
6%
6
Month-on-month house price changes in 'speculation zones' (LHS)
5%
Policy rate (RHS)
5
September 2002:
Introduced LTV limits
4%
4
3%
2%
September 2009:
Tightened DTI
October 2003:
Lowered LTV in
speculative areas
1%
3
February 2007:
Tightened DTI
2
0%
June 2003:
Lowered LTV in
speculative areas
-1%
-2%
2000 - Jan
July 2009:
Lowered LTV in
non-speculative
areas
August 2005:
Introduced DTI limits
1
0
2001 - Apr
2002 - Jul
2003 - Oct
2005 - Jan
2006 - Apr
2007 - Jul
2008 - Oct
2010 - Jan
Conclusions

Benign neglect might be dead, so …

Emerging consensus that leveraged bubbles (real
estate in particular) are dangerous

What to do. Still many open questions:




Monetary policy remains blunt instrument
Fiscal impractical. Perhaps helpful on liability structures
Macroprudential tools promising …
But it will take time:



Develop new macro models
Design/calibrate macroprudential tools
Build institutions to control them
Limited liability and speculators
Unlevered
consumer
Levered
speculator
H-P(1+r)
q
1-q
H-P(1+r)
q
L-P(1+r)
1-q
L-P(1+r)