Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

Description

STUDIESREPORT OF THE WORKINGGROUP ON THE SENSITIVITY OFBANK BALANCE SHEETS TOBENCHMARK RATE CHANGES1(1992-1993 DATA)The working group on the sensitivity of bank balance sheets tochanges in benchmark rates was convened by the Banque de Franceunder the chairmanship of André Icard, Head of the ResearchDepartment. It comprises representatives of credit institutions andthe Banque de France.The group's objective is to improve the understanding of howmonetary policy decisions regarding benchmark rates affect the realeconomy, in particular through the reactions of the banking system.The role of credit institutions in providing financing for theeconomy has remained preponderant despite the transformations ofthe 1980s, which thoroughly modified the environment in whichbanks operate. The lending rates applied by financial institutions aretherefore still an essential economic variable, and as a result, creditinstitutions' reactions to changes in the level and structure ofbenchmark interest rates are a key factor in the transmission ofmonetary policy effects.JEAN-PIERRE BARILLASMonetary Resarch and Statistics DivisionMonetary Analysis and Statistics UnitThe objective of the working group differs from that of previous work on the subject. Its intention isnot to assess the interest rate risk exposure of credit institutions, nor to measure the effect of variationsin interest rates on their results or on the structure of their balance sheets.1This study was ...

Informations

• Publié par: Ewli
• Nombre de lectures: 16
• Langue: English

Extrait

STUDIES

REPORT OF THE WORKING GROUP ON THE SENSITIVITY OF BANK BALANCE SHEETS TO BENCHMARK RATE CHANGES (1992-1993 DATA) 1 The working group on the sensitivity of bank balance sheets to changes in benchmark rates was convened by the Banque de France under the chairmanship of André Icard, Head of the Research Department. It comprises representatives of credit institutions and the Banque de France. The group's objective is to improve the understanding of how monetary policy decisions regarding benchmark rates affect the real economy, in particular through the reactions of the banking system. The role of credit institutions in providing financing for the economy has remained preponderant despite the transformations of the 1980s, which thoroughly modified the environment in which banks operate. The lending rates applied by financial institutions are therefore still an essential economic variable, and as a result, credit institutions' reactions to changes in the level and structure of benchmark interest rates are a key factor in the transmission of monetary policy effects.

J EAN -P IERRE B ARILLAS Monetary Resarch and Statistics Division Monetary Analysis and Statistics Unit The objective of the working group differs from that of previous work on the subject. Its intention is not to assess the interest rate risk exposure of credit institutions, nor to measure the effect of variations in interest rates on their results or on the structure of their balance sheets.

1 This study was published in the Etudes Supplement to the Banque de France Bulletin No. 18 (June 1995).

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

33

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

The work launched in October 1992 with the participation of several credit institutions resulted in a first study centralizing data on nine large banks in Paris. It was published in the first quarter 1994 Studies supplement to the Banque de France Bulletin ("Etudes"), and the seventh issue of the English- language Digest. The present study of data collected in 1992 and 1993 is based on a wider sample of eleven large credit institutions, which together account for three-fourths of the bank credit distributed. The methodology jointly developed by the members of the working group for the first study remained unchanged for the second study. It is described in detail in appendice 2. The outstanding assets and liabilities of all credit institutions surveyed were broken down according to their terms of remuneration. The data was collected directly from the institutions, because the usual surveys provide little information relevant to this approach. The results give a more precise view of the sensitivity of bank balance sheets to interest rate variations. In addition, it is now possible to analyze developments over a period of time by comparing the present results with those of the previous year.

1. The macroeconomic context The study related to the twelve months between the fourth quarters of 1992 and 1993. This period was marked by an unprecedented decline in interest rates despite the tension on foreign exchange markets in the first half of 1993. Interbank rates plummeted between the last quarters of 1992 and 1993: the three-month PIBOR lost 4 basis points while the daily average weighted overnight rate (TMP) fell by 3.5 basis points, settling at average quarterly rates of 6.74% and 6.83%, respectively. At the same time, the average quarterly index of the underlying long-term securities of MATIF operations dropped from 8.18% to 5.85%, sharpening the previous years' decline in long-term interest rates, which eased by 134 basis points in 1991 and by 63 basis points in 1992. Nevertheless, the inversion of the term structure persisted throughout the period, with short-term rates higher than long-term rates. The gap narrowed somewhat in the second half of the year with a more rapid decline in short-term rates, allowing the beginning of a return to a more normal term structure. The difference between the three-month PIBOR and long-term rates on the pool of securities underlying the MATIF notional contract narrowed from about 260 basis points at the end of 1992 to less than 90 basis points in December 1993. Moreover, American long-term interest rates stopped declining in the last months of 1993 to veer abruptly upwards. Ten-year Treasury bond yields, which had fallen from 6.77% in December 1992 to 5.30% in October 1993, rose again to 5.75% two months later, indicating that an overall rate increase might be in the offing. And finally, it should be remembered that outstanding credits, whose growth had slowed significantly in the previous months, began falling in May 1993.

2. Results 1 2.1. Balance Sheet Structure The results available relate to eleven credit institutions (see appendices) and cover the last quarters of 1992 and 1993. At those dates, the credits distributed by the institutions within the sample accounted

1 It should be remembered that this study is not a comprehensive presentation of the impact of interest-rate variations on the results of credit institutions and ultimately on the economy. These consequences are numerous and contradictory, and greatly dependent on the average length of the credit and the speed with which banks adjust their lending rates.

34

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

for 76.0% and 75.3% of total credits distributed by banks. The year-on-year decrease in outstanding credits of 3.9% was slightly higher than the 3.0%-drop reported by banks overall. Sight and time deposits recorded under liabilities by the eleven banks accounted for 73.2% and 92.7%, respectively, of the total deposits collected by the banks at the end of 1992. One year later, their sight deposits accounted for 75.4% and time deposits for 87.6% of the total. The analysis of the restatements made according to the methods described in the appendice shows that the aggregate balance sheet total remained stable, although there were small variations in the relative amounts of the main items. In particular, the breakdown of balance sheet items into fixed or variable rates changed slightly. The portion of fixed-rate deposits tended to increase, while developments in the terms of household and business credits diverged. 2.1.1. Key Asset Characteristics Credits constitute the main component of the assets restated according to the relevant methodology. In 1993 as in 1992, they accounted for almost four-fifths of loans, not taking into account the bias arising on the net credit balance of refinancing transactions in 1993. Almost two thirds were at fixed rates, while 10% were indexed to the base lending rate and 20% to money market rates. The decline in the portion of credits indexed to the base lending rate and the matching increase in credits linked to money market indexes have sharpened markedly since the reform of the money market. This trend can be attributed to the introduction of negotiable debt securities in the mid-80s, which enabled companies to raise financing directly from underwriters and gave them access to bank credit at conditions close to market rates.

LOAN BREAKDOWN BY INTEREST RATE AND TYPE OF BORROWER AT END-1992 Households Companies Total FRF billions (%) FRF billions (%) FRF billions (%)

FIXED RATE........................................... 1,022.0 87.7 794.9 48.8 1 816.9 65.0 VARIABLE RATE................................... 143.7 12.3 835.5 51.2 979.2 35.0 – Base lending rate.................................. 20.6 1.8 271.7 16.7 292.3 10.5 – Money market rates ............................. 54.9 4.7 539.0 33.1 593.9 21.2 – Bond market rates ................................ 6.6 0.6 7.0 0.4 13.6 0.5 – Other.................................................... 61.6 5.3 17.8 1.1 79.4 2.8 TOTAL ................................................... 1,165.7 100.0 1,630.4 100.0 2,796.1 100.0

LOAN BREAKDOWN BY INTEREST RATE AND TYPE OF BORROWER AT END-1993 Households Companies Total FRF billion (%) FRF billion (%) FRF billion (%)

FIXED RATE........................................... 1'022.9 85.8 736.3 49.3 1,759.2 65.5 VARIABLE RATE................................... 169.7 14.2 756.9 50.7 926.7 34.5 – Base lending rate.................................. 18.9 1.6 234.4 15.7 253.3 9.4 – Money market rates ............................. 87.9 7.4 488.2 32.7 576.2 21.5 – Bond market rates ................................ 12.0 1.0 7.1 0.5 19.1 0.7 – Other.................................................... 50.9 4.3 27.2 1.8 78.1 2.9

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

35

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

TOTAL ................................................... 1,192.6 100.0 1,493.2 100.0 2,685.8 100.0 The tables breaking down outstanding credits by beneficiary reveal a major discrepancy in the loan conditions offered to the various types of non-financial agents. Almost 90% of household loans were granted at fixed rates, whereas approximately half of business loans bore variable rates. This breakdown varied little year on year. The main change related to household credit, where the portion of fixed-rate loans dipped slightly from 87.7% in 1992 to 85.8% at the end of 1993, while that of variable-rate loans rose correspondingly from 12.3% to 14.2% of the total loans granted. The increase in variable-rate loans was caused solely by loans linked to money market rates, which increased from 4.7% to 7.4% of total loans. Business credit showed the opposite trend, although the changes were far more limited, with a minimal decline in the portion of variable-rate loans from 51.2% to 50.7%. Outstanding credits calculated from the base lending rate, in particular, declined by one point from 16.7% of total credits in 1992 to 15.7% one year later. Total variable-rate loans shrank by only 9.4% against a 13.7% decline in loans linked to the base rate, reflecting the weaker demand for credit, which put borrowers in a better position to obtain market rates. It should also be noted that the decline in outstanding business loans observed in 1993 occurred at a time when their rate of self-financed investment taken in the broadest sense 1 exceeded 130%. The portion of loans indexed to the base lending rate (TBB) in total credits dropped by over one point from 10.5% to 9.4%. This can also be interpreted as resulting from lending institutions' tendency to prefer the instant adjustments provided by a money market index when granting loans, rather than the less flexible base lending rate. Moreover, the relative weight of the securities portfolios in the total assets of the banks included in the sample decreased in the wake of the decline of holdings of fixed-rate securities, which accounted for only 4.4% of the balance sheet total, compared to 6.0% at the end of 1992 (see appendice 1). 2.1.2. Key Funding Characteristics Traditional sources of funding represented the lion's share of the surveyed banks' liabilities, restated according to the methodology described in the appendice. Sight deposits accounted for somewhat less than one-fifth of liabilities, and time deposits for slightly under 40%. About one-sixth of banks' funding stemmed from certificates of deposit and medium-term negotiable notes (BMTN). This was more than the 14% generated by bond borrowings in 1993. The main developments observed between the two reference periods were an increase in the portion of interest-rate bearing and sight deposits and a matching decline in funds obtained through the issue of certificates of deposit. The increase in the share of interest-bearing deposits stemmed solely from fixed-rate deposits, as variable-rate deposits dropped to only 6.3% of liabilities in 1993, compared to 7.6% in 1992. This probably reflected the changing attitudes of depositors, who have adapted to the constant interest-rate decline by moving out of variable-rate instruments, as can be seen by the increase in investments at regulated rates, in particular passbook accounts, as of spring 1993. However, almost all variable-rate deposits continued to be indexed to money market rates. The amount of certificates of deposit and medium-term negotiable notes shrank, since the sizeable 3.2-point drop in the share of variable-rate resources was not fully offset by the rise in fixed-rate resources. A conjunction of factors contributed to this development. Resources collected by short-term mutual funds, which are the main subscribers to certificates of deposit, declined substantially in the 1 Savings + net capital transfers __________________________________________________________________________ Gross fixed investment + inventory variations + net purchases of land and intangible assets (see 1993 Report of the National Credit Council)

36

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

second half of 1993 and weak credit demand reduced banks' funding requirements, while customer deposits rose. At the same time, banks boosted their bond issues, almost exclusively at fixed rates. 2.2. Valuation of Indexation Spreads

INDEXATION SPREAD OF ELEVEN LARGE BANKS (including hedging transactions)

FRF billion (%) FRF billion (%) 1992 1993 1992 1993 1992 1993 1992 1993

FRENCH FRANC FIXED- FRENCH FRANC FIXED- RATE ASSETS .................. 2,474.8 2,424.2 70,7 69,1 RATE LIABILITIES ........ 2,652,8 2,859.4 75.8 81.5 – Credits............................ 1,816.9 1,759.3 51.9 50.2 – Sight deposits............... 665.4 689.1 19.0 19.6 – Securities portfolios ....... 209.5 154.3 6.0 4.4 – Interest-bearing deposits ........................ 1.058.1 1,132.2 30.2 32.3 – Non-interest rate – Certificates of deposit sensitive assets ............... 448.4 510.6 12.8 14.6 and medium-term negotiable notes ........... 179.4 230.2 5.1 6.6 – Bond issues .................. 354.8 385.9 10.1 11.0 – Non-interest rate sensitive liabilities ....... 395.1 422.0 11.3 12.0 FRENCH-FRANC FRENCH-FRANC VARIABLE-RATE VARIABLE-RATE ASSETS.............................. 1,025.3 1,082.9 29.3 30.9 LIABILITIES ................... 847.3 647.7 24.2 18.5 – Base lending rate............ 292.3 253.0 8.4 7.2 – Base lending rate ....... 6.3 10.1 0.2 0.3 – Money market rates........ 630.1 610.1 18.0 17.4 – Money market rates ... 756.2 577.2 21.6 16.5 – Bond rates ...................... 23.4 34.4 0.7 1.0 – Bond rates.................. 40.7 43.4 1.2 1.2 – Other .............................. 79.5 78.1 2.3 2.2 – Other.......................... 3.9 17.0 0.1 0.5 – Interbank rates ............... 107.3 0.0 3.1 – Interbank rates ........... 40.2 1.1 0.0 TOTAL ASSETS TOTAL LIABILITIES IN FRF ............................... 3,500.1 3,507.1 100.0 100.0 IN FRF ............................. 3,500.1 3,507.1 100.0 100.0

There was little change in the breakdown between fixed and variable-rate assets between the two periods covered by the study. However, the banks surveyed appear to have endeavoured to increase the share of fixed-rate liabilities in their total funding. The portion of fixed-rate liabilities grew by 6 points, including 2 points for interest-bearing deposits and 1.5 points for certificates of deposit, while variable- rate liabilities posted a symmetrical decline due to the drop in the amount of outstanding resources linked to money market rates. The balance sheets of the banks included in the sample showed a high surplus of French-franc denominated, variable-rate assets over the matching liabilities. This gives rise to an indexation spread corresponding to the difference between the proportion of assets and liabilities whose remuneration may be revised over periods not exceeding one year. The interest margins of credit institutions are protected against changes in key rates if the spread is nil; they will grow in response to interest-rate increases if the spread is positive, and vice versa. At the end of 1992, the interest margin was protected by a positive five-point spread between variable-rate assets and liabilities, and was even due to rise somewhat had interest rates increased. However, this spread was based chiefly on the base lending rate, with a difference of 8.2 points, whereas the spread between assets and liabilities remunerated at money market rates was a negative 3.6 points. As can be seen, the margin on such rates was, in fact, vulnerable to rate rises.

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

37

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

A year later, banks had raised the total spread to over 12 points. Nevertheless, whereas at the end of 1992 most of the difference was based on the base lending rate, which therefore was a determining factor in banks' intermediation margin, they subsequently cut back their assets indexed to this rate by 1.2 points. As a result, the spread between the base-rate linked assets and liabilities shrank to less than 7 points. This was accompanied by a reduction in the proportion of variable-rate liabilities, in particular those linked to money market rates, which declined by 5.1 points. It is difficult to determine whether these developments were prompted by banks or their customers. The decline in the portion of bank liabilities remunerated at money market rates, which had fallen to almost one point below that of similar assets, can be interpreted as a sign that financial agents expected interest rates to rise. It is also possible that the reduction in the proportion of banks' variable-rate liabilities was not entirely voluntary. It may have been caused partly by a shift in customer investment patterns towards the more attractive fixed-rate products after the interest decline in 1993. 2.3. The Impact of Benchmark Rates on Bank Balance Sheets

IMPACT OF BENCHMARK RATES ON BALANCE SHEETS

(as a % of assets or liabilities) Benchmark rates Variable-rate assets/liabilities 1992 1993 TMM (a) TBB (b) TMO (c) Other 1992 1993 1992 1993 1992 1993 1992 1993 CREDITS ...................................................... 28.0 26.4 17.0 16.4 8.4 7.2 0.4 0.5 2.3 2.2 – Household................................................. 4.1 4.8 1.6 2.5 0.6 0.5 0.2 0.3 1.8 1.4 – Business.................................................... 23.9 21.6 15.4 13.9 7.8 6.7 0.2 0.2 0.5 0.8 SECURITIES PORTFOLIOS ....................... 1.3 1.4 1.0 1.0 – – 0.3 0.4 – – INTERBANK REFINANCING............................................ 3.1 TOTAL ......................................................... 29.3 30.9 18.0 17.4 8.4 7.2 0.7 0.9 2.3 2.2 INTEREST-BEARING DEPOSITS ............. 7.6 6.3 7.2 5.4 0.2 0.2 0.1 0.1 0.1 0.5 CERTIFICATES OF DEPOSIT AND MEDIUM-TERM NEG. NOTES ................. 12.5 9.2 12.5 9.2 – – – – – – BOND ISSUES ............................................. 3.0 3.0 1.9 1.9 – – 1.1 1.1 – – INTERBANK REFINANCING............................................ 1.1 TOTAL ......................................................... 24.2 18.5 21.6 16.5 0.2 0.2 1.2 1.2 0.1 0.5 (a) Average monthly money market rate (b) Base lending rate (c) Average bond market yield

The balance sheet structure of the banks included in the sample appeared characterized by a significant proportion of variable-rate assets, accounting for some 30% of the total. Three-fourths of these assets were based on money market rates, and the remainder on the base lending rate. There was a less significant share of variable-rate liabilities, representing less than one-fifth at the end of 1993, which were almost entirely indexed to money market rates. A one-point rise in money market rates in 1992 would have increased the cost of funding by 0.22 point and the average yield on lending by 0.18 point, thereby eroding banking margins if the base lending rate had not been adjusted accordingly.

38

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

By the end of 1993 credit institutions had changed their money market asset/liability mix. Liabilities were kept lower than assets, so that in principle increases in money market rates would not negatively impact margins. On the contrary, they would even have had a slightly positive effect. Although the quantitative impact of the base lending rate seems to have weakened since it concerns a smaller portion of assets, it still appears to play a key role in margin developments. The impact of bond rate variations as a reference rate for variable-rate transactions, however, is minimal, both in terms of securities portfolios and bond issues.

The data available for the period under review appears to indicate that banks felt the combined effects of a decline in fixed-rate assets, reduced recourse to the base lending rate as a credit index and a diminution in variable-rate liabilities. This resulted in a better match between assets and liabilities linked to money market rates, which are more sensitive to changes in key rates.

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

39

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

BREAKDOWN OF BALANCE SHEET TRANSACTIONS BY INTEREST RATE CATEGORY (including hedging transactions) FRF billion (%) At end -1992 At end-1993 1992 ASSETS (FRF) ..................................................................... 3,500.1 3,507.1 100.0 Credits................................................................................. 2,796.2 2,685.8 79.9 Fixed rates .......................................................................... 1,816.9 1,759.3 51.9 Variable rates ..................................................................... 979.3 926.5 28.0 – Base lending rate .............................................................. 292.3 253.0 8.4 – Money market rates .......................................................... 593.9 576.2 17.0 – Bond rates ........................................................................ 13.6 19.2 0.4 – Other ................................................................................ 79.5 78.1 2.3 Securities portfolio.............................................................. 255.5 203.4 7.3 Fixed rates .......................................................................... 209.5 154.3 6.0 Variable rates ..................................................................... 46.0 49.1 1.3 – Money market rates .......................................................... 36.2 33.9 1.0 – Bond rates ........................................................................ 9.8 15.2 0.3 Interbank transactions ........................................................ 107.3 Non-interest rate sensitive assets........................................ 448.4 510.6 12.8 LIABILITIES (FRF) ............................................................. 3,500.2 3,507.1 100.0 Sight deposits ...................................................................... 665.4 689.1 19.0 Interest-bearing deposits .................................................... 1,322.9 1,352.5 37.8 Fixed rates .......................................................................... 1,058.1 1,132.2 30.2 Variable rates ..................................................................... 264.8 220.3 7.6 – Bank lending rate.............................................................. 6.3 10.1 0.2 – Money market rates .......................................................... 250.5 189.1 7.2 – Bond market rates............................................................. 4.1 4.1 0.1 – Other ................................................................................ 3.9 17.0 0.1 Certificates of deposit and medium-term negotiable notes............................................ 617.6 553.2 17.6 Fixed rates .......................................................................... 179.4 230.2 5.1 Variable rates ..................................................................... 438.2 323.0 12.5 Bond issues ......................................................................... 459.0 490.3 13.1 Fixed rates .......................................................................... 354.8 385.9 10.1 Variable rates ..................................................................... 104.1 104.4 3.0 – Money market rates .......................................................... 67.5 65.1 1.9 – Bond rates ........................................................................ 36.6 39.3 1.0 Interbank transactions ........................................................ 40.2 1.1 Non-interest rate sensitive liabilities ................................... 395.1 422.0 11.3

1993 100.0 76.6 50.2 26.4 7.2 16.4 0.5 2.2 5.8 4.4 1.4 1.0 0.4 3.1 14.6 100.0 19.6 38.6 32.3 6.3 0.3 5.4 0.1 0.5 15.8 6.6 9.2 14.0 11.0 3.0 1.9 1.1 12.0

Methodology The sensitivity of bank balance sheets to key rate changes and trends contributes to the understanding of how monetary policy affects the real economy. The purpose of this study is to analyze the impact of variations in benchmark rates on the costs of the assets and liabilities of the participating credit institutions, and to assess their sensitivity to macro- financial changes. The study does not attempt to assess the interest-rate risk exposure of these balance sheets, nor does it measure the impact of interest rate variations on banks' profitability. Moreover, since the study does not cover all balance sheet items, the results cannot be applied to all banking activities. A specific methodology was defined to harmonize the rules adopted by credit institutions to restructure their balance sheets by interest rate category in order to obtain a common functional

40

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

framework for analysis. This implied determining the overall range of transactions covered, establishing a common balance sheet structure by asset and liability heading, agreeing on a means of matching selected balance sheet items, defining the extent to which off-balance sheet transactions should be taken into account, grouping benchmark rates by main categories and specifying the valuation method for assets and liabilities.

1. Defining the Overall Scope Because of their immediate availability, the study was limited to franc-denominated balance sheet operations carried out in mainland France for residents and non-residents. Non-customer-related treasury operations were excluded because the purpose of the study was to measure the intermediation activity of credit institutions. In practice, some interbank transactions were nevertheless taken into account corresponding to the excess of claims over liabilities. Finally, foreign currency transactions were not taken into account because they are more influenced by euro-currency rates than domestic market rates. Data was collected on a non-consolidated basis for reasons of ease and availability. However, consolidated information proved to be indispensable for banking groups whose customer transactions are handled mainly by specialized subsidiaries.

2. By Asset and Liability Heading Because they also affect balance sheet sensitivity to interest rate movements, structural assets (tangible and intangible assets, non-interest bearing claims, adjustment accounts, and reserve requirements) were taken into account in addition to customer lending (credits and securities portfolio) corresponding to assets directly related to financing economic activity. Customer loans were recorded in terms of type of borrower to reflect the different customer categories. Because of the difficulty in isolating customer loans, trading account securities were excluded from the securities portfolios in order to focus only on securities held for sale, investment securities, and equity securities of all issuers. Similarly, on the liabilities side, long-term funds and capital resources such as bonds, capital, reserves and provisions were taken into account as well as customer funds such as deposits and certificates of deposit. Deposits were divided into sight deposits and interest-bearing deposits (which include term deposits, special savings accounts, and short-term notes (bons de caisse). In addition, an aggregation was carried out of all CDs on the one hand, and all bonds issued, including subordinated debt, on the other, for all holders, whether customers or not, because here again the statistical tool cannot identify the economic nature of the counterparties.

3. Matching Balance Sheet Operations By agreement, the balance of interbank transactions (overnight or forward) matching loans with funding was identified under assets or liabilities. Refinancing operations with the Central Bank were considered to be recorded under this heading.

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

41

STUDIES Sensitivity of Bank Balance Sheets to Benchmark Rate Changes

4. The Treatment of Off-Balance Sheet Items Hedging instruments (such as "interest rate swaps, "caps," interest rate options, etc.) constitute an increasingly important area of activity for credit institutions, since they can be used to neutralize the interest rate exposure arising on many transactions. As such, they influence the interest rate sensitivity of credit institutions' balance sheets. In order to take this factor into account and to determine the interest rate for a particular balance sheet heading, but also to keep matters as simple as possible, it was agreed to consider only those off- balance sheet items that serve as hedges, whether micro or macro, for assets and liabilities.

5. Grouping of Benchmark Rates Based on the above, the breakdown of balance sheet items and of part of off-balance sheet commitments by interest rate category was carried out on the basis of contractual interest rates. Outstanding assets and liabilities were grouped according to the terms of remuneration, in order to distinguish between fixed and variable interest rates. The variable interest rate indexes were divided into four categories: the base lending rate, money market rates (the TMP, or "taux moyen pondéré", for the daily average overnight money market rate; the T4M, or "taux moyen mensuel du marché monétaire", for the monthly average money market rate; the TAM, or "taux annuel monétaire", for annual money market rate, and PIBOR, for the Paris Interbank Offered Rate), bond market rates (TME, or "taux moyen mensuel de rendement des emprunts d'État à long terme", for the average monthly yield on long-term French government bonds; TMO, or "taux du marché obligataire" for the bond market rate, etc.), and other rates, such as composite rates. It was agreed that deposits and loans earning or paying a variable interest rate with an interest-fixing period exceeding one year would be classified, at the time the information was gathered, under fixed- rate instruments. Moreover, within fixed-rate loans or deposits, those with maturities of three months or less were segregated further. At the time that the balance sheet information of the nine banks was aggregated, such redemptions were reclassified within variable-rate loans or deposits, proportionally to the various categories of variable rates, on the basis of the information provided by the reporting credit institutions.

6. Loan and Deposit Valuation Method The study was based on the average fourth-quarter data of the two years under review.

42

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

SPEECHES AND STATEMENTS

THE FUTURE OF EUROPE AS SEEN FROM A CENTRAL BANKER'S PERSPECTIVE Speech at the 1995 Natwest Lecture in London by the Governor of the Banque de France, Jean-Claude Trichet, 16 May 1995

My Lord, Mr Governor, Mr Ambassador, Ladies and Gentlemen, It is always a great pleasure for me to be here in London. First, I would like to thank Lord Alexander for giving me this opportunity to address such a distinguished audience. I would also like to stress how pleased I am to be here in the company of my friend and colleague, Eddie George — although I suspect that he might find the following remarks rather familiar. Indeed, as agreed with Lord Alexander, I will try and give my assessment of the future of Europe, as seen from a central banker's perspective. As you know, the Member States of the Community are now engaged in an historical process of monetary integration. The Treaty on European Union has been ratified democratically by 15 countries. It sets out a clear objective: the instauration of a monetary union, a single monetary policy and a single currency. One is, of course, fully justified in holding depassionate debates on such a crucial issue, as Eddie George recently suggested. I would here like to submit to you my own assessment, and I will elaborate on the following two ideas: – First, the objective of monetary union serves the fundamental needs and interests of the European economy: indeed, the future of Europe hinges in part on the success of the single market and, for several reasons, which I will address later, there is no such thing as a real single market without an appropriate monetary organisation and, ultimately, a monetary union. – Second, the objective of monetary union is being pursued in a progressive, realistic and pragmatic way, within the framework of a process which involves the active participation of central banks.

BANQUE DE FRANCE BULLETIN DIGEST – No. 19 – JULY 1995

43