Transcript Croatia`s Credit Growth in Comparative Perspective

Household Lending in Croatia:
a Comparative Perspective
Evan Kraft
Advisor to the Governor
Croatian National Bank
The views expressed in this paper are the author’s own and not
necessarily those of the Croatian National Bank.
Motivation
• Literature, both theoretical and empirical, points to links
between lending booms and banking and currency crises
• A previous paper at this conference (Kraft and Jankov
2005) noted that Croatia’s post-2001 lending boom may
be less dangerous than 1995-98 due to greater share of
household loans
• Croatia’s current lending boom has been associated with
current account problems. But Herrman and Jochem
(2005) argue that high current account deficits are
normally associated with the “catching-up” process and
should reduce as GDP/capita rises. Of course, this does
not mean that there are no dangers here!
• Both the growth rate of household loans and share in
GDP in Croatia are exceptionally high relative to other
transition countries
Main questions
• Are repayment rates on household loans still holding up?
And if so, how can this be reconciled with the rapid rate
of growth of household credit?
• Are there any special macroeconomic implications of this
heavy share of household lending? In particular, how
does consumption lending get reflected in balance of
payments issues?
• Is Croatia's seemingly high share of household loans
actually high in broader cross-country comparison, and
what determinants can be found for this high share?
M
ay
Au 00
g0
N 0
ov
-0
Fe 0
b0
M 1
ay
Au 01
g0
N 1
ov
-0
Fe 1
b0
M 2
ay
Au 02
g0
N 2
ov
-0
Fe 2
b0
M 3
ay
-0
Au 3
g0
N 3
ov
-0
Fe 3
b0
M 4
ay
-0
Au 4
g0
N 4
ov
-0
Fe 4
b0
M 5
ay
Au 05
g0
N 5
ov
-0
Fe 5
b06
Loan growth since 2000
50
40
30
to enterprises
to households
overall
20
10
0
-10
-20
do
ni
a
ua
ni
a
a
R
om
an
ia
ia
-M
on
te
ne
gr
o
Sl
ov
ak
R
.
Sl
ov
en
ia
Po
la
nd
ac
e
Li
th
La
tv
i
y
H
un
ga
r
to
ni
a
R
.
at
ia
ch
Es
C
ze
C
ro
Bu
lg
ar
ia
na
Al
ba
ni
a
eg
ov
i
er
z
M
ia
-H
Se
rb
Bo
sn
Household lending to GDP in
transition countries, 2005
40,0%
35,0%
30,0%
25,0%
20,0%
15,0%
10,0%
5,0%
0,0%
Current account and fiscal balance,
%GDP
0
-1
current account
consolidated general goverment
-2
-3
-4
-5
-6
-7
-8
-9
-10
2000
2001
2002
2003
2004
2005
Foreign debt, % GDP
90
80
70
60
50
FDI
Other sectors
Banks
Government
40
30
20
10
0
2000
2001
2002
2003
2004
2005
Policy measures to limit lending
growth and foreign debt
• 2003: “tax” on lending growth—if loans grow more than
4% per quarter, banks must purchase low-yield central
bank securities (twice as much as the overrun)
• 2003: prudential measures—banks with rapid credit
growth must retain a portion of dividends (effectively
higher capital adequacy)
• 2004: marginal reserve requirement—extra reserve
requirement on increases in banks’ foreign liabilities.
Rate initially 24%, now 55%
• 2005-6: retirement of some government foreign debt and
replacement with domestic debt
• 2006: guidelines on banks’ management of foreign
currency induced credit risk (FCICR)
-9
Se 9
p9
D 9
ec
-9
M 9
ar
-0
Ju 0
n0
Se 0
p0
D 0
ec
-0
M 0
ar
-0
Ju 1
nSe 01
p0
D 1
ec
-0
M 1
ar
-0
Ju 2
nSe 02
p0
D 2
ec
-0
M 2
ar
-0
Ju 3
n0
Se 3
p0
D 3
ec
-0
M 3
ar
-0
Ju 4
nSe 04
p0
D 4
ec
-0
M 4
ar
-0
Ju 5
nSe 05
p0
D 5
ec
-0
M 5
ar
-0
6
Ju
n
Past-due loans to households, %
3,50
3,00
2,50
2,00
1,50
1,00
0,50
0,00
Detailed break-down of past-dues
2,5%
Housing
Collateralized with real estate
Car purchase
Credit card
2,0%
Other
Total
1,5%
1,0%
0,5%
0,0%
Mar-04
Jun-04
Sep-04
Dec-04
Mar-05
Jun-05
Sep-05
Dec-05
Mar-06
Distribution of household loans by
disposable income
debt payments as % of disposable income
9,0
8,0
1999
2000
2001
7,0
2002
2003
2004
6,0
5,0
4,0
3,0
2,0
1,0
0,0
D1
D2
D3
D4
D5
D6
disposable income deciles
D7
D8
D9
D10
Prudential indicators still do not
show any obvious danger signs
• Of course, both loan-loss provisioning and
past-dues are backward-looking
• Certainly, in a recession, we would expect
increased past-dues and defaults.
Econometric strategy for crosscountry regressions
• Use large sample of 90 countries to
establish broad determinants of household
lending to GDP
• Then look at residuals for transition
countries
• Analyze these residuals with variables
available for transition countries only to
see if Croatia is “exceptional”
Determinants of household lending
•
•
•
•
•
•
•
GDP per capita
Macroeconomic stability, particular low inflation
Enterprise use of direct finance and equities
Institutional development and quality
Legal origin
Regional effects
Transition country effects
Basic messages of the first-stage
regressions
•
•
•
•
•
Strong explanatory power of GDP/capita
Inflation record also significant
Legal origin variables perform well
Corruption significant
Direct finance (market capitalization of
listed companies) nearly significant
Variables used for the secondstage regressions
•
•
•
•
Banking sector reform (EBRD index)
Privatization indices
Enterprise reform scores
Non-bank financial institution reform
Second-stage results
• Strong, significant estimates for banking
reform, privatization and enterprise reform
• Most interestingly, Croatia’s residuals are
no longer outliers.
Conclusion: explaining high
consumer lending in Croatia
• Victim of success: low inflation, strong banking
reform lead to high household lending
• Slow reforms: stagnant privatization in recent
years, slow enterprise reform, high and
unchanging corruption, continued weaknesses
in institutional environment
• Conclusion: EU accession provides an ideal
opportunity to speed up reform, which, among
other things, will make it easier for banks to lend
to enterprises and decrease some of the focus
on lending to households