‘Running for the exit’: How cross-border bank lending fell

Neeltje van Horen

Cross-border bank lending fell dramatically in the aftermath of Lehman Brothers’ failure as funding constraints forced banks to reduce their foreign exposures. While this decline was partly driven by lower demand for international bank credit, it was substantially aggravated by a retrenchment of international banks from cross-border lending. But banks did not cut their cross-border lending in a uniform manner. Instead, they reallocated their foreign portfolios towards countries that were geographically close, in which they had more experience, in which they had close connections with domestic banks or in which they operated a subsidiary. The crisis thus showed that deeper financial integration is associated with more stable cross-border credit when large global banks are hit by a funding shock.

How do banks contract their cross-border lending?

Syndicates – groups of financial institutions that jointly provide large loans – are a key source of cross-border finance for firms in both developed and emerging countries. In 2007, international syndicated loans made up over 40% of all cross-border debt funding of US borrowers and more than two-thirds of cross-border flows to emerging markets. But in the year following the collapse of Lehman Brothers syndicated cross-border lending to private borrowers shrank by 58% according to data from Dealogic Loan Analytics.

As Figure 1 illustrates, the magnitude of this reduction differed substantially across countries. While 46 out of 59 recipient countries experienced a reduction in lending, the decrease varied from a drop of 1 percent to a complete lending stop. Furthermore, 13 recipient countries actually experienced and increase in cross-border flows they received after the collapse of Lehman Brothers. While this partly reflects differences in the adjustment of economic activity and credit demand, I show in a paper co-authored with Ralph De Haas (published in the Review of Financial Studies) that it also reflects variation in how banks reduced their credit-supply across countries.

If banks face a funding or other shock which leads them to cut cross-border lending and focus on domestic credit, they have to decide how to allocate the contraction across different foreign markets. Broadly speaking, international operating banks can follow one of three strategies.

First, they could simply cut cross-border lending uniformly across the board. A second strategy is to cut their credit supply on the basis of broad recipient country characteristics that are common across all lenders, such as political or macroeconomic risk. Whilst this second strategy implies that banks withdraw more from some countries relative to others, both strategies imply that all banks adjust their portfolio by the same magnitude in each recipient country, i.e. there is a generalized run for the exit.

But they could adopt a third strategy, where they cut lending on the basis of characteristics that are specific to bank-country pairs. Banks compare the ‘franchise’ value of future cross-border lending to determine where to retrench more and where to reduce credit less.  Unlike the first two strategies, under this approach two foreign lenders might react differently to a given borrower, and there may not be a generalised run for the exit.

Economic theory suggests several reasons why different banks might react differently to a given lender. One reason is that banks find it easier to overcome information asymmetries when they are geographically closer to borrowers. This channel becomes especially important when default risk increases during a crisis. Another reason is that if a bank has established lending relationships in a country it may also possess local market power, which it can exploit during a crisis. Related, banks with significant experience in a country can lend at lower costs as they know the local business sector well, know more domestic banks they can co-lend with, and are familiar with the legal, institutional, and accounting environment. If banks withdraw less from countries that are relatively ‘close’ to them in a geographic sense or in terms of lending relationships, they will adjust their portfolios differently for each recipient country.

Empirical evidence from syndicated loan data

A useful way to study how lending contracted is by looking at adjustments in syndicated lending. For each syndicated loan all lenders and the borrower are known. So we can observe how much each bank lends to each country in the year after the collapse in Lehman Brothers and compare it to its lending to that country in the year before. As multiple banks lend to the same country we can control for changes in credit demand and how much the country was affected by the financial crisis. We can thus analyse how a particular bank – given a certain funding shock – changes its lending to a particular country compared to another bank.

Using a dataset of syndicated cross-border lending by 117 international banks to 59 recipient countries we find that during the financial crisis international banks did not cut cross-border lending in an indiscriminate manner. In other words, there was no overall run for the exit. Rather, and in line with economic theory, banks followed the third strategy.

We find that bank-borrower closeness was strongly related to the resilience of cross-border credit. Banks continued to lend more to countries that were geographically close, where they were integrated in a network of domestic co-lenders, and where they had built up more lending experience. Banks that operated a local subsidiary were more stable providers of cross-border credit too, in particular in countries with weaker institutions.

What did the crisis reveal about the resilience of cross border lending?

The global financial crisis provided a stark natural experiment in how international credit contractions play out. It demonstrated that cross-border lending can be very unstable in times of stress which can have important negative consequences for firm performance and the real economy. However, alongside the aggregate contraction, the crisis also demonstrated that not all cross-border lending is equally unstable. This yields several important insights:

First, when international banks that provide cross-border bank lending are more entrenched in the recipient country they are less likely to cut lending. This entrenchment naturally results from cultivating long-term lending relationships with domestic banks and firms.

Second, a local presence in terms of a subsidiary or branch also tends to increase lending stability, especially in countries with weaker institutional environments. Not only do foreign-bank subsidiaries provide a relatively stable credit source themselves in these countries, their presence also tends to stabilize cross-border lending flows provided by the parent banks.

Third, a country’s vulnerability to capital outflows depends on the geographical proximity and experience of its creditors. For countries and firms that depend on international banks that are remote and have less local experience, the risk of a significant homebound retrenchment when foreign creditors are faced by a funding shock will be higher.

The crisis thus showed that shocks to the core of the global financial system can easily transmit to other parts of the world via a sharp reduction in cross-border lending. But when two countries are more deeply integrated financially, cross border lending flows between them are likely to be more resilient.

Neeltje van Horen works in the Bank’s Research Hub Division

If you want to get in touch, please email us at bankunderground@bankofengland.co.uk or leave a comment below.

Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.