Chapter

Appendix I Growth in Banks’ International Assets

Author(s):
John Lipsky, Peter Keller, Donald Mathieson, and Richard Williams
Published Date:
July 1983
Share
  • ShareShare
Show Summary Details

Alternative Measures of Bank Claims

BIS Quarterly Press Release on International Banking Developments

The main data series used in this study to analyze international banking flows is the BIS measure of “net international bank credit,” which is given in its quarterly reports on “international banking developments.” International bank credit thus defined represents total foreign lending by banks in the BIS (quarterly) reporting area41 after (partly estimated) adjustment to exclude double counting due to redepositing among reporting banks. The BIS also supplies data on the gross foreign claims of banks in the reporting area (including redeposits). Table 24 shows both the gross and (as a memorandum item) the net series. Gross claims are shown on a territorial basis42 in Table 15.

Table 24.Gross Foreign Claims of Banks Inside and Outside BIS Reporting Area, 1977–821(In billions of U.S. dollars; and in percent)
197719781979198019811982
Gross foreign claims of banks in BIS reporting area6908931,1111,3221,5491,687
Growth in percent219.931.124.419.017.28.9
Eurocurrency market3385502640751847867
Growth in percent22.630.427.517.312.82.4
Other305391471571702820
Growth in percent216.532.120.521.223.016.8
Gross claims or banks in certain offshore centers outside BIS reporting area477107135175236
Growth in percent39.026.228.134.9
Gross foreign claims of deposit money banks, IFS series8661,1361,4381,7492,1122,238
Growth in percent25.531.226.621.620.86.0
Memorandum items:
Net foreign claims of banks in the BIS reporting area4305356658109451,020
Growth in percent222.725.624.321.816.77.9
Redepositing among banks in the BIS reporting area260358446512604667
Growth in percent215.640.024.614.818.010.4
Sources: Bank for International Settlements; and International Monetary Fund, International Financial Statistics (IFS).

Gross claims include, and net claims exclude, redepositing among banks within the BIS reporting area.

Growth rates do not always correspond to stock figures because of breaks in the data series.

Foreign currency claims of banks in Europe.

Claims of non-U.S. banks in the Bahamas, the Cayman Islands, Hong Kong, Panama, and Singapore, and claims of all banks in Bahrain and the Netherlands Antilles.

Sources: Bank for International Settlements; and International Monetary Fund, International Financial Statistics (IFS).

Gross claims include, and net claims exclude, redepositing among banks within the BIS reporting area.

Growth rates do not always correspond to stock figures because of breaks in the data series.

Foreign currency claims of banks in Europe.

Claims of non-U.S. banks in the Bahamas, the Cayman Islands, Hong Kong, Panama, and Singapore, and claims of all banks in Bahrain and the Netherlands Antilles.

Until 1981, interbank claims of banks within the BIS reporting area tended to grow at a rate less than or equal to the rate of growth of the sum of claims on nonbank borrowers in the reporting area and on bank and nonbank borrowers outside the area; only in 1978 did such interbank claims grow at a rate substantially faster than claims net of redeposits among reporting banks, and this has been attributed to turbulence in exchange markets. In 1981, and again in 1982, inter-bank credits within the (quarterly) reporting area grew faster than “net” lending. The creation of International Banking Facilities (IBFs) in the United States in 1981 would have contributed to the more rapid growth of interbank claims to the extent that there was a shift of assets of non-U.S. banks from offshore branches to the IBFs.

The BIS quarterly series of gross claims include data on the narrowly defined Eurocurrency market43 (also shown in Table 24), which indicate that this market has usually grown more slowly than the foreign claims of banks in the entire reporting area (although the Euro-currency market still accounts for more than half of foreign claims of banks in the reporting area).

BIS Semiannual Press Releas e on Maturity Distribution of International Lending

The BIS also reports semiannually on the aggregate external position of banks located in the Group of Ten countries, Switzerland, Austria, Denmark, and Ireland, and a number of their affiliates domiciled in other countries, vis-a-vis countries outside the reporting area. A technical note is appended to the semiannual press release that covers in some detail the coverage, limitations of the data, and the differences in statistical coverage compared with the data in the BIS quarterly press release on “international banking developments.”

Data derived from the semiannual press release are given in Table 35. The claims of the reporting banks on individual countries are classified according to residual maturity—i.e., between claims with residual maturities of up to one year (shown in Table 35), over one, up to and including two years, and over two years, respectively.44 A maturity breakdown is not available in the case of external liabilities. In addition to the data on maturity distribution of claims, Table 35 shows the BIS data on undisbursed credit commitments, also provided on a country-by-country basis.

Important differences between the BIS semiannual data and those contained in the quarterly press release may be noted. First, the scope of the “reporting area in the semiannual release” is broader than that in the quarterly release: (1) besides the external claims (and liabilities) Of U.S. branches in the Bahamas, the Cayman Islands, Hong Kong, Panama, and Singapore, the include those of all other foreign affiliates of U.S throughout the world; (2) for the other reporting countries (except Italy in the case of liabilities) the data include the external assets and liabilities vis-a-vis countries outside the reporting area of some or all affiliates in a number of offshore centers.45 Second, the fairly larg e residuals with respect to claims contained in Table 7 of the BIS quarterly press release have been allocated to individual countries in the semiannual report.

Aggregate claims on countries “outside the reporting area ” are larger in the semiannual than in the BIS quarterly press release for the same periods; this difference increased in 1982. At the end of 1981 the stock of claims was $489.0 billion in the semiannual, and $461.6 in the quarterly, press releases; at the end of 1982 the totals were $532.0 billion and $491.2 billion, respectively, i.e., the total in the quarterly release declined from 94.4 percent to 92.3 percent of that in the semiannual release. The significance of this difference can readily be seen in the discrepancy between aggregate new lending to countries outside the BIS reporting area, particularly in the second half of 1982, on the basis of the two data series:46

Quarterly Series
1982
First halfSecond halfTotal
Current U.S. dollars (billion)13.416.229.6
Constant U.S. dollars47 (billion)26.512.539.0
Semiannual Series
1982
First halfSecond halfTotal
Current U.S. dollars (billion)17.327.743.0
Constant U.S. dollars47 (billion)30.322.652.9

Further Initiatives of Fund and BIS

Table 24 also contains two extensions of the series of gross claims of banks within the BIS quarterly reporting area. First, as noted above, the BIS quarterly reporting area includes branches of U.S. banks in five offshore banking centers.48 The BIS now also estimates the gross foreign claims of non-U.S. banks in those five offshore centers plus the gross foreign claims of all banks in Bahrain and the Netherlands Antilles. Although data for these non-U. S. banks are not yet available for 1982, In previous years the gross foreign claims of this group grew at a rate faster than those of banks in the reporting area. Second, the Fund’s publication International Financial Statistics (IFS) provides a data series on the foreign claims of deposit banks prepared on a broader basis than the BIS reporting area. The IFS series offers wider coverage, as it includes both the foreign claims of some deposit banks that are located in the BIS reporting area but not included in the BIS data and the foreign claims of deposit banks located outside the BIS reporting area. This IFS series has tended to grow somewhat faster than the BIS gross claims series.

The BIS has for some years provided a geographic analysis of the foreign claims (and liabilities) of banks in the BIS reporting area. This has been a useful source of information on the external indebtedness of reported countries (which comprise most countries of the world). Working closely with the BIS, the Fund extended the range of the BIS statistics to include a bank/nonbank classification of this geographic analysis, and the BIS has recently begun to publish that analysis aggregated over the 13 reporting countries for which such an analysis is available.

The Fund, in collaboration with the BIS, is now seeking regular returns that give a geographic analysis of the bank/nonbank classification of their deposit banks’ foreign claims and liabilities from a wider reporting area, covering 38 international banking centers. The centers include the 15 countries that already report to the BIS (these countries are already providing data to the Fund), together with other centers that have not been represented, or that have been only partially represented, in the BIS data. Collectively the 38 centers account for 99 percent of the world total of deposit banks’ foreign assets as identified in IFS. All but 2 have already agreed to provide data to the Fund. Owing to their wide coverage, the data provided under these arrangements will provide better estimates of the recourse of all countries of the world to the international banking system.

Econometric Analysis of the Growth in Claims

In the last two International Capital Markets reports,49 the growth of the gross international assets of banks (including redepositing) was analyzed with the help of an econometric model. The model related the quarterly growth rate of gross external assets of banks (GRQ) to the quarterly growth of world imports in U.S. dollars (GRWIMP), the quarterly growth rate of the sum of the U.S. dollar value of the GNP of the United States, Canada, Japan, the Federal Republic of Germany, and the United Kingdom (GRSUM), and the quarterly growth rate in the ratio of the sum of the absolute values of the trade deficits and surpluses of 56 countries to world imports (GRIMB). The latest reestimation of the single, reduced form equation of the model is based on a sample of data from the second quarter of 1970 to the second quarter of 1982.50 Thus, these estimates are not affected by the financial market disturbances of late 1982. The model was used to forecast the growth in international lending in 1983. Moreover, actual lending and that “predicted” by the model for the second half of 1982 were compared to see how well the model could forecast into a period of some turbulence in financial markets. In light of the model’s generally good predictive qualities in the past, any divergence between the actual and forecast growth rates perhaps could be interpreted as a rough measure of the impact of the financial crises on “normal”market growth.

For 1982, the World Economic Outlook indicates that world imports (measured in terms of U.S. dollars) declined by 6.4 percent, the U.S. dollar value of the GNPs of the five industrial countries declined by 0.2 percent, and the absolute value of world trade deficits of surpluses and deficits relative to world imports fell by 9.9 percent. Using these values of the explanatory variables, the model “predicts” a growth of gross international bank assets of deposit banks for 1982 of 12 percent, i.e., a decline of 2.4 percentage points below the trend rate of growth inherent in the equation. The actual growth rate was approximately 10 percent. While the forecast for the growth of bank assets was quite close for the first half of the year (a forecast of 2 percent expansion versus a 1.9 per cent actual increase), the projected growth for the second half of the year exceeded the actual growth by a wider margin (a forecast of 10 percent growth versus an actual increase of roughly 8 percent). The current model is primarily demand oriented, hence it is not surprising that it will tend to overestimate the growth of banks’ international assets at times when the willingness of banks to continue lending to a number of countries was seriously impaired.

There is one sense in which the model may not have overestimated the growth of international bank lending. As noted earlier in the discussion of the differences between BIS semiannual and quarterly data, the semiannual data indicated somewhat more rapid growth of international bank lending in the second half of 1982 than did the quarterly data. In part, this reflected lending out of foreign affiliates not included in the quarterly survey and, in addition, some reporting banks rebooked loans from their domestic offices to foreign affiliates that were not included in the quarterly surveys. For these reasons, the reported quarterly data for the second half of 1982 may have understated the actual growth of bank lending. As the econometric model has not been modified to reflect this rebooking activity, the forecast of a 12 percent growth in international bank lending should be compared with the reported quarterly growth plus the expansion of lending out of foreign affiliates that would normally have taken place out of domestic offices.

The World Economic Outlook forecasts that the U.S. dollar value of world imports will grow by 1 percent in 1983. The total nominal GNP measured in U.S. dollars of the five major industrial countries included in the model is anticipated to rise by 6.2 percent in 1983, while the payments imbalance ratio is expected to decline by 10.3 percent. The projected recovery of world trade and GNP growth will tend to increase the demand for trade-related financing and thereby will stimulate the growth of gross external assets. However, the decline in overall payments imbalances is likely to slow the expansion of international lending by reducing the need for balance of payments financing. Under the assumption of normal financial market conditions, the model predicts an increase of nearly 18 percent in the gross external assets of banks. This projection includes the forecast of a relatively slow expansion (4.4 percent) of gross external assets during the first half of 1983 but more rapid growth (12.8 percent) during the second half of the year.

However, in the Fund staffs judgment, it appears unlikely that such a sharp acceleration in the growth of banks’ gross external assets will be forthcoming in 1983. In part, a divergence between the projected expansion for 1983 and the actual growth may be taken as a broad indication of the size of the impact of the recent financial disturbances on international bank lending. While there is potential demand to support a significant increase in international bank lending, and while banks could technically provide such an increase, banks in present circumstances appear to have adopted a restrictive stance on international lending to developing countries for 1983. Such supply factors are, however, not explicitly included in the model and the model does not attempt to project flows to various analytical country groupings or to reflect explicitly developments in the international bond markets.

Data reported by banks in the Group of Ten countries, Switzerland, Austria, Denmark, Ireland, and the branches of U.S. banks in the Bahamas, the Cayman Islands, Panama, Hong Kong, and Singapore.

That is, figures for banks in the various BIS reporting countries (e.g., United States and Japan) cover all reporting banks operating in those countries, both those owned by residents and those that are foreign owned.

That is, the externel positions, denominated in foreign currency of banks in the European sector of the BIS reporting area.

For claims arising out of rollover credits the residual maturity is calculated on the basis of the latest data on which repayment is due to be made by the borrower.

Defined by the BIS as the Bahamas, Barbados, Bahrain, Bermuda, the Cayman Islands, Hong Kong, Lebanon, Liberia, Netherlands Overseas Territories, Panama, Singapore, Vanuatu, and other British West Indies.

Aggregate new lending to countries outside the BIS reporting area is the difference between the stocks at the beginning and the end of the period.

The information on flows in constant dollar terms is given in the text of the press releases. In addition to new lending outside the reporting area, the quarterly release provides data on such lending with in the reporting area and on unallocated residuals (which are allocated in the semiannual report). In constant dollar terms, the unallocated flows were $3.5 billion in the first half and $6.5 billion in the second half of 1982.

The Bahamas, the Cayman Islands, Hong Kong, Panama, and Singapore.

International Monetary Fund, International Capital Markets: Developments and Prospects, 1982, Occasional Paper No. 14 (July 1982), and International Capital Markets: Recent Developments and Short-Term Prospects, 1981, Occasional Paper No. 7 (August 1981).

For the period from the second quarter of 1970 through the second quarter of 1982, the ordinary least-squares estimates of this relationship are

GRQ = a0 + a1 GRWIMP + a2 GRSUM + a3 GRIMB + a4 HERS + a5 Q4 + a6 Ql

a0a1a2a3a4a5a6
Coefficient2.590.130.680.12–5.105.84–2.35
t-value3.722.003.534.062.175.812.89

R2 = 0.76, R̄2 = 0.72, D.W. = 2.31

where HERS =dummy variable with one in third quarter of 1974

to represent aftermath of collapse of the Herstatt Bank
Q4 =fourth-quarter dummy
Q1 =first-quarter dummy

As a comparison, the results for a sample through the fourth quarter of 1980 are

a0a1a2a3a4a5a6
Coefficient2.880.140.620.11–5.405.52–2.41
t-value3.031.942.543.352.164.982.64

R2 = 0.73, R̄2 = 0.68, D.W. = 2.33.

    Other Resources Citing This Publication