1 Fiscal Policy with Credit Constrained Households Werner Roeger and Jan in ’t Veld DG ECFIN - Economic and Financial Affairs European Commission June 2009 e views expressed here are those of the author and should not be attributed to the European Commissio
38
Embed
1 Fiscal Policy with Credit Constrained Households Werner Roeger and Jan in ’t Veld DG ECFIN - Economic and Financial Affairs European Commission June.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Fiscal Policy with
Credit Constrained Households
Werner Roeger and Jan in ’t Veld
DG ECFIN - Economic and Financial AffairsEuropean Commission
June 2009
The views expressed here are those of the author and should not be attributed to the European Commission.
2
Revival of interest in discretionary fiscal policy
Earlier scepticism on effectiveness of fiscal policy: • Dominance of supply shocks in the past• Financial liberalisation
Severity of financial crisis• sharp fall in aggregate demand• tightening credit constraints• limits of monetary policy
Policy response:• European Economic Recovery Plan• Fiscal packages announced by EU member states : 1.1% of
GDP in 2009, 0.6% of GDP 2010.• G-20 : 1.8% of GDP in 2009, 1.3% in 2010 ($1.35 trl)
3
Structure of the presentation
1. Literature review of fiscal multipliers
2. Empirical evidence credit constraints
3. QUEST III with credit constrained households
4. Impulse responses to government spending and tax shocks (with and without credit-constrained households)
5. The role of monetary policy
4
Overview of empirical literature
VARs
• Results in the VAR literature differ widely. • The highest multipliers are obtained for US
data (Blanchard and Perotti (2002)), • Estimates for European data seem to be less
significant (Perotti (2005), De Castro and Fernandez de Cos (2006), Afonso and Sousa (2009))
5
Overview of empirical literature:DSGE models
Since the result obtained by Blanchard and Perotti is often regarded as representing a stylised fact, several attempts have been made to come up with DSGE models which can generate positive consumption co-movement.
– Ravn et al. (2007) introduce a market structure into the model which implies a strong decline in the mark up in the case of a government spending shock in order to generate a positive consumption effect.
– Monacelli et al. (2008) introduce a utility function which implies a stronger comovement between hours worked and consumption in order to generate the same effect.
– Gali et al. (2007) generate a positive effect on private consumption by introducing substantial capital market imperfections in the form of liquidity constrained households.
6
Overview of empirical literature: DSGE But empirically estimated DSGE models have problems
to obtain such an effect:– Ratto et al. (2009a) estimate a 1st year multiplier for
gov. cons. shocks of around 0.6 with an estimated share of liquidity constrained households of about 30% for the euro area, similar for gov. inv. but lower for transfers. But it is impossible to replicate the Gali result despite liquidity constrained consumers.
– Also Coenen and Straub (2005) find for a similar share of non-Ricardian households a decline in aggregate consumption.
Credit constraints constitute an attractive alternative hypothesis. Given the uncertainty about income and wealth developments of borrowers, banks typically impose collateral constraints.
This paper therefore explores the consequences for fiscal policy of this credit market friction.
7
Figure 1: Euro area:
Credit standards applied to the approval of loans to households (net percentages of banks reporting tightening credit standards)
-20
-10
0
10
20
30
40
50
loans for house purchases consumer credit and other loans
8
Figure 2: US:
Credit standards applied to the approval of loans to households (net percentages of banks reporting tightening credit standards)
Net Percentage of Domestic Respondents Tightening Standards for Mortgage Loans
( if Δ i = 0 => r < 0 )3. Composition (spending > revenue)4. Duration (temporary > permanent)
Current financial crisis:• Increase credit-constrained households • Monetary policy more likely to be
accommodative Higher fiscal multipliers
10
QUEST III housing model
Extension of the QUEST III model with a housing sector and credit-constrained consumers along the lines of the financial accelerator literature.(Kiyotaki and Moore (1997), Iacoviello (2005), Iacoviello and Neri (2008), Monacelli (2007), Calza, Monacelli and Stracca (2007))
Disaggregation of the household sector into borrowers and lenders.
Credit-constrained households: intertemporal optimising over consumption, leisure, housing subject to borrowing constraint (collateral constraint – endogenously linked to nominal value of asset (housing)
Positive co-movement consumption and residential investment
11
Households
Ricardian/lenders: intertemporal optimising (utility separable in consumption, leisure and housing services)
• Full access to financial markets
Credit-constrained / borrowers: optimising utility• Borrow from Ricardian households • Face collateral constraint on borrowing
Liquidity-constrained (hand-to-mouth):• Consume their current disposable income
Households 1: Ricardian households - lenders
Period utility function separable in C, leisure and housing services HRicardian hh hold government bonds and bonds issued by domestic and foreign hh, real capital of T and NT sector
Intertemporally optimising (as Ricardians) (i.e. not hand-to-mouth) but:
1. higher rate of time preference βc<βr and 2. they face a collateral constraint on their borrowing : They
borrow Bc from domestic Ricardian households
c
tHt
ct
tc
tt
ct
ct
HcHt
ct
tc
t
ct
ct
t
cLSt
t
tWctt
Wt
ctt
ct
cHt
Ht
Ht
ct
Ct
ct
tcct
t
ct
ct
ct
tcc
HpB
HJH
TW
WLwtBrBItpCpt
HLCUVMax
)1(
)1(
2)1())1()1()1(
),1,(
00
1,
00
0
,
1
2
1,
0
000
Housing investment:Shadow price of housing capital ςt = PDV of ratio of the marginal utility of housing services H and consumption C
CC:
)1()1()1(
1
)1(
)1)((
)1( 11
1
111
1 Hct
ht
rt
ct
Ht
GDPtt
tHt
ct
rt
Ct
ctt
rtr
ct
Ht
rt
tptrptH
pthCC
tp
)1()1()1(
)1()1(
)1(
)1)((
)1( 11
1
111
1 Hct
Ht
ct
ct
Ht
GDPtt
tttH
tct
ct
Ct
ctt
ctc
ct
Ht
ct
tptrptH
pthCC
tp
Note: Lagrange multiplier of the collateral constraint ψ - acts like premium on interest rate (fluctuates positively with tightness of constraint)
Ric:
Consumption:
)1(
)1()(
1
1
t
tc
tct
tctt r
hCC
hCC
)1()(
1
1t
r
trt
trtt r
hCC
hCC
Ric:
CC:
15
(24) WtC
t
tCt
Wt
ltc
lrtc
rctc
c
ltL
lrtL
rctL
c
P
W
t
t
UsUsUs
UsUsUs
)1(
)1(
,,,
,1,1,1
(25)
))1(())1((//11 11 tWtt
WtW
Wt sfwsfw 10 sfw
(23) lLSt
ltt
ltt
wt
lt
ct
ct TTRLWtCPt ,)1()1(
Households 3: Liquidity-constrained households
“Hand-to-mouth”: Consume entire disposable income (no intertemporal optimisation)
Wage setting
Trade union maximises a joint utility function(distributed equally – population weigths si )
Wage rule :
Wage mark up:
16
(26a) lt
lct
crt
rt CsCsCsC
(26b) lt
lct
crt
rt LsLsLsL with l
tct
rt LLL .
(27) rtc
rct B
s
sB .
Liquidity constrained households do not own financial assets:
0 lt
Flt
lt KBB
Credit constrained households only engage in debt contracts with Ricardian households:
Aggregation:
17
Household decision rules in a special case: 01 ct
IH tH
Consumption/savings decision
(20) )1(
)1()(
1
1
t
tc
tct
tctt r
hCC
hCC
Housing investment decision
(22') Ct
Ht
tHH
tt
tct
cct
p
pi
hCCH
)(
)(1
1
1
A tightening of the credit constraint ( 0 t ) leads to 1) A decline in the housing investment to consumption ratio 2) A reduction in current consumption
18
Figure 3: Response of consumption to changes in current income (absolute deviations)
Housing investment: Non-constrained and credit constrained hh
0 1 2 3 4 5 6 7 8 9 1 0-0 .2
0 .0
0 .2
0 .4
0 .6
0 .8
1 .0
RIC_WR CC_WR
Real wages
0 1 2 3 4 5 6 7 8 9 1 0-0 .1 5
-0 .1 0
-0 .0 5
0 .0 0
0 .0 5
0 .1 0
0 .1 5
0 .2 0
RIC_INF CC_INF
Inflation
0 1 2 3 4 5 6 7 8 9 1 0-0 .0 5
0 .0 0
0 .0 5
0 .1 0
0 .1 5
0 .2 0
0 .2 5
32
Fiscal multipliers in model with credit-constraints
Table 1 First year GDP effects of fiscal shocks of 1% of GDP
fiscal measures:
Permanent stimulus
Temporary stimulus
(one year)
Temporary with monetary
accommodation (1) Investment subsidy 0.46 1.37 2.19 Government investment 0.84 1.07 1.4 Government consumption 0.36 0.99 1.4 Consumption tax 0.37 0.67 0.99 Government transfers 0.22 0.55 0.78 Labour tax 0.48 0.53 0.68 Corporate profit tax 0.32 0.03 0.05 Note: GDP percentage difference from baseline for global shocks of 1% of (baseline) GDP, assuming long run financing through labour tax increases. (1) unchanged nominal interest rates for 1 year.
33
Spill-oversTable 2 First year GDP effects of fiscal shocks:
Fiscal stimulus in: EU RoW Global Government consumption EU GDP 0.74 0.26 0.99 RoW GDP 0.09 0.96 1.04 Government investment
EU GDP 0.84 0.24 1.07 RoW GDP 0.08 1.04 1.12 Government transfers
EU GDP 0.40 0.15 0.55 RoW GDP 0.05 0.53 0.58 Labour tax
EU GDP 0.41 0.12 0.53 RoW GDP 0.04 0.52 0.56 Consumption tax
EU GDP 0.49 0.18 0.67 RoW GDP
0.06 0.64 0.70
Note: GDP difference from baseline in first year of simulation, for resp. EU acting alone, RoW acting alone and global coordinated expansions. All shocks are credibly temporary for one year and of equal size, 1% of baseline GDP.
34
Fiscal Policy in the current crisis
Temporary fiscal policy measures can be effective to support growth in the current crisis due to
• increase in credit constraints and • monetary policy more likely to be
accommodativeGlobal expansion leads to positive spillovers
(coordination needed to avoid countries taking a free ride)
Design (composition) matters.Need for credible temporary policies (guarantees
that expansion does not become permanent): avoid adverse financial market reaction
35
36
Credibility of temporary fiscal expansions:
Effects of permanent changes in spending and taxes are much smaller, and generally become negative in the long run (agents anticipate future increases in taxes and reduce their consumption and save more)
Smaller multipliers if stimulus is not perceived as temporary but permanent.
Risk premia: unsustainable fiscal strategies may give rise to increase in risk premia (sovereign bonds, (corporate bonds))
Smaller multipliers if stimulus leads to higher borrowing costs for government (and private sector)
Need credible medium-term strategies to deal with increase in debt
37
Figure 4 Credibly temporary vs. permanent shocks:increase government consumption and increase in risk premia
Temp.+mon.acc : 1 year increase gov. cons (1% of GDP) with nominal interest rates unchangedTemp. : 1 year increase government consumption (1% of GDP) Perm. : permanent increase gov. consumption (financed by tax increases)Perm.+sov.rp : permanent increase plus sovereign bond risk premium 100 bp.Perm.+rp : permanent increase plus sov. bond risk premium 100bp plus risk premium 25bp
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
0 1 2 3 4 5 6 7 8 9 10
year
GD
P (
% d
iff.
fro
m b
ase)
temp+mon.acc temp. perm. perm+sov.rp perm.+rp
Higher fiscal multipliers in financial crisis:Effect of credit-constraints and monetary policy