1. The main text discusses the set of core and encouraged FSIs. This appendix sets out ideas that arose during the drafting of, and consultation on, the Guide for additional definitions of the FSIs and for related data series. Issues relevant for the monitoring of financial conglomerates are also discussed. Information provided in this appendix may be useful to compilers when developing FSIs for use in their own national context.
2. Chapter 6 brings together the concepts and definitions set out in Part I of the Guide to explain how each FSI for deposit takers is to be calculated. Additional definitions of some FSIs were proposed during discussions on the preparation of the Guide, which are set out below. In some instances, more disaggregated data series would be needed to compile these FSIs.
3. The Guide recommends that sector-level data compiled to calculate FSI ratios include any intrasector positions in debt and financial derivatives on a gross basis (paragraph 5.49). This approach allows the interrelationships among groups in the sector, and hence potential contagion risks, to be identified. However, for FSI ratios where gross assets or liabilities are either the denominator or the numerator-for example, return on assets and the capital-to-assets ratio-they could also be calculated excluding intra- sector positions in debt and financial derivatives, so that both the numerator and the denominator of the ratio exclude intrasector transactions and positions.
4. The debt-to-capital ratio is another measure of financial leverage that could be considered in addition to the capital-to-assets ratio.
5. Return on equity could be calculated including purchased goodwill in the denominator, which would amount to using a measure of capital and reserves closer to that used in commercial accounting.
6. With a view to providing a broader measure of nonperforming assets, this FSI could be calculated using total debt claims in the numerator and not just loans.
7. This FSI could also be calculated for resident and nonresident borrowers separately. The approach might be relevant in the context of differing economic circumstances prevailing in the domestic and foreign markets.
8. In economies where collateral is widely used, nonperforming loans net of provisions and collateral to capital is an alternative FSI that might give a more realistic picture of the potential for losses by deposit takers than the FSI ratio, which is calculated by excluding collateral. 1 Any dissemination of this ratio would need to be supplemented with detailed meta- data on collateral rules in use, including the valuation approach adopted by national supervisors.
9 The number of large exposures at various percentages of regulatory capital could be considered, such as the total number of individual large exposures above 10 percent but below 20 percent of regulatory capital, between 20 percent and 40 percent of Page 204 regulatory capital, and above 40 percent of regulatory capital.
10. To identify the location of the counterparties, the number of large exposures could be divided between resident and nonresident counterparties.
11. To monitor concentrated lending by deposit takers, as peer groups or as for the sector as a whole, FSIs could be constructed that relate to the sectoral- particularly by industry-and geographic distribution of loans. Indications of a buildup of concentrated positions derived from these data could allow compilers to specify sectors and/or countries for which more detailed information might be required.
12. Other approaches to monitoring concentrated lending include (1) specifying a minimum exposure amount in nominal terms at which any search for concentrated lending by deposit takers could begin, and (2) developing a credit concentration ratio (for example, the ratio of the total exposures to the largest 20 borrowers by each bank to the total exposures of banks).
13. Some economies rely on credit registers to monitor large exposures. Through such registers, the total exposure of the deposit-taking sector (and indeed of the financial system) to each individual borrower can be measured, and reports could, for example, be generated each quarter on the exposures to the 100 largest borrowers. An identification code attributed to each borrower would allow consistency of recording. However, the exposures of the foreign branches and subsidiaries of resident deposit takers might not be covered by such registers.
14. There may be analytical interest in presenting the net open position in equities by country to identify any large exposures to equity holding in particular economies.
15. This FSI could be calculated using very short- term liabilities-three months or less-as the denominator. Such liabilities would be closer to the liquidity concept used for liquid assets. Moreover, this FSI could be calculated excluding short-term customer deposits from short-term liabilities; that is, excluding those short-term liabilities considered to be a more stable, less volatile form of funding. This FSI could also be calculated excluding financial derivatives positions-that is, calculating the ratio taking short- term debt only into consideration-particularly if a net derivative asset position is significantly affecting the ratio.
16. To identify the sectoral concentration of NPLs, this FSI could be calculated for each sector (using the same sectors as in calculation of the sectoral distribution of loans to total loans).
17. A more disaggregated view of lending to the other financial corporations sector could be provided through dissemination of the ratios for loans to the five subsectors, 2 defined in Appendix VII, the Glossary of Terms.
18. An additional possibility is to classify loans by type of borrower using the International Standard Industrial Classification of all Economic Activities (ISIC). This approach might be particularly relevant when an economy has systemically important industries, such as petroleum and agriculture. The ISIC has 17 major categories of economic activity in the resident economy and places more emphasis on the type of activity undertaken than on the economic nature of the business, which is the basis of the sector distribution described in Chapter 2. The categories and short definitions of the activities covered in each category are set out in Box A3.1 of this appendix. An alternative approach is to classify loans by type, such as retail, commercial, and industrial.
19 If this FSI is compiled on a cross-border consolidated basis to also capture loans by deposit takers' branches and subsidiaries abroad, a complementary, but far more ambitious, approach would be to attribute loans by sector regardless of the residence of the borrower. For instance, total lending to nonfinancial entities worldwide, regardless of residence, could be compiled. In this way, exposures of deposit takers in the reporting population to similar activities worldwide could be monitored.
Box A3.1. The International Standard Industrial Classification of All Economic Activities (ISIC)
The ISIC is an industrial classification developed by the United Nations that groups establishments that have the same principal activity by industry. An establishment is defined as an enterprise, or part of an enterprise, that is situated in a single location and in which only a single productive activity is carried out or in which the principal productive activity accounts for most of the value added.
The industries identified in the ISIC are as follows:
- Agriculture, hunting, and forestry, including related service activities
- Fishing, including fish farming and service activities incidental to fishing
- Mining and quarrying, including service activities incidental to oil and gas extraction, excluding surveying
- Electricity, gas, and water supply
- Wholesale and retail trade, repair of motor vehicles, motorcycles, and personal and household goods
- Hotels and restaurants
- Transport, storage, and communications
- Financial intermediation
- Real estate, renting, and business activities -such as computer and related activities, and research and development
- Public administration