Transcript Diapositiva 1

HOUSEHOLD LEVERAGE AND SOVEREIGN RISK
Comments
Juan F. Jimeno (Research Division)
CONFERENCE: “Debt and Credit, Growth and Crises”
Madrid
18-19 June 2012
THE PAPER (I)
• Main objective
• Explanation of cross-country GDP and employment growth in the Eurozone
(EZ) during the crisis
• Using the US experience as a starting point (Mian and Sufi, 2010, Midrigan and
Philippon, 2010)
• In the EZ: A two-stage story of the crisis (Two stages ?):
• 2007-2010: Deleveraging in the private sector
• 2010-?: Fiscal consolidation and sovereign crisis
• Cross-country differences in household leverage and financing costs to
Governments, respectively, can explain cross-country differences in GDP,
consumption, and employment growth in 2007-2009 and 2010-2011.
RESEARCH DIVISION
2
THE PAPER (II)
• Method
•Calibration of DSGE model with credit frictions (still preliminary)
• “Islands” with representative consumers
• Cash-in-advance-constraint with two sources of “money”
• Two sectors (tradables, non-tradables) with sector-specific labor
• Sticky wages
• Intertemporal substitution modeled as adjustment costs to asset holdings
• Government and private lenders are perfect substitutes. No monetary
policy (M=1). Fiscal transfers providing money.
• Extension (and simplification) of Midrigan and Philippon (2010) to account for
fiscal policy
•Two exogenous shocks:
• collateral constraint to private credit
• financing costs of Governments
RESEARCH DIVISION
3
MAIN QUESTION
• Do the model’s assumptions fit well with the European experience to be
explained?
• Empirical evidence on the credit-consumption channel
• The different nature of the crisis across EZ countries
• Asymmetries and imbalances in the EZ
• Credit boom, deleveraging and sovereign risk premia
RESEARCH DIVISION
4
COMMENT #1: RELATIONSHIP BETWEEN CREDIT
AND CONSUMPTION IN EUROPE
Alberola, Molina and del Rio (2012)
RESEARCH DIVISION
5
COMMENT #1: RELATIONSHIP BETWEEN CREDIT
AND CONSUMPTION IN EUROPE
1.
The role of housing as a source of liquidity
•
UK : Increases in housing prices led to more borrowing
Bridges, Disney and Gathergood (2009).
•
DK: Introduction of home equity loans led to increases in consumption.
Leth-Petersen (2010)
•
Less evidence for other economies, as equity withdrawal less intense
SP: Bover (2005): Estimation of wealth effects. Not very large.
Marques and Nieto (2003): Mortgage refinancing did not increase as a result of higher house prices.
(Update: only in 2005 increase in mortgage debt > increase in housing investment)
2. But even without home equity withdrawals, mortgage credit conditions may affect household
consumption (SP: Masier and Villanueva, 2011).
• Longer mortgage maturity partly allows to diminish the mortgage payment and increase consumption by
recent home owners
• Specially important if expect income to be higher later in life
3. On the downside: Are households/firms credit-constrained?
• Disentangling demand-supply of credit
SP: Hernando and Villanueva (2012): “relatively small magnitude of credit supply factors”
RESEARCH DIVISION
6
Table 2: OLS impacts of interest rates and maturity on non-durable consumption
Impact of a 100 bp Impact of a one year
Income growth
increase in R
increase in M
50-65 years
Panel A: Average impact
(1)
(2)
(3)
1.Total effect
-.0335
.0133
(.0180)*
(.0069)**
Panel C: By schooling of hhold head
2. Basic schooling
.0216
(.026)
.016
(.011)
-.126
(.071)*
3. High school
.023
(.032)
.029
(.010)**
-.099
(.077)
-.11
(.026)**
.007
(.008)
-.14
(.08)*
4. College
Estimation method: group -specific OLS. Robust standard errors in parentheses.
Sample of home owners observed in 2002/2005 who acquired their house with a mortgage between 1991 and 2006.
RESEARCH DIVISION
7
Table 3: The impact of R and M on non-durable consumption
Impact of a 100 bp
Impact of a one year
increase in R
increase in M
(1)
(2)
Panel B: By income group
2. Below median income
.0046
-.003
(.042)
(.014)
3. 50th< Y < 75th centile
-.028
(.035)
Sample
size
.025
(.012)**
454.2
615.2
4. Top income quartile
-.066
.013
785.4
(.039)*
(.01)
Estimation method: group -specific OLS. Robust standard errors in parentheses.
Sample of home owners observed in 2002/2005 who acquired a mortgage in 1992-2006.
Other Xs: permanent income, age at purchase, year of purchase, revaluation of house value, fa
savings bank dummy, 4 dummies with current age of spouse
RESEARCH DIVISION
8
LEVERAGE AS DRIVING CONSUMPTION: EVIDENCE
FOR SPAIN
• Spanish Survey of Household Finances, (EFF 2002, 2005, 2008)
• A panel that allows tracking the consumption, labor supply, wealth and debt
of households
• The evolution of household expenditure in vehicles between 2002 and 2008
by the leverage ratio in 2002
• Some evidence of higher cyclicality of the expenditure in durables among
indebted households
RESEARCH DIVISION
9
RESEARCH DIVISION
10
MAIN QUESTION
• Do the model’s assumptions fit well with the European experience to be
explained?
• Empirical evidence on the credit-consumption channel
• SOME EMPIRICAL SUPPORT. LIKELY TO BE RELEVANT. BUT
PROBABLY LESS STRONG THAN IN THE US
• The different nature of the crisis across EZ countries
• Asymmetries and imbalances in the EZ
• Credit boom, deleveraging and sovereign risk premia
RESEARCH DIVISION
11
COMMENT #2. SAME SHOCK? SAME (LABOR
MARKET) TRANSMISSION MECHANISMS?
0.1
Luxembourg
Malta
Austria
Finland France
Greece
Italy
SloveniaEA17
Belgium
Germany
Cyprus
Netherlands
0.05
0
Slovakia
Portugal
-0.05
Spain
-0.1
G
D
P
Ireland
-0.15
Estonia
-0.2
-0.2
-0.15
-0.1
-0.05
0
0.05
0.1
Employment growth (%) vs. GDP growth (2007q2-2010q2)
RESEARCH DIVISION
12
COMMENT #2. SAME SHOCK? SAME (LABOR
MARKET) TRANSMISSION MECHANISMS?
0.15
Estonia
0.1
0.05
Luxembourg
Germany
Malta
Belgium
Netherlands
France
EA17
Italy
Eslovenia
Slovakia
Austria
0
Finland
Cyprus
Ireland
Portugal
Spain
-0.05
G
D
P
-0.1
Greece
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
0.08
0.1
-0.15
0.12
Employment growth (%) vs. GDP growth (20010q2-2011q4)
RESEARCH DIVISION
13
MAIN QUESTION
• Do the model’s assumptions fit well with the European experience to be
explained?
• Empirical evidence on the credit-consumption channel
• The different nature of the crisis across EZ countries
• DIFFERENT SHOCKS.
• DIFFERENT LABOR MARKET RESPONSES DIFFICULT TO EXPLAIN
WITH SAME DEGREE OF LABOR MARKET RIGIDITIES ACROSS
COUNTRIES
• Asymmetries and imbalances in the EZ
• Credit boom, deleveraging and sovereign risk premia
RESEARCH DIVISION
14
COMMENT #3. ASYMMETRIES AND IMBALANCES
• Lower productivity growth in the periphery during 1999-2007…
• … and higher wage pressure, led to
Alberola, Molina and del Rio (2012)
RESEARCH DIVISION
15
COMMENT #3. ASYMMETRIES AND IMBALANCES
• …and together with lax fiscal policies to
Alberola, Molina and del Rio (2012)
RESEARCH DIVISION
16
COMMENT #3. ASYMMETRIES AND IMBALANCES
• but could have a more restrictive fiscal policy avoided the increase in
external debt?
Assume government consumption would have stayed constant, in per
capita terms, at its 19981s level for 10 years (Scenario #1) or 20 years
(Scenario #2). Beyond 2008 in Scenario #1, 2018 in Scenario #2,
government consumption represents again a 17:3% of GDP in each
period.
Gavilán, Jimeno, Hernandez de Cos, and Rojas (2010)
RESEARCH DIVISION
17
MAIN QUESTION
• Do the model’s assumptions fit well with the European experience to be
explained?
• Empirical evidence on the credit-consumption channel
• The different nature of the crisis across EZ countries
• Asymmetries and imbalances in the EZ
• COMPETITIVENESS, EXTERNAL IMBALANCES ARE KEY TO
UNDERSTAND EZ CRISIS
• Credit boom, deleveraging and sovereign risk premia
RESEARCH DIVISION
18
COMMENT #4. SOURCES OF SHOCKS
• Endogenous leverage
• The role of banks (and shadow banking) at generating credit cycles
• Nuño and Thomas (2012)
• Martinez-Miera and Suarez (2012)
• Factors behind the financing cost of EZ Governments ?
• Only fiscal variables?
• Contagion effects?
• Interactions public-private sectors?
• Other variables?
RESEARCH DIVISION
19
FINANCING COST OF GOVERNMENTS
03-Jun-12
03-May-12
3,500
03-Apr-12
3,500
03-Mar-12
4,000
03-Feb-12
4,000
03-Jan-12
4,500
03-Dec-11
4,500
03-Nov-11
5,000
03-Oct-11
5,000
03-Sep-11
5,500
03-Aug-11
5,500
03-Jul-11
6,000
03-Jun-11
6,000
03-May-11
6,500
03-Apr-11
6,500
03-Mar-11
7,000
03-Jan-11
7,000
03-Feb-11
Spain: 10y bond yield
July 7th: 4.71 %
July 11th: Spain wins FIFA World Cup
August 27th: 3.98 %
October 14th: 3.92%
RESEARCH DIVISION
20
THANKS FOR YOUR ATTENTION ¡¡¡
RESEARCH DIVISION
BACK
RESEARCH DIVISION
22