The BIS consolidated international banking statistics

Press release  | 
02 February 2001

International bank claims stood at $7,484 billion at the end of the third quarter of 2000, a decline of $86 billion which was more than accounted for by the reduced dollar values of claims denominated in euros and other non-dollar currencies.(2)The data are based on banks’ worldwide consolidated international claims reported to the BIS, which net out inter-office claims.(3)On the basis of these consolidated data, worldwide claims on non-bank borrowers stood at $3,972 billion, compared to $3,380 billion on a locational reporting basis. At the same time, consolidated claims on banks amounted to $3,512 billion, compared to $6,763 on a purely locational basis.

The $86 billion decline in total consolidated claims was the result of two opposing developments during the third quarter of 2000. Currency movements accounted for an underlying reduction in outstanding positions by about $245 billion when reported in US dollar terms, due to the lower value of euro and other non-dollar currency positions. The more limited decline in reported outstanding positions indicates, however, that this was partly offset by new bank lending of about $170 billion.

Claims on developed countries fell by $72 billion to $5,959 billion but, as explained above, this was more than accounted for by currency movements. On an exchange rate adjusted basis, there was in fact substantial new lending amounting to $180 billion. Claims on developing countries fell by $19 billion to $841 billion, with exchange rate movements responsible for two thirds of the reduction. Aggregate claims on Asia and Africa declined, while positions on Latin America were unchanged. Developing countries in Europe obtained some net additional financing once valuation effects are taken into account.

Substantial lending to telecoms in Europe

The currency valuation effect was most pronounced in claims on European developed countries - where about half of banks’ positions are denominated in euros. After adjustment for exchange rate movements, there was about $125 billion of estimated new lending. Rising bond spreads in the third quarter led telecommunications firms to return to short-term bank funding for bridge loans to finance third-generation wireless licences, while deals to finance mergers and acquisitions also rebounded.(4)Contractual claims on Switzerland increased by $18 billion, owing largely to new short-term interbank claims by US banks.

Rise in claims on US non-bank private borrowers and public borrowers

Long-term positions to finance the public and private non-bank sectors in the United States increased by an adjusted $30 billion. With over 80% of international banks’ cross-border claims on United States residents being denominated in US dollars, valuation effects had a smaller impact. Japanese banks increased their holdings of long-term public sector bonds by about $5 billion,(5)while German and Swiss banks increased short- and long-term lending to US corporates by $27 billion. Other European banks, in particular French institutions, reduced their outstanding positions vis-à-vis US residents, mostly banks, by $13 billion.(6)Concerns about rapidly rising spreads in the financial sector (towards end-2000 spreads on US bank bonds were approaching levels last seen during the financial market disturbances of autumn 1998) may have discouraged interbank lending. UK and French banks increased local positions in US currency by $21 billion and $12 billion respectively, more than offsetting a $13 billion reduction by Swiss and German banks.(7)

European banks raise claims on Cayman Islands

Total claims on offshore centres expanded by about $6 billion to $609 billion.(8)European banks, those from Germany in particular, raised their direct claims on the Cayman Islands by $11 billion, mostly to the non-bank private sector, under which hedge funds tend to be classified. Japanese banks’ claims on the Cayman Islands were unchanged, but shortened somewhat as $29 billion of long-term positions moved into the one- to two-year maturity bracket. Japanese banks’ positions on Singapore and Hong Kong fell by $4.7 billion each. In the case of Hong Kong, this cutback was largely offset by increased lending reported by UK and Swiss banks.

Developing countries in Asia repay short-term loans

Outstanding positions on developing countries in Asia declined by 3.5% to $274 billion. The 11% of outstanding positions denominated in euros accounted for about a third of the $10 billion decline, but loans, mainly short-term, amounting to $6 billion were also repaid. In an environment of rebounding foreign direct investment, corporations in China repaid $2.2 billion to banks from Japan, Europe and the United States. US banks’ positions with banks in South Korea declined by about $1.5 billion, while European and Japanese banks’ positions on private corporations in Thailand were cut by about $1.8 billion. These banks also reduced their lending to the banking sector in Taiwan. In contrast, US banks doubled their claims on Taiwan to $3.1 billion by increasing lending to banks by $1.6 billion. Spanish banks drive local expansion of euro area banks in Latin America

Cross-border and local foreign currency claims on developing countries in Latin America and the Caribbean remained largely unchanged at $281 billion. With 81% of these claims denominated in US dollars, exchange rate effects were limited. A $1.6 billion reduction in outstanding dollar positions was partly offset by slight increases in positions in euros, yen and unidentified other currencies.

Long-term claims on Mexico increased by $4 billion, mainly due to new lending to public and corporate counterparties. With the acquisition of a bank in Mexico, Spanish banks increased their outstanding international claims (including equity participations) by $4 billion and at the same time boosted local assets (and liabilities) in local currency by the equivalent of about $40 billion.(9)Spanish banks thus continued their five-year expansion in Latin America, decisively replacing US banks as the largest lenders in the region. Spanish banks had already overtaken German banks in their initial expansion phase during 1996- 97 (see graph).

Italian and Spanish banks provided Argentina with an additional $1.2 billion of short-term loans during the third quarter. In contrast, German banks cut their claims by $0.5 billion and obtained additional guarantees, reducing claims net of guarantees received from third parties (ie net risk exposure) by $1.6 billion. While US banks maintained their positions, they also obtained additional guarantees, reducing net risk exposure by $2.6 billion.

Maturity shifts elsewhere in Latin America

The share of short-term claims on Chile rose to 42% of the total, compared with 33% at the end of 1999, but these claims were still more than covered by foreign exchange reserves. Claims on Peru remained at $13 billion, but the short-term component rose to 73% of the total, compared with 63% at the end of 1999, as the maturity of Spanish and Italian banks’ positions shortened. In contrast, the maturity profile of positions on Venezuela lengthened, with corporations reducing debt to US and Italian banks by $0.8 billion.

Short-term positions on Brazil, in particular the corporate sector, fell by $2.7 billion, with loans to European and Japanese banks being repaid. Improved market confidence in Brazil’s economic performance and declining external debt were reflected in narrowing spreads on international sovereign debt issues. Standard & Poor’s recently upgraded their Brazil rating from B+ to BB-.

Claims on Turkey rise, restructuring in Russia

Banks extended new lending of $2.5 billion to developing countries in Europe, after taking into account the valuation effect on the 35% share of positions denominated in euros. Claims on Turkey rose by $1.5 billion to $44 billion, largely as a result of European banks increasing their short-term interbank claims on Turkey. Claims on Russia declined by $4.5 billion, with mostly long-term claims of German banks accounting for about $3 billion and claims of Italian banks for another $1 billion of the reduction. In the context of debt relief on Soviet-era debt by the London Club of commercial creditors, claims on Russian banks declined by $7 billion altogether, while restructuring of outstanding credits in the form of new bonds partly explained the rise in claims of about $1 billion each vis-à-vis corporations and the public sector. As a result of the restructuring, claims on banks in Russia dropped from 60% to 50% of total contractual claims. The increase in the share of the corporate sector to 38% was, however, partly due to new lending.

Claims in euros on Africa and the Middle East decline

Banks’ loans to Africa and the Middle East declined by $4 billion, of which $2.5 billion was due to valuation effects. About 25% of outstanding positions are denominated in euros, and these fell by nearly $5 billion, even in valuation-adjusted terms. Most of the decline was, however, offset by increased lending in dollars and other currency positions. French and German banks accounted for much of the decline, with reductions vis-à-vis public and corporate borrowers partially offset by new lending to banks. In Iran, European banks extended new credits to replace maturing claims. In consequence, the proportion of short-term claims continued to rise, from 65% of the total at end-1999 to 74% in the third quarter of 2000.

1The statistics are available on the BIS website ( and will be reproduced in the statistical annex of the BIS Quarterly Review: international banking and financial market developments, to be released on 5 March 2001. The BIS international banking data are also included in the quarterly release of the joint BIS-IMF-OECD-World Bank statistics on external debt (
2 In contrast to the previous quarter, when euro and yen exchange rates ended very close to where they had started out, the declines over the third quarter were substantial, so that valuation effects on the stock of outstanding claims need to be taken into account. Although the consolidated banking statistics do not contain detailed information on the currency composition of bank claims, the impact of exchange rate fluctuations can be estimated using currency information which is available from the BIS locational banking statistics (which are collected on the basis of the residence of the reporting banks and will be published in the forthcoming BIS Quarterly Review on 5 March 2001).
3The consolidated banking statistics are based on a different reporting population from the locational statistics. For a detailed explanation, see: "A tale of two statistics: the BIS locational and consolidated international banking statistics", in BIS Quarterly Review, June 2000, p 16.
4 See "Syndicated credits in the third quarter of 2000", in BIS Quarterly Review, November 2000, p 17.
5 Based on data on banks’ holdings of securities from the locational banking statistics, which are currently becoming available, and on comments provided by reporting central banks.
6 An additional $7 billion decline was due to a change in the reporting population of Dutch banks.
7 The sectoral composition of this type of lending is not reported.
8 Valuation effects on positions in offshore centres were minor. Banks’ consolidated claims on offshore centres are only about half of what is reported on a residence basis in the locational banking statistics, because lending intermediated by banks’ own offices in offshore centres is attributed to the country in which the final borrower resides rather than to the offshore centre itself. This difference might appear to limit the ability to estimate the currency composition using the locational banking data. However, since positions in euros or yen are comparatively small and positions in US dollars dominate, valuation adjustments are unlikely to play a big role.
9 Local currency positions of reporting banks’ foreign affiliates with local residents are excluded from the data shown in Tables 1-6 and 8, but are shown in Table 7.