Download Research on global financial stability: the use of BIS international PDF

TitleResearch on global financial stability: the use of BIS international
File Size3.9 MB
Total Pages265
Table of Contents
                            Research on global financial stability: the use of BIS international financial statistics
List of participants
The long or short of it: determinants of foreign currency exposure in external balance sheets
	1. Introduction
	2. Analytical issues
		2.1 Conceptual framework
		2.2 Moving from theory to empirics
		2.3 Components of the net foreign currency asset position
	3. Data
	4. Foreign-currency exposure: decomposition
	5. Econometric analysis
		5.1 Regression specification
		5.2 Results for FXAGG
			5.2.1 Pooled estimation
			5.2.2 Time series variation
			5.3 Results for FXDEBTAGG
			5.4 Results for subcomponents and NETFX
	6. Summary of stylised facts
	7. Conclusions
The US dollar shortage in global banking
	The long and short of banks’ global balance sheets
		Banks’ global expansion
		Banks’ foreign currency positions
		Maturity transformation across banks’ balance sheets
		The shortage of US dollars
		Concluding remarks
Networking financial centres: What BIS international financial statistics tell us
	1. Introduction
	2. A brief survey of the literature
	3. Our data
	4. Some simple algebra
	5. Our main results
		5.1 The size of the cross-border market for interbank deposits
		5.2 In- and out-degrees: a look at the microstructure of the market
		5.3 Who stands at the centre of the market?
International banking centres: a network perspective
	From size to network structure
	Identifying international banking centres by network methods
Banking globalization and monetary transmission
	1. Introduction
	2. Identification strategy
	3. The data
	5. Economic effects of banking globalization
	6. Conclusions
	Data appendix
Is there a cross-border bank lending channel? Evidence from US banks’ international exposure
	1. Introduction
	2. Hypothesis and methodology
		2.1 Two-step estimations of the effect of monetary policy on foreign claims
		2.2 Estimations of the effect of monetary policy controlling for host-country conditions
	3. Data
	4. Results
		4.1 Two-step estimations of the effect of monetary policy on foreign claims
		4.2 Cross-border claims and monetary policy at the bank-country level
		4.3 Economic significance
		4.4 Is there a balance sheet channel?
	5. Conclusions
Foreign asset risk exposure, degree of internationalization,and performance: An analysis of Canadian banks
	1. Introduction
	2. Literature review
	3. Framework for testing the DOI–performance relationship
		3.1 Econometric concerns
	4. Data and descriptive statistics
	5. Regression results
	6. Conclusions and policy implications
What lies beneath the euro’s effect on financial integration? Currency risk, legal harmonization, or trade?
	1. Introduction
	2. Empirical specification and data
		2.1 Specification
		2.2. Data
			Dependent variable
			Exchange rate regime
			Legislative harmonization
	3. Results
		3.1. Total effect138
		3.2. Channels
			Exchange rate risk
			Legislative Harmonization
			All channels
			Sensitivity Analysis
	4. Conclusion
	5. Data appendix
The geographical composition of national external balance sheets: 1980–2005
	1. Introduction
	2. Data construction
		2.1 Country selection and treatment of financial centres
		2.2 General approach for FDI, equity, and debt
		2.3 FDI
			2.3.1 Data
			2.3.2 Estimation
		2.4 Equity
			2.4.1 Data
			2.4.2 Estimation
		2.5 Debt
			2.5.1 Data
			2.5.2 Estimation
		2.6 Reserves
	3. A look at the data
		3.1 Financial network – undirected
		3.2 Financial network – directed
		3.3 Financial network – asset composition
		3.4 Comparison with the trade network
	4. Conclusions
	Appendix: Measures of network centrality
Openness and geographic neutrality: How do they contribute to international banking integration?
	1. Introduction
	2. Integration indicators: definitions and properties
		2.1 Degree of financial openness
		2.2 Degree of regularity of direct financial connections
		2.3 Degree of regularity of total financial connections
		2.4 Degree of financial integration
		2.5 Other global indicators
		2.6 Bipartite decomposition of the factors affecting financial integration
	3. Data
	4. Empirical evidence: the integration of the international banking systems
		4.1 Degree of financial openness
		4.2 Degree of regularity of financial connections
		4.3 Degree of financial integration
		4.4 Global indicators
	5. On the relative positions between bank flows' directions
	6. Concluding remarks
Document Text Contents
Page 1

Committee on the Global
Financial System

CGFS Papers
No 40

Research on global
financial stability: the use of
BIS international financial
Proceedings of the second CGFS workshop held at the BIS in
December 2008, chaired by Aviram Levy of the Bank of Italy

June 2010

Page 2

Copies of publications are available from:

Bank for International Settlements
CH-4002 Basel, Switzerland

E-mail: [email protected]

Fax: +41 61 280 9100 and +41 61 280 8100

This publication is available on the BIS website (

© Bank for International Settlements 2010. All rights reserved. Brief excerpts may be
reproduced or translated provided the source is cited.

ISBN 92-9131-832-9 (print)

ISBN 92-9197-832-9 (online)

Page 132

Table 3 shows the sum of the coefficients and standard errors on the monetary policy
measure from the second stage estimations outlined in (2). The table is divided into two
panels. Panel B differs from Panel A, in that it includes the change in log real GDP and its
four lags (not shown) as regressors in the second stage of the estimation. We use the
effective (nominal) federal funds rate and the real federal funds rate, our preferred measure,
as monetary policy indicators. Higher values indicate tighter monetary policy.

Table 3

Two-Step Estimation of the Impact of Monetary Policy on Banks’ Foreign Lending

This table presents results from the two-step estimation methodology developed in Kashyap and
Stein(2000). In the first step (results not shown), we estimate the sensitivity of the banks’
international activity to their own liquid assets represented by the parameter  from the following

   

 


1 1


1 1

log log


k itit it k

it k ik itit

y y L iqu id ity C ap ita liza tion

N onperform ing A ssets F R B

    it

   

 

 

     

  

 

where y is the value of foreign claims in US dollars, i indexes countries, t indexes time. The second
step uses the estimated ’s from the previous model to estimate the following equation:



4 3

0 1

t k t k t j t

k j

r Trend Quarter u    
 

     

where r is a measure of monetary policy. The parameter of interest is  and it is shown in Panel A
for different measures of monetary policy and foreign claims. Panel B shows results for the same
parameter, but adding controls for the current growth rate of GDP and its four lags in the second
step of the empirical specification.

Panel A: Excluding GDP controls

Monetary policy indicator Total foreign claims Cross-border claims Foreign-office claims

0.104** 0.1009** 0.0425
Federal funds rate
(nominal) [0.0466] [0.0489] [0.0679]

0.1224** 0.1147** 0.0368
Federal funds rate
(nominal) [0.0551] [0.0488] [0.0706]

Panel B: Including GDP controls

Monetary policy indicator Total foreign claims Cross-border claims Foreign-office claims

0.1039** 0.1007* 0.0449
Federal funds rate
(nominal) [0.0485] [0.053] [0.0653]

0.1366** 0.135** 0.0338
Federal funds rate
(nominal) [0.0519] [0.0589] [0.0543]

Newey-West robust standard errors in brackets: *** p < 0.01, ** p < 0.05, * p < 0.1.

In Panel A, the coefficients on both monetary policy indicators are positive and significant for
total foreign claims and cross-border claims. This finding is consistent with a cross-border
bank lending channel. As monetary policy gets tighter, only banks with liquid balance sheets
are able to continue lending to their foreign borrowers. In contrast, the coefficients for

124 CGFS – The second workshop on the use of BIS international financial statistics

Page 133

foreign-office claims on local residents are positive but not significantly different from zero.
The latter result is not surprising. Cetorelli and Goldberg (2008), using data for a longer time
period, find that lending by foreign affiliates of US banks is less reliant on the parent banks’
balance sheet during periods of monetary tightening. As mentioned before, foreign offices—
in some cases—are able to collect deposits on foreign residents, making them less
dependent on the parent bank’s funding.

The results in Panel B are consistent with the findings in Panel A. The sum of coefficients for
the nominal federal funds rate do not differ significantly from the values observed when GDP
growth is excluded. The effect of monetary policy on cross-border lending becomes even
stronger in this specification when the real federal funds rate is used as proxy for the policy
indicator. The change in foreign-office claims remains not significant after we include GDP

Although these results are consistent across specifications, there could be unmeasured
effects that may be driving the correlation between cross-border claims and monetary policy.
The most relevant are the characteristics of banks’ foreign borrowers and the fluctuation of
the foreign demand for credit. In the next set of tests, we control for these factors.

4.2 Cross-border claims and monetary policy at the bank-country level

In this section, we describe a series of tests of the cross-border bank lending hypothesis that
use controls at the bank-country level and take into account cyclical movements in the
foreign demand for credit. We estimate (3) using detailed bank-country foreign claim
information. We want to assess whether US banks’ cross-border claims react to monetary
policy. To identify credit supply movements from changes in credit demand, we exploit
information on US banks’ foreign-office claims as a proxy for local demand conditions. In
addition, all estimations have bank-country fixed effects to control for unobserved lender-
borrower characteristics. The test relies on the variation within each bank-country
relationship and its correlation to US monetary policy.

Table 4 shows results from the main specification.15 The dependent variables are the
logarithmic transformation of real cross-border claims. In columns (1) and (3), foreign credit
demand is proxied by the value of US banks’ foreign-office claims on local residents. The
value of domestic deposit banks’ claims on the private sector is included as the control for
foreign credit demand in columns (2) and (4). The first two columns use the nominal federal
funds rate as the monetary policy indicator and the last two use the real federal funds rate.
We use all bank-country relationships with positive cross-border claims.

The last two rows in the table report the sum of coefficients on the monetary policy indicator
and its standard errors—clustered by bank. Our main hypothesis states that a cross-border
bank lending channel exists if the sum of these coefficients is negative. A tightening in
monetary policy in the US reduces the supply of credit to foreign residents through cross-
border lending. The sum of coefficients in all specifications is negative and significant. This
implies that we are not able to reject our hypothesis that banks reduce their level of cross-
border claims in tight money periods. This result confirms the findings in the two-step method
estimations, after controlling for the creditor country’s demand for credit.

15 All specification include (not shown) an indicator variable for mergers, and three bank-specific variables—

defined in section 3—to control for the banks’ balance sheet “health”: Capitalization, Liquidity, and

CGFS – The second workshop on the use of BIS international financial statistics 125

Page 264

256 CGFS – The second workshop on the use of BIS international financial statistics

Page 265

CGFS – The second workshop on the use of BIS international financial statistics 257

Similer Documents