Download Market Confidence and Liquidity Hoarding

Market Con…dence and Liquidity Hoarding
Volha Audzei
Czech National Bank and CERGE-EI
(Prague)
May 2016
Motivation
Credit crunch and central banks policy
No agreement about the policy e¤ect in the literature:
-Curdia and Woodford (2011) and Taylor and Williams (2009): was not
e¢ cient or irrelevant
-Del Negro et al. (2011) and Christensen et al. (2014), Gertler and
Karadi (2011): helped avoid more severe recession
Liquidity hoarding
Change in sentiment
Motivation
Liquidity hoarding
Motivation
Impact on bank’s lending standards
Motivation
Paper contribution
Counterparty risk in the interbank market
Liquidity hoarding
Policy exercises:
liquidity provision, targeted liquidity provision, declining policy rate,
relaxing collateral constraints
Model
Overview
DSGE framework (Gertler and Karadi (2011))
Continuum of banks, indexed by i, lend to the real sector and to
each other
Two types of assets:
safe (reserves), pays Rtres
risky, pays Rtk+1
Banks di¤er by their beliefs about risky asset return,
Êti Rtk+1 N (R̄tk ,KF , PtKF )
Model
Overview
Assumption 1:
Rtk =
α PKt Yt t + (1
Qt
δ) Qt ζ t
1
Model
Overview
Assumption 1:
Rtk =
α PKt Yt t + (1
Qt
δ) Qt ζ t
1
Assumption 2:
ζ t = ρζ ζ t
1
+ µt + εζ ,t
(1)
Model
Overview
Assumption 1:
Rtk =
α PKt Yt t + (1
Qt
δ) Qt ζ t
1
Assumption 2:
ζ t = ρζ ζ t
1
+ µt + εζ ,t
µt is a persistent shock
µt = ρ µ µt
1
+ vt
(1)
Model
Overview
Assumption 1:
Rtk =
α PKt Yt t + (1
Qt
δ) Qt ζ t
1
Assumption 2:
ζ t = ρζ ζ t
1
+ µt + εζ ,t
(1)
µt is a persistent shock
µt = ρ µ µt
Experts’opinions
1
+ vt
µit = µt + θ it
(2)
Model
Bank’s balance sheet
Assets
Manufacturers’claims
Reserves
Interbank lending
Liabilities
Deposits
Interbank borrowing
Net worth
Model
Bank’s problem
Each bank maximizes expected return, choosing αit , hti :
Model
Bank’s problem
Each bank maximizes expected return, choosing αit , hti :
risky asset:
αit
Êti Rtk+1
Model
Bank’s problem
Each bank maximizes expected return, choosing αit , hti :
risky asset:
αit
Êti Rtk+1
reserves:
hti
Rtres
Model
Bank’s problem
Each bank maximizes expected return, choosing αit , hti :
risky asset:
αit
Êti Rtk+1
reserves:
hti
Rtres
interbank market lending:
pti 1
αit
hti
Rtib
Model
Bank’s problem
Each bank maximizes expected return, choosing αit , hti :
risky asset:
αit
Êti Rtk+1
reserves:
hti
Rtres
interbank market lending:
pti 1
αit
hti
Rtib
Êti Rtk+1
Rtib
Lit
or interbank market borrowing:
Model
Bank’s problem
Each bank maximizes expected return, choosing αit , hti :
risky asset:
αit
Êti Rtk+1
reserves:
hti
Rtres
interbank market lending:
pti 1
αit
hti
Rtib
Êti Rtk+1
Rtib
Lit
or interbank market borrowing:
Li interbank loan
Lit =
λb
0, for lenders
(net worth), for borrowers
Model
Bank’s problem continued
Collecting the terms by αit :
αit Êti Rtk+1
pti Rtib
αit = 1 for Êti Rtk+1 pti Rtib , αit = 0 for Êti Rtk+1 < pti Rtib
Note: for each lender all borrowers are identical
pti = Prob Êti Rtk+1 (1 + λb ) > Rt dt + λb Rtib
A lender lends if pti Rtib > Rtres , hoards otherwise
A borrower borrows if Êti Rtk+1 Rtib
Model
Interbank market and beliefs
Model
Crisis and policy responses
"Fundamental" shock: ζ t = ρζ ζ t
Sentiment shock:
µ̂it
=
Policy: rpt = κ p Rtk+1
1
+ µt + εζ ,t
µt + η it
Rt
(Rk
R)
untargeted Qt Kt +1 + Rest = Dt + ψt (Qt Kt +1 + Rest )
targeted Qt Kt +1 + Rest = Dt + ψtarg
Q t K t +1
t
interest rate Rtres rpt
collateral constraint λb rtλ
Policy costs: τψt (Qt St + Rest ) or τψtarg
(Qt St + Rest )
t
Model
Overview
Results
Impulse Responses to Sentiment (5%) and Fundamental Shock (5%)
Eξ','ξ
policy response
hat
premium
0
10
1
5
0.5
-2
-4
0
10
20
30
0
10
Y
20
30
10
C
0
0
0
-2
-2
-5
10
20
30
10
K
20
30
0
-10
20
30
20
30
20
30
Q
3
2
1
0
-1
10
10
L
-5
30
-10
-4
-4
20
Res
0
-2
-4
-6
-8
10
N
20
30
20
30
10
R
0
0.2
0
-0.2
-0.4
-0.6
-10
-20
-30
10
20
GK basic
30
10
No expect.shock
Expect.shock
Pure expect.shock
Results
Role of the interbank market
policy response
prem
2
Y
0.2
0
0.1
0
-0.1
-0.2
0
-2
5
10
C
15
20
5
10
Res
15
20
5
10
K
15
20
5
10
R
15
20
5
10
15
share of borrowers
20
0
0
0
-0.1
-0.05
-0.2
-0.1
-0.4
5
10
L
15
20
5
10
N
15
20
0
0
0
-0.5
-0.2
-0.4
-0.2
-0.02
-1
5
10
Rres
15
20
0.05
5
10
15
interbank lending
20
0
0
-0.04
0
-2
-10
-4
-0.05
5
10
15
20
-20
Expect. shock
5
10
15
20
Pure expect. shock
-6
5
10
No expect. shock
15
20
Results
Policy e¤ects
premium
Y
C
Res
-5
1.5
-2
-2
-4
-4
-10
1
0.5
-15
0
5
10
K
15
20
5
10
L
15
20
5
10
Q
15
20
0
-6
-8
-10
-12
-14
5
10 15
Rres
20
0.2
0
-0.2
-0.4
-0.6
0.2
0
-0.2
-0.4
5
10 15
Rib
20
-10
-4
-15
-20
5
10 15 20
policy response
15
20
untargeted
20
5
20
0.1
10
0.05
-4
10
5
10 15
policy costs
20
-2
5
5
10 15 20
interbank lending
-2
-6
2
0
0.6
0.4
0.2
0
-0.2
5
10 15 20
share of borrowers
5
10 15 20
priv ate lev erage ratio
20
-20
-10
5
10 15 20
share of inv estors
15
-15
-5
-2
10
N
-10
2
0
5
5
10
15
20
0
5
10
15
20
0
targeted
10
15
Results
Policy e¤ects continued
premium
Y
3
-2
2
-4
1
-6
0
-8
5
10
K
15
20
-5
C
-4
-6
5
10
L
15
20
4
2
0
-2
-4
-6
-10
-15
5
10
R
15
20
Res
-5
-10
-15
-20
-25
-2
5
10
Q
15
20
0
-10
-5
-10
-20
5
10 15 20
share of inv estors
5
10 15 20
share of borrowers
-10
-1
-1
-20
20
5
10 15 20
priv ate lev erage ratio
10
4
0
5
10 15 20
policy response
10
15
20
no response
20
5
20
0.1
10
0.05
-5
5
5
10 15
policy costs
20
0
0
5
10 15 20
interbank lending
-20
5
2
20
20
0
10 15
Rib
15
40
1
-0.5
5
10
N
-30
-15
0
0
5
5
10
collateral
15
20
0
untargeted
5
10
15
20
targeted
0
10
interest rate
15
Conclusion
Investors’expectations generate long and large responses in model
variables
Banks hoard some liquidity provided by central bank due to their low
sentiment
Liquidity provision mitigates crisis slightly, but does not stop it, nor
decreases its duration
Future Work
?
Motivation
Investor Sentiment