Measuring the vulnerability of emerging market economies to sudden capital withdrawals through the banking system

BIS Quarterly Review  |  December 2011  | 
12 December 2011

(Extract from pages 16-17 of BIS Quarterly Review, December 2011)

The steady growth of foreign credit to emerging market economies has been one of the most significant regularities in international banking over the past couple of years. Nevertheless, some commentators have expressed concerns that the flow of capital to these economies may be adversely affected by the sharp rise in global financial volatility which began in the third quarter of 2011. How vulnerable were emerging markets on the eve of the turmoil?

The BIS international banking statistics can be used to construct four measures of the degree to which a country is vulnerable to sudden capital withdrawals through the banking system. First, the fraction of short-term international claims relative to total international lending measures the degree to which an economy is exposed to a non-renewal of short-term foreign bank credit to its residents. Second, the share of cross-border lending in total foreign lending sends a signal about the stability of funding from foreign banks since cross-border claims tend to be much more volatile than their locally booked counterparts.1 Third, the proportion of cross-border claims held in the form of tradable debt securities (as opposed to non-tradable loans) quantifies the ease with which foreign creditors could dispose of the claims they have on the residents of a given country. Finally, the foreign bank participation rate gives an indication of the fraction of total credit to non-banks in a given economy that is provided by foreign-owned banks.2 While none of these four indicators is perfect on its own, taken as a group they can paint a fairly informative picture of the vulnerability of various emerging market economies to sudden capital withdrawals.

Based on the first two indicators, Asia-Pacific appears to be the region most exposed to sudden capital withdrawals. As of the end of June 2011, close to two thirds (63%) of all international claims on residents of that region had a remaining maturity of less than one year (Graph A, righthand panel). In addition, cross-border claims represented more than half (52%) of all foreign lending to the area (Graph A, left-hand panel). Nevertheless, the signals sent by the other two indicators for Asia-Pacific were less worrying. First, foreign bank participation rates in the area were relatively low compared to those in the other three emerging market regions. In addition, debt securities represented only about a tenth of all cross-border claims on the region

Emerging Europe and Latin America and the Caribbean were mirror images of Asia-Pacific along three of the four dimensions of vulnerability examined in this box. Namely, they had substantially lower shares of cross-border claims (38% and 31%, respectively) and short-term claims (37% and 47%, respectively) than Asia-Pacific. By contrast, foreign bank participation rates were significantly higher across emerging Europe and Latin America and the Caribbean than in Asia-Pacific (Graph A, left-hand panel). The only dimension along which the three regions looked similar was the share of debt securities, which was relatively small in all of them. It was slightly over a fifth (21%) in Latin America and the Caribbean and barely over a tenth in emerging Europe (11%).

The picture in Africa and the Middle East was mixed. On the one hand, it was the only emerging market region other than Asia-Pacific for which the share of cross-border claims exceeded one half. On the other hand, the share of debt securities in the region was very small (5%). Furthermore, foreign bank participation rates were much lower than in emerging Europe, and the share of short-term claims in international claims (46%) was substantially smaller than in Asia-Pacific.


1 See R McCauley, P McGuire and G von Peter, "The architecture of global banking: from international to multinational?", BIS Quarterly Review, March 2010, pages 25-37.
2 This variable is constructed by combining BIS data on cross-border credit and foreign credit with IMF data on domestic credit. See P McGuire and N Tarashev, "Bank health and lending to emerging markets", detailed description of the construction of the measure, BIS Quarterly Review, December 2008, pages 67-80, for a more.