BIS Working Papers
No 108
The determinants of private
sector credit in industrialised
countries: do property prices
matter?
by Boris Hofmann
Monetary and Economic Department
December 2001
Abstract
Episodes of boom and bust in credit markets have often coincided with
cycles in economic activity and property markets. The coincidence of these
cycles has already been widely documented in the literature, but few studies
address the issue in a formal way. In this study we analyse the
determinants of credit to the private non-bank sector in 16 industrialised
countries since 1980 based on a cointegrating VAR. Cointegration tests
suggest that the long-run development of credit cannot be explained by
standard credit demand factors. But once real property prices, measured as
a weighted average of real residential and real commercial property prices,
are added to the system, we are able to identify long-run relationships
linking real credit positively to real GDP and real property prices and
negatively to the real interest rate. These long-run relationships may be
interpreted as long-run extended credit demand relationships, but we may
also capture effects on credit supply. Impulse response analysis based on a
standard Cholesky decomposition reveals that there is significant two-way
dynamic interaction between bank credit and property prices. We also find
that innovations to the short-term real interest rate have a strong and
significant negative effect on bank credit, GDP and property prices.
BIS Working Papers are written by members of the Monetary and Economic Department of the Bank
for International Settlements, and from time to time by other economists, and are published by the
Bank. The papers are on subjects of topical interest and are technical in character. The views
expressed in them are those of their authors and not necessarily the views of the BIS.
Copies of publications are available from:
Bank for International Settlements
Information, Press & Library Services
CH-4002 Basel, Switzerland
E-mail: publications@bis.org
Fax: +41 61 280 9100 and +41 61 280 8100
This publication is available on the BIS website (www.bis.org).
© Bank for International Settlements 2001. All rights reserved. Brief excerpts may be reproduced
or translated provided the source is cited.
ISSN 1020-0959
i
Table of contents
1. Introduction.................................................................................................................................... 1
2. Credit, economic activity and property prices ............................................................................... 2
3. Data issues.................................................................................................................................... 9
4. Long-run relationhips................................................................................................................... 10
5. Dynamic interaction..................................................................................................................... 15
6. Conclusions................................................................................................................................. 26
References............................................................................................................................................. 27
Appendix................................................................................................................................................ 30
1
1. Introduction
1
Over the last two decades most industrial countries have experienced episodes of boom and bust in
credit markets. These credit cycles have often coincided with cycles in economic activity and property
markets. The coincidence of these cycles has already been widely documented in the policy-oriented
literature (eg IMF (2000), BIS, (2001a)), but few studies assess the relationship between credit
aggregates, economic activity and property prices in a formal way. In particular, the role of property
prices has not been explored in any great detail. This paper attempts to partially fill this gap, by
modelling the determinants of credit to the private non-bank sector as a function of economic activity,
interest rates and property prices for a sample of 16 industrialised countries.
2
Economic activity, interest rates and property prices may affect credit via both credit demand and
supply channels. Economic conditions and prospects determine consumption and investment demand,
and thus the demand for credit. On the other hand, changes in economic activity are reflected in firms’
cash flow position and households’ income. Cash flow and income determine the ability of firms and
households to repay their debts, so that changes in economic activity may also affect the willingness
of banks to extend credit. The state of economic activity may therefore also determine the supply of
credit.
Financing costs, represented by market interest rates, have a negative effect on credit demand. When
interest rates go up, loans become more expensive and loan demand is reduced. A monetary
tightening, reflected by an increase in interest rates, may also induce banks to cut back credit supply.
A reduction in credit supply may also arise from reduced creditworthiness of firms and households due
to a deterioration in their financial positions following a monetary tightening (balance sheet channel of
monetary transmission). A tightening of monetary policy, operated via open market sales by the
central bank, may also drain reserves and thus loanable funds from the banking sector, which may
also cause a reduction of loan supply (bank lending channel of monetary transmission).
Property prices may also affect both credit demand and credit supply. Property accounts for a
substantial share of household assets, so that changes in property prices may have a significant
wealth effect on credit demand. Since loans are often secured with real estate collateral, property
prices may also have a significant effect on the borrowing capacity of the private sector. An increase in
property prices increases the value of collateralisable assets and thus the creditworthiness of firms
and households. As a result, banks are more willing to extend loans, so that the supply of credit to the
private sector increases.
Thus, economic activity, interest rates and property prices may affect both credit demand and credit
supply. The problem of identifying demand and supply effects in the analysis of credit aggregates is
well known and is most likely one of the reasons why there are so few studies analysing the
determinants of credit aggregates. Nevertheless, we still think that it is important to understand which
factors drive the development of credit aggregates, even if it is not possible to clearly identify the
demand and supply effects.
3
Based on Johansen’s (1988, 1991, 1995) approach to cointegration analysis we show that the long-
run development of credit cannot be explained by standard credit demand factors, ie real GDP and the
real interest rate. But once real property prices, measured as a weighted average of real residential
and real commercial property prices, are added to the system, we are able to identify long-run
relationships linking real credit positively to real GDP and real property prices and negatively to the
real interest rate. Credit is found to adjust significantly to the cointegrating relationship, implying that
there is a long-run relationship linking credit to GDP, property prices and interest rates.
1
Boris Hofmann, Zentrum für Europäische Integrationsforschung, University of Bonn, e-mail: bhofmann@uni-bonn.de. I wish
to thank Claudio Borio, Charles Goodhart, Philip Lowe, Jürgen von Hagen and seminar participants at the Bank for
International Settlements for helpful comments and Steve Arthur and Wilhelm Fritz for their help with the data. The usual
disclaimer applies.
2
In the following we refer to credit to the private non-bank sector as private sector credit.
3
The identification problem could possibly be at least partly overcome if we were to model credit demand and credit supply
simultaneously. We do not, however, attempt to explicitly model a credit supply function, since time series data on important
credit supply factors, such as banking sector profitability or the degree of liberalisation and competition in banking markets,
are not readily available.
