impact of the credit crunch on the polish economy · credit crunch increase in lending restrictions...
TRANSCRIPT
Motivation Model Calibration Preliminary Results Summary
Impact of the Credit Crunch on the PolishEconomy
(preliminary and incomplete)
Micha l Brzoza-Brzezina and Krzysztof Makarski
Economic Institute NBP and Warsaw School of Economics
June 2009
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Plan of the Presentation
1 Motivation
2 Model
3 Calibration
4 Preliminary Results
5 Summary
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Plan of the Presentation
1 Motivation
2 Model
3 Calibration
4 Preliminary Results
5 Summary
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in Poland
Credit crunch
increase in lending restrictionsincrease in spreads between the intebank rates and the creditrates.decline in loans to households and firms
Decline in external demand for Polish goods and services (anddecline in export)
Depreciation of the Polish zloty.
Capital outflow?
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in PolandLending restrictions in Poland. (Source: NBP)
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
2004 2005 2006 2007 2008 2009
LTV_MORTGAGE LTV_ENTERPRISE
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in Poland
Credit crunch
increase in lending restrictionsincrease in spreads between the intebank rates and the creditrates.decline in loans to households and firms
Decline in external demand for Polish goods and services (anddecline in export)
Depreciation of the Polish zloty.
Capital outflow?
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in PolandSpreads on loans to households in Poland. (Source: NBP)
-.05
-.04
-.03
-.02
-.01
.00
.01
.02
.03
.04
1998 2000 2002 2004 2006 2008
SPREAD_F SPREAD_H
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in Poland
Credit crunch
increase in lending restrictionsincrease in spreads between the intebank rates and the creditrates.decline in loans to households and firms
Decline in external demand for Polish goods and services (anddecline in export)
Depreciation of the Polish zloty.
Capital outflow?
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in PolandLoans to households in Poland. (Source: NBP)
-.4
-.3
-.2
-.1
.0
.1
.2
.3
.4
1998 2000 2002 2004 2006 2008
LOAN_H_HP LOAN_F_HP
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in Poland
Credit crunch
increase in lending restrictionsincrease in spreads between the intebank rates and the creditrates.decline in loans to households and firms
Decline in external demand for Polish goods and services (anddecline in export)
Depreciation of the Polish zloty.
Capital outflow?
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in PolandExport in Poland. (Source: Central Statistical Office)
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in Poland
Credit crunch
increase in lending restrictionsincrease in spreads between the intebank rates and the creditrates.decline in loans to households and firms
Decline in external demand for Polish goods and services (anddecline in export)
Depreciation of the Polish zloty.
Capital outflow?
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in PolandReal exchange rate in Poland. (Source: NBP)
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in Poland
Credit crunch
increase in lending restrictionsincrease in spreads between the intebank rates and the creditrates.decline in loans to households and firms
Decline in external demand for Polish goods and services (anddecline in export)
Depreciation of the Polish zloty.
Capital outflow?
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Main recent shocks in PolandCapital inflow. (Source: NBP)
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Response of GDP (Source: Central Statistical Office)
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Question
What is the role of financial frictions in this process?
How much of GDP decline is generated by financial frictions?
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Methodology
Dynamic Stochastic General Equilibrium (DSGE) of Polisheconomy.
The advantages of DSGE models
They take into account the influence of future on today’sdecisions in a coherent way (rational expectations)MicrofoundationsCalibration/Estimation of the so called deep parameters(describing e.g. preferences, technologies) - gives rise to thepossibility of studies on changes in policies (this approach isimmune to the Lucas critique)
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Plan of the Presentation
1 Motivation
2 Model
3 Calibration
4 Preliminary Results
5 Summary
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Key Features
New Keynesian model of business cycle
Non-neutrality of monetary policy due to:
Nominal rigidities - prices and wages are set according to aCalvo scheme (Calvo, 1983)
Monetary Policy
Taylor rule (Taylor, 1993)
Financial frictions - we extend Iacoviello (2005):
stochastic shocks to credit constraints and spreads,small open economy.
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Basic structure
Standard new Keynesian model of business cycle fluctuations.
Households
Monetary and fiscal authorities
Three stages of production: final good producers,intermediate good producers, and wholesale good producers(entrepreneurs)
Housing and capital producers.
Prices of intermediate goods and wages are sticky.
Banking sector, nominal rigidities, spreads plus creditconstraints.
Perfect competition in the other markets
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Households
There are two types of households: patient, P, and impatient,I . The type i ∈ {P, I} maximizes utility
E0 ∑t(βi )tu(ct , χt , nt , εt), subject to
the budget constraint and the borrowing constraint
RHL,tL
it ≤ mH
t Et
[Pχ,t+1 (1− δχ) χi
t
]where mF
t ∼ AR(1). Thus consumers need housing ascollateral to get a loan.
In equilibrium patient HHs save and impatient HHs borrow.
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Wholesale good producers (entrepreneurs)
Do not work, run firms. Sell their product in a competitivemarket.
Are impatient (βE = βI ).
Own capital use it as collateral for loans.
Maximize utility
E0 ∑t(βE )tu(ct , εt), subject to
the flow of funds, the production function and the borrowingconstraint
RFL,tL
Ft ≤ mF
t Et [Pk,t+1 (1− δk) kt ]
where mFt ∼ AR(1). Thus producers need capital as collateral
to get a loan.
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Other producers
Other producers are standard.
Intermediate good producers operate in monopolisticenvironment. Have sticky prices - Calvo pricing. Do not owncapital.
Final good producers put together domestic and foreignintermediate goods (imperfect substitutability) and sell finalgoods in a competitive market. Do not own capital.
Housing and capital good producers are standard (cost ofinstallation).
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
BankingDeposits
Policy rate - the interbank rate Rt .
Banks collect deposits from patient households and depositthem in the interbank market. zD,t ∼ AR(1) affects thespread between the interbank rate Rt and the HHs depositrate RH
D,t (follows from DIB,t = zD,tDt).
There are nominal rigidities (Calvo) that slow down theadjustment of the interest rates. The flexible deposit rate is(in log deviations from the steady state) RH
D,t = zD,t + Rt .With nominal rigidities this relationship becomes morecomplex.
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
BankingLoans
Furthermore, banks take loans in the interbank market andgrant loans to impatient HHs and firms. zL,t ∼ AR(1) affectsthe spread between the interbank rate Rt and the lending rateRL,t (follows from Lt = zL,tLIB,t).
There are nominal rigidities (Calvo) that slow down theadjustment of the interest rates. The flexible HHs loan rate is(in log deviations from the steady state) RH
L,t = Rt − zHL,t and
for firms RFL,t = Rt − zF
L,t . With nominal rigidities thisrelationship becomes more complex.
Borrowing constraints tied to housing (HHs) and capital(firms).
Banks have also access to the international nominal interestrate adjusted for risk premium.
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Monetary and fiscal authority
Central bank - the Taylor rule; log-linear version of the form:
Rt = γR Rt−1 + (1− γR)[γππt + γGDPˆGDPt ] + ϕt
The government budget, very simplified (to keep Ricardianequivalence only patient HHs are taxed)
Gt = TPt
where
Gt+1 =(
1− ρg
)µg + ρgGt + εg ,t+1, εg ,t ∼ N [0, σg ]
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Definition of Decentralized Equilibrium
Definition
An equilibrium is a set of allocations and prices, that:
solves the decision problems of all agents populating theeconomy
satisfies the government budget constraints
satisfies the Taylor rule
clears all markets
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Plan of the Presentation
1 Motivation
2 Model
3 Calibration
4 Preliminary Results
5 Summary
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Solution
The model was solved with 1-st order perturbation(linearization)
The model parameters were crudely calibrated.
In future we plan Bayesian estimation and/or propercalibration.
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Calibrated model parameters
A group of parameters are taken from the literature(parameters that affect business cycle behavior).
A group of parameters is obtained from the data (parametersthat affect the steady state).
Crude calibration of the shocks affecting financial markets
the initial shocks to spreads were calibrated to match theobserved change to spreads.the initial shocks to LTVs were calibrated to match (givenprevious calibration of the shocks to spreads) the observeddecline in loans.persistence was not calibrated, it was set at reasonable values.
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
IRF to a monetary policy shock
10 20 30 40−3
−2
−1
0
1x 10
−3 gdp
10 20 30 40−4
−2
0
2x 10
−3 c
10 20 30 40−0.01
−0.005
0i
10 20 30 40−10
−5
0
5x 10
−4 pi
10 20 30 40−8
−6
−4
−2
0x 10
−3 l_f
10 20 30 40−0.03
−0.02
−0.01
0
0.01l_h
10 20 30 40−10
−5
0
5x 10
−3 spread_f_l
10 20 30 40−10
−5
0
5x 10
−3 spread_h_l
10 20 30 40−5
0
5
10x 10
−3 R
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
IRF to a 10% households LTV shock
10 20 30 40−5
0
5
10x 10
−3 gdp
10 20 30 40−0.01
0
0.01
0.02c
10 20 30 40−5
0
5
10
15x 10
−3 i
10 20 30 40−1
0
1
2
3x 10
−3 pi
10 20 30 40−4
−2
0
2
4x 10
−3 l_f
10 20 30 40−0.05
0
0.05
0.1
0.15l_h
10 20 30 40−6
−4
−2
0
2x 10
−4 spread_f_l
10 20 30 40−6
−4
−2
0
2x 10
−4 spread_h_l
10 20 30 40−5
0
5
10x 10
−3 R
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
IRF to a 10% firms LTV shock
10 20 30 40−1
0
1
2
3x 10
−3 gdp
10 20 30 40−1
0
1
2
3x 10
−3 c
10 20 30 40−5
0
5
10
15x 10
−3 i
10 20 30 40−2
0
2
4x 10
−4 pi
10 20 30 40−0.05
0
0.05
0.1l_f
10 20 30 40−2
0
2
4
6x 10
−3 l_h
10 20 30 40−4
−2
0
2x 10
−4 spread_f_l
10 20 30 40−4
−2
0
2x 10
−4 spread_h_l
10 20 30 400
1
2
3
4x 10
−4 R
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
IRF to a spread on loans to HHs shock
10 20 30 40−5
0
5
10x 10
−4 gdp
10 20 30 40−1
0
1
2x 10
−3 c
10 20 30 40−1
0
1
2x 10
−3 i
10 20 30 40−2
0
2
4x 10
−4 pi
10 20 30 40−4
−2
0
2x 10
−4 l_f
10 20 30 40−5
0
5
10
15x 10
−3 l_h
10 20 30 40−1
−0.5
0
0.5
1x 10
−4 spread_f_l
10 20 30 40−3
−2
−1
0
1x 10
−3 spread_h_l
10 20 30 40−5
0
5
10x 10
−4 R
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
IRF to a spread on loans to firms shock
10 20 30 40−5
0
5
10
15x 10
−5 gdp
10 20 30 40−6
−4
−2
0
2x 10
−4 c
10 20 30 400
1
2
3x 10
−3 i
10 20 30 40−2
−1
0
1x 10
−5 pi
10 20 30 400
1
2
3
4x 10
−3 l_f
10 20 30 40−2
−1
0
1
2x 10
−4 l_h
10 20 30 40−4
−3
−2
−1
0x 10
−3 spread_f_l
10 20 30 40−10
−5
0
5x 10
−6 spread_h_l
10 20 30 40−2
0
2
4x 10
−5 R
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Plan of the Presentation
1 Motivation
2 Model
3 Calibration
4 Preliminary Results
5 Summary
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Simulation description
The goal of the simulation is to find the impact of the creditcrunch alone.
Shocks to spreads set to match values to match observedchange in spreads.
Shocks to LTVs set to match (given the previously calibratedinitial shocks to spreads) observed decline in loans.
Persistence set at reasonable values.
Two questions:
do we need shocks to LTVs?does the credit crunch alone have an important effect?
Given that the crudeness of the calibration results should beonly treated as a documentation of the model’s possibilities.
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Spreads in the model
0 5 10 15 20 25 30 35 40-5
0
5
10
15
20x 10
-3
spread_h_lspread_f_l
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Spreads as a source of the credit crunch in the model
0 5 10 15 20 25 30 35 40-0.08
-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0
0.01
l_hl_f
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Spreads and LTVs as a source of the credit crunch in themodel
0 5 10 15 20 25 30 35 40-0.35
-0.3
-0.25
-0.2
-0.15
-0.1
-0.05
0
0.05
l_hl_f
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Credit crunch and GDP in the model
0 5 10 15 20 25 30 35 40-20
-15
-10
-5
0
5x 10
-3
gdp
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Plan of the Presentation
1 Motivation
2 Model
3 Calibration
4 Preliminary Results
5 Summary
M. Brzoza-Brzezina and K. Makarski Credit Crunch
Motivation Model Calibration Preliminary Results Summary
Summary and work for the nearest future
Model seems to be very promising in terms of its ability toreplicate the credit crunch impact.
A proper calibration or Bayesian estimation needs to be done
It should result in a proper decomposition of shocks.
Financial frictions seem to have an important effect on theeconomy.
Spreads alone may not be enough to quantitatively replicatethe observed behavior of financial variables thus also shocks toLTVs are needed.
Possibly policy analysis.
M. Brzoza-Brzezina and K. Makarski Credit Crunch