Profits and balance sheet developments at U.S. commercial banks in 1997.
U.S. commercial banks had another excellent year m 1997. Their return on equity remained in the elevated range that it has occupied for five consecutive years, and their return on assets reached a new high (chart 1). Banks maintained their profitability while also adding significantly to assets. The year's strong economic growth increased demand for credit, and banks more than met that demand, gaming market share. In addition, banks departed from the pattern of recent years by sharply increasing their holdings of securities. Compared with 1996, banks earned a slightly lower, average rate on their interest-earning assets and paid a bit more on their liabilities, but these developments were more than offset by higher fee income and increased efficiency. Loan losses remained low relative to loans.(1)
[Chart 1 ILLUSTRATION OMITTED]
The advance in bank profits helped boost bank holding company stock prices substantially last year. With banks retaining a slightly larger fraction of income, dividend growth slowed relative to the large increases of recent years. The resulting increase in retained income helped boost bank capital, which grew about in line with assets. As has been true for several years, virtually all bank assets were at institutions classified as "well-capitalized" at the end of 1997. Only one bank--a small one--failed last year.
Consolidation continued in 1997. In June, most of the remaining legal restrictions on interstate mergers were removed, and several bank holding companies combined subsidiary banks that had been operating in separate, regions. Partly as a result, the number of banks declined to 9,217, down from 9,575 at the end of 1996 and far below the peak, reached in 1984, of about 14,500 (chart 2). At year-end 1997, the largest 100 banks accounted for two-thirds of bank assets, up from about half in 1991.
[Chart 2 ILLUSTRATION OMITTED]
BALANCE SHEETS DEVELOPMENTS
Bank assets grew 9 1/4 percent in 1997, the fastest growth in more than a decade (table 1). Demand for credit was strong, and banks were generally willing lenders. As a result, loans increased 8 1/4 percent, a bit faster than in 1996.(2) In addition, securities, which had been about unchanged for the past few years, expanded nearly 9 percent. Non-interest-earning assets, which make up about 13 percent of total assets, expanded 11 1/4 percent, in large part because of growth in the gross positive fair value of derivatives. Core deposit growth picked up, but not enough to keep pace with assets; managed liabilities and equity made up the difference.
1. Annual rates of growth of balance sheet items, 1998-97
Percent Item 1988 1989 1990 Assets 4.33 5.35 2.64 Interests-earning assets 4.04 5.61 2.23 Loans and leases (net 5.93 6.24 2.37 Commercial and industrial 1.84 2.97 -.67 Real estate 12.43 12.69 8.79 Booked in domestic offices 11.99 13.02 8.55 One- to four-family residential 14.60 16.13 14.00 Other 9.84 10.34 3.62 Booked in foreign offices 27.03 2.99 16.64 Consumer 7.64 6.18 .38 Other loans and leases -3.09 -.94 -5.68 Loan loss reserves and unearned income -4.20 10.29 .35 Securities 3.27 5.08 8.46 Investment account 2.93 4.04 8.19 U.S. Treasury -5.80 -13.79 3.50 U.S. government agency and corporation obligations 22.54 33.41 24.02 Other -2.46 -5.35 -6.70 Trading account 8.58 20.62 11.87 Other -5.82 2.49 -11.70 Non-interest-earning assets 6.45 3.50 5.51 Liabilities 4.05 5.43 2.37 Core deposits 5.48 5.75 7.58 Transaction deposits 2.65 .93 2.43 Savings and small time deposits 7.29 8.71 10.51 Managed liabilities(1) 2.27 5.13 -6.15 Deposits booked in foreign offices -7.77 -1.07 -5.88 Large time 9.22 5.00 -5.68 Subordinated notes and debentures -4.25 16.98 20.99 Other managed liabilities 5.45 9.86 -8.06 Other -.06 3.29 4.43 Equity capital 8.76 4.18 6.64 MEMO Commercial real estate loans(2) n.a. n.a. n.a. Mortgage backed securities 19.06 41.00 34.39 Item 1991 1992 1993 Assets 1.33 2.19 5.68 Interests-earning assets 1.98 2.53 6.56 Loans and leases (net -2.65 -1.04 6.05 Commercial and industrial -9.10 -4.10 .52 Real estate 2.73 1.94 6.13 Booked in domestic offices 2.90 2.57 6.17 One- to four-family residential 7.76 7.53 11.08 Other -1.93 -2.86 .22 Booked in foreign offices -2.35 -17.80 4.67 Consumer -2.55 -1.66 9.06 Other loans and leases -4.91 -4.24 9.97 Loan loss reserves and -3.78 -4.85 -5.82 unearned income 16.23 12.29 12.26 Securities 14.42 11.44 8.11 Investment account 32.01 23.95 7.24 U.S. Treasury U.S. government agency and 15.88 12.77 9.62 corporation obligations -2.56 -5.20 6.09 Other 38.88 21.01 51.84 Trading account 2.82 1.57 -7.90 Other -3.10 -.32 -.86 Non-interest-earning assets 1.01 1.35 5.12 Liabilities 5.25 5.09 1.49 Core deposits 3.38 14.62 5.47 Transaction deposits 6.24 .18 -.85 Savings and small time deposits -6.19 -6.07 12.30 Managed liabilities(1) Deposits booked in foreign 3.81 -5.85 15.06 offices -19.73 -26.20 -9.21 Large time Subordinated notes and 4.69 34.20 10.82 debentures -1.39 6.94 22.18 Other managed liabilities -4.18 -1.02 15.30 Other Equity capital 5.98 13.75 12.58 MEMO Commercial real estate loans(2) -2.54 -2.54 -.60 Mortgage backed securities 19.27 19.27 9.66 Item 1994 1995 1996 Assets 8.06 7.61 6.03 Interests-earning assets 5.77 7.76 5.60 Loans and leases (net 9.83 10.63 8.02 Commercial and industrial 9.33 12.26 7.24 Real estate 7.90 8.33 5.44 Booked in domestic offices 7.64 8.48 5.50 One- to four-family residential 10.09 10.06 4.65 Other 4.35 6.25 6.75 Booked in foreign offices 18.35 2.81 3.18 Consumer 16.01 9.98 4.44 Other loans and leases 5.29 14.23 22.28 Loan loss reserves and unearned income -2.22 .47 .06 Securities -2.61 .59 .82 Investment account -8.46 -1.55 -1.14 U.S. Treasury -8.46 -19.21 -14.30 U.S. government agency and corporation obligations .87 6.43 3.62 Other 2.49 4.33 1.71 Trading account -20.46 18.51 14.44 Other 3.25 7.64 -.90 Non-interest-earning assets 25.65 6.63 8.89 Liabilities 8.31 7.23 5.89 Core deposits -.17 3.97 4.12 Transaction deposits .33 -3.09 -3.45 Savings and small time deposits .08 8.37 8.34 Managed liabilities(1) 17.57 10.61 9.48 Deposits booked in foreign offices 30.89 5.13 4.27 Large time 8.72 19.61 21.16 Subordinated notes and debentures 9.23 6.61 17.74 Other managed liabilities 12.91 11.63 7.83 Other 79.17 20.50 2.57 Equity capital 5.24 12.07 7.66 MEMO Commercial real estate loans(2) 4.00 6.35 7.66 Mortgage backed securities -3.12 .67 2.03 MEMO: Dec. 1997 Item 1997 level (billions of dollars Assets 9.22 4,971 Interests-earning assets 8.88 4,281 Loans and leases (net 8.38 2,885 Commercial and industrial 12.15 791 Real estate 9.20 1,235 Booked in domestic offices 9.42 1,207 One- to four-family residential 9.69 713 Other 9.04 494 Booked in foreign offices .34 28 Consumer -2.17 544 Other loans and leases 13.78 372 Loan loss reserves and unearned income -.69 58 Securities 8.85 1,006 Investment account 8.66 861 U.S. Treasury 8.88 151 U.S. government agency and corporation obligations 14.19 500 Other 11.20 210 Trading account 9.97 145 Other 12.81 390 Non-interest-earning assets 11.35 690 Liabilities 9.12 4,557 Core deposits 4.52 2,494 Transaction deposits -4.58 757 Savings and small time deposits 9.05 1,737 Managed liabilities(1) 13.83 1,720 Deposits booked in foreign offices 11.13 526 Large time 20.13 379 Subordinated notes and debentures 21.00 62 Other managed liabilities 12.22 753 Other 23.77 344 Equity capital 10.34 414 MEMO Commercial real estate loans(2) 9.85 496 Mortgage backed securities 14.18 380
NOTE. Data are from year-end to year-end.
n.a. Not available.
(1.) Measured as the sum of deposits in foreign offices, large time deposits in domestic offices, federal funds purchased and securities sold under agreements to resell, demand notes issued to the U.S. Treasury, subordinated notes and debentures, and other borrowed money.
(2.) Measured as the sum of construction and land development loans secured by real estate; real estate loans secured by nonfarm nonresidential properties; real estate loans secured by multifamily residential properties; and loans to finance commercial real estate, construction, and land development activities not secured by real estate.
Loans to Business
The value of commercial and industrial (C&I) loans on bank balance sheets expanded nearly 12 1/4 percent, the second largest annual increase in seventeen years. C&I loans increased in part because inventory accumulation and fixed investment by businesses apparently outstripped their internally generated funds last year; at nonfinancial corporations, the excess of capital expenditures over internal funds rose to $45 billion, up from $6 1/2 billion, in 1996 (chart 3). The borrowing needs of nonfinancial corporations were further increased because, as has been true for several years, they retired a large volume of equity, on net, via stock buybacks and during corporate acquisitions.
[Chart 3 ILLUSTRATION OMITTED]
Banks expanded their share of outstanding nonmortgage business credit to its highest level since 1989 (chart 4). In part, this expansion reflected a substantial rise in the number of mergers and acquisitions among middle-market firms, which are more likely to be financed by bank loans than are the combinations of large corporations. Those respondents that reported stronger demand for business loans on the Federal Reserve's quarterly Senior Loan Officer Opinion Survey on Bank Lending Practices (BLPS) last year most commonly attributed the increased demand to mergers and acquisitions; m addition, banks cited financing for inventories and for plant and equipment.
[Chart 4 ILLUSTRATION OMITTED]
Banks also expanded their market share by competing more vigorously for business loans. Although only small fractions of the respondents to the BLPS said they had eased standards on business loans m 1997, large fractions indicated they had eased loan terms, particularly the spreads of loan rates over their bank's cost of funds (chart 5). These results are somewhat at odds with the Federal Reserve's quarterly Survey of Terms of Business Lending (STBL), which showed a slight widening of the average spread on business loans last year (chart 6). Nevertheless, the average of measured spreads reported in the STBL narrowed over the preceding several years, so results from both surveys are broadly indicative of aggressive pricing of business loans.
[Chart 5 and 6 ILLUSTRATION OMITTED]
Spreads on the largest loans are the narrowest relative to historical norms. Partly as a result, several large banks established programs last year to package and sell "collateralized loan obligations" (CLOs)--securities backed by large commercial and industrial loans. Respondents to the November 1997 BLPS attributed the recent interest in CLOs to a desire by banks to deploy their capital more efficiently by moving relatively high quality loans (which have the same regulatory capital requirement as riskier loans) off their balance sheets.(3) With these loans off the books, the measured growth in business loans last year understates the expansion of bank-originated credit. However, the understatement was small because the bulk of the CLO activity involved loans by foreign rather than U.S. banks. Nonetheless, the survey results suggest that CLOs have the potential to shift significant amounts of C&I loans off the books of domestic banks over time.
Banks' holdings of commercial real estate loans increased more than 9 3/4 percent last year. The growth of these loans has been picking up for the past four years following a sharp pullback in the early 1990s. A variety of indicators show continued improvement in the condition of the commercial real estate industry, including falling vacancy rates and rising prices for properties. Commercial real estate loans have grown much more rapidly in recent years at smaller banks; after adjusting for the effect of mergers, these loans at the largest 100 banks increased 4 1/4 percent last year, whereas growth at the remaining banks was 15 1/2 percent. The losses on such loans in the early 1990s were concentrated at large banks, which may therefore remain relatively more cautious. Growth at large banks may also have been held down somewhat by the issuance of commercial mortgage-backed securities (CMBSs); many of the respondents to the August 1997 BLPS--particularly the largest banks--reported that they had issued CMBSs. These securities were virtually nonexistent ten years ago. In 1997, however, the increase in the outstanding dollar amount of CMBSs exceeded the increase in the commercial mortgage loans on the books of commercial banks; by this measure, CMBSs and bank loans were the two leading sources of finance for commercial real estate activities.
Loans to households
In contrast to business loans, consumer loans on banks' books contracted 2 1/4 percent last year. Several factors contributed to the decline, including reduced demand for such loans by households, which in turn partly reflected a substitution toward home equity loans; a tightening of terms and standards on consumer loans by some banks; and the increased securitization of consumer loans. Consumer credit from all sources grew 4 1/2 percent last year, down from 8 percent in 1996 and 14 percent in the two preceding years.
As is typical over an economic expansion, the deceleration has been more pronounced for consumer credit than for spending on consumer durables; the growth of the latter slowed from 9 1/4 percent in 1994 to 4 percent last year. In the early stages of an expansion, net increases in consumer debt tend to be large because an upturn in spending for consumer durables boosts loan originations, while past low levels of originations keep debt repayments low. But as the economy continues to expand, the growth of repayments provides more of an offset to new originations, resulting in smaller net additions to the stock of debt.
As discussed below, the average repayment performance of consumer loans deteriorated significantly in 1995-96 and remained poor last year. In response, for the past two years a large percentage of banks tightened their standards on consumer loans, according to responses to the BLPS (chart 7). But the percentages reporting tightening were lower in the second half of 1997, suggesting that many banks felt that they had altered their standards sufficiently. Some banks also reported imposing lower credit limits on credit cards and rinsing finance charges on outstanding balances. While these adjustments may have made credit card lines harder to acquire for less creditworthy customers, banks apparently remained eager to attract more creditworthy borrowers. Reportedly, credit card solicitations continued at a record pace, and the value of credit card lines grew 11 3/4 percent. By the end of 1997 the aggregate credit card Utilization rate--credit drawn on credit card lines relative to the total size of such lines--had fallen to less than one-third.
[Chart 7 ILLUSTRATION OMITTED]
Despite the decline in consumer loans on the books of banks, outstanding consumer loans originated by banks grew nearly 4 percent last year. The decline in loans on the books resulted from an increase of more than 20 percent in the volume of loans originated by banks and then securitized; at the end of the year, these off-balance-sheet amounts accounted for nearly 30 percent of consumer loans originated by banks (chart 8). Given the marked deterioration in the performance of consumer loans in recent years, banks may be inclined to reduce the amount of such loans appearing on their balance sheets. In addition, banks evidently find securitization frequently to be a less expensive way of funding consumer loans than funding them on their balance sheets.(4)
[Chart 8 ILLUSTRATION OMITTED]
Consumer loans were likely also depressed last year because many households refinanced them with loans backed by real estate, which typically have lower interest rates and for which interest payments are generally tax deductable. Borrowing under home equity lines of credit at banks increased 15 percent in 1997, and closed-end residential real estate loans secured by junior liens increased 10 3/4 percent. Real-estate-secured borrowing from nonbanks, particularly finance companies, was also strong. When asked in !he February 1997 BLPS to account for the strength in home equity loans, most banks cited increased demand from households or specific encouragement by the banks to consolidate unsecured consumer credit with such loans.
Home mortgages secured by first liens also accelerated in 1997, expanding 8 3/4 percent. Some of the strength in mortgages reflected the high level of residential construction activity last year, although increased construction activity generally rinses the level of mortgages only gradually. Toward the end of the year, the low level of mortgage interest rates induced large numbers of borrowers to refinance existing mortgages (chart 9). The resulting increase in refinancing activity likely contributed to the expansion of banks' holdings of mortgages because some households increase their mortgage size and take cash out when refinancing. The added cash may be used to pay down other debts, so the heightened level of refinancing probably contributed a bit to the weakness in consumer loans discussed above. The high level of refinancing activity may also have boosted mortgage loans on banks' books temporarily, as previously securitized loans were replaced by new loans that appear on the balance sheets of the refinancing institution, at least for a while.
[Chart 9 ILLUSTRATION OMITTED]
As has been true for several years, real estate loans at banks were also boosted somewhat by banks' acquisition of savings institutions; last year such acquisitions added about 2 percentage points to the growth of real estate loans at banks. Banks have been absorbing savings institutions since the savings and loan crisis in the late 1980s. Partly as a result, banks have become bigger players in the mortgage business, which had previously been dominated by thrift institutions. At the end of the fourth quarter, single family mortgages accounted for nearly as large a share of bank assets (14 1/2 percent) as did C&I loans (15 3/4 percent).
Banks' holdings of securities increased more than 8 3/4 percent last year after declining in 1994 and changing little in 1995 and 1996. Securities also expanded sharply earlier in the 1990s, but that increase occurred during a period of weak loan demand and strong inflows of core deposits; it may also have reflected banks' efforts to comply with new regulatory standards that imposed larger capital requirements on loans than on securities.(5) The strength in securities last year, by contrast, coincided with substantial loan growth and thus does not seem to indicate any diminution in the demand for loans or in the willingness of banks to provide them. Indeed, responses to the May 1998 BLPS indicated that the growth in securities was due in part to an increased willingness on the part of some banks to boost leverage in an effort to raise return on equity. Many responses also attributed the growth to mergers: Some banks were expanding their balance sheets m line with capital accumulated because their holding companies had recently participated in pooling-of-interest mergers and therefore were constrained from buying back stock.
Securities in investment accounts at banks expanded 8 3/4 percent last year. Within investment accounts, mortgage-backed securities, which account for about half of the securities in such accounts, increased much more rapidly than the remaining types of securities. The 10 percent growth an interest-earning trading account securities was concentrated in securities booked at domestic offices; foreign office trading accounts declined somewhat. The decline in trading account securities booked in foreign offices was largely in the fourth quarter, when the turmoil in East Asia probably induced banks to sell some securities and reduced the value of some that they kept.
Non-interest-earning trading account assets were boosted by a $42 1/2 billion rise in the gross positive fair value of derivatives written on interest rates, exchange rates, and equity, commodity, and other prices (see box "Off-Balance-Sheet Activity"). Much of this gain was probably the result of an increase in the value of exchange-rate contracts in the fourth quarter, most likely because of the large depreciations of several East Asian currencies at that time. Banks typically hold offsetting positions in such contracts, so large movements in exchange rates generally increase bank assets and liabilities without greatly affecting net worth. Exchange-rate-based derivative contracts in a negative position, which are recorded as liabilities on the balance sheet, also increased substantially in the fourth quarter.
Bank core deposits grew 4 1/2 percent last year, about the same rate as in 1996 and only half as fast as the rise in bank assets. Core deposit growth was relatively slow, in part because banks set deposit rates low in comparison to market rates, as they have done for several years. For example, the average rate paid by banks on their interest-bearing core deposits was 1 1/4 percentage points below the yield on six-month Treasury bills last year. By contrast, the average difference was only 1/4 percentage point from 1987 to 1993 (chart 10). Yields available on core deposits were especially low relative to the returns on bond and stock mutual funds last year, and households substitution toward such funds likely continued to depress the growth of core deposits.
[Chart 10 ILLUSTRATION OMITTED]
Within core deposits, savings accounts expanded rapidly, mainly because of the ongoing introduction of "sweep" programs. These programs automatically move funds out of transactions deposits, against which banks must hold non-interest-bearing reserves, into savings accounts, against which banks do not have to hold reserves. Sweep programs thus release funds that banks can invest in interest-earning assets. In 1997, banks slowed the initiation of programs that sweep funds out of NOW accounts--until last year the most common form of retail sweep program--but they accelerated the creation of programs that sweep funds from household demand deposits; on net, the amount moved by the initiation of new sweep programs--$84 billion--was about 25 percent lower than in 1996.
With core deposits growing more slowly than assets, banks funded their robust growth last year largely with managed liabilities, which increased 13 3/4 percent. Large time deposits, deposits in foreign offices, and subordinated debt all expanded at double-digit rates last year. During the past five years of substantial growth of bank assets, the share funded by managed liabilities rose from 28 1/4 percent to nearly 34 percent, a level just below that in the late 1980s.
Historically, banks have increased their reliance on managed liabilities during periods of rapid asset growth (chart 11), perhaps because they cannot profitably attract new core deposits quickly enough to keep up with rapidly growing assets. Managed liabilities can generally be raised in large amounts with little or no change in the rates paid for the funds. The public's demand for core deposits, however, is much less sensitive to rates, so banks would have to increase deposit rates substantially to induce large inflows over relatively short periods of time. Thus, even though core deposits on average are less expensive than managed liabilities, the latter may still be the more profitable means for banks to finance rapid growth in assets, with reliance on those liabilities declining when asset growth is weak.
[Chart 11 ILLUSTRATION OMITTED]
Bank borrowing from the Federal Home Loan Bank System (FHLB) grew significantly last year, rising by more than a half and reaching about 1 percent of bank assets by year-end. Membership in the FHLB was limited to thrift institutions until 1989, when the Financial Institutions Reform, Recovery, and Enforcement Act allowed qualifying commercial banks to join; more than half of all commercial banks had become members by the end of 1997.
Bank equity grew 10 1/4 percent last year, a bit faster than assets. More than one-third of the growth m capital arose from a 30 percent increase in retained income. Retained income increased so much partly because net income was strong, but also because the proportion of income retained by banks rose, from 24 1/4 percent in 1996 to 27 3/4 percent last year. With the increased rate of retention, dividends rose just 8 percent, well below the 29 percent annual rate posted between 1993 and 1996. About one-fourth of the increase in capital was new capital, acquired generally from the issuance of stock or the rejection of funds from parent holding companies. Most of the remaining growth in capital arose from two sources: the increase in goodwill arising from bank mergers and the increase in net unrealized gains on investment account securities available for sale.(6)
Capital for regulatory purposes, which excludes both goodwill and net unrealized gains on investment account securities, increased only 7 1/4 percent, a bit less than assets; hence, the average leverage ratio edged down over the year (chart 12). Industry-average risk-weighted capital ratios (total and tier 1) also declined slightly over the year.(7) Even though securities, which generally have low risk than loans, which generally carry high risk weights, risk-weighted assets increased more rapidly than total assets because of rapid growth in the selected off-balance-sheet items that are included in risk-weighted assets on a credit-equivalent basis. The risk-weighted credit-equivalent amount of these items increased 30 percent from year-end 1996 to year-end 1997, raising their share of risk-weighted assets to nearly 20 percent. Despite the slight declines, average capital ratios remain high relative to regulatory stan well-capitalized banks.(8)
[Chart 12 ILLUSTRATION OMITTED]
TRENDS IN PROFITABILITY
The 1997 rise of 12 3/4 percent in the net income of U.S. commercial banks boosted the industry's return on assets to 1.25 percent, a new record, and its return on equity to more than 14 3/4 percent (table 2). With profits strong, bank holding company stock prices advanced rapidly over the first three quarters of the year (chart 13). In the fourth quarter, however, bank holding company stocks, especially those of money center banks, were buffeted by concerns that economic problems in Asia would depress earnings. Nonetheless, for the year as a whole, stock prices of the money center companies about matched the broader market, while those of regional banking companies easily surpassed both.
2. Selected income expense items as a proportion of assets, 1991-97
Percent 1991 1992 1993 1994 Net interest income 3.62 3.89 3.90 3.78 Noninterest income 1.81 1.95 2.13 2.00 Noninterest expense 3.75 3.86 3.94 3.75 Loss provisioning 1.03 .78 .47 .28 Realized gains on investment .09 .11 .09 -.01 account securities Income before and .73 1.32 1.70 1.73 extraordinary items Taxes and extraordinary items .22 .41 .50 .58 Net income (return on assets) .51 .91 1.20 1.15 Dividends .45 .41 .62 .73 Retained income .07 .49 .58 .42 MEMO Return on equity 7.71 12.64 15.32 14.63 Percent 1995 1996 1997 Net interest income 3.72 3.73 3.67 Noninterest income 2.02 2.18 2.23 Noninterest expense 3.64 3.71 3.61 Loss provisioning .31 .37 .40 Realized gains on investment .01 .03 .04 account securities Income before and 1.81 1.85 1.93 extraordinary items .63 .65 .68 Taxes and extraordinary items 1.18 1.20 1.25 Net income (return on assets) Dividends .75 .91 .90 Retained income .43 .30 .35 MEMO Return on equity 14.69 14.53 14.87
[Chart 13 ILLUSTRATION OMITTED]
Rates of commercial bank profitability averaged over the past five years are higher than in the previous five-year period and significantly exceed longer-term averages. For example, the industry's 14 3/4 percent average return on equity over the past five years was about 5 1/2 percentage points higher than the average over the previous five years and 4 percentage points higher than the average for the forty years from 1948 to 1987.(9) The improvement in the 1993-97 returns over the 1988-92 returns is primarily the result of a much-reduced level of loss provisioning relative to loans. The decline in provisioning in turn resulted from the vastly improved quality of assets: Troubled sovereign and commercial real estate credits extended in the 1970s and 1980s were worked out, and the sustained economic expansion contributed to a low level of losses on more recent lending. The high level of profits also reflects banks' efforts to limit costs, which have helped lower the share of revenue needed to cover noninterest expenses. Over a longer period, noninterest income has accounted for an increasing share of bank revenue as banks have shifted away from traditional intermediation and toward such fee-based activities as servicing loans funded by others and selling and servicing mutual funds and annuities.
Interest Income and Expense
Net interest income as a percentage of average assets declined 6 basis points last year because of a decline in banks' net interest margin net interest income as a percentage of interest-earning assets, chart 14). The narrowing of the net interest margin was produced by a slight decline in the average rate received on interest-earning assets and an increase in the average rate paid on interest-bearing liabilities. Although shorter-term market rates on balance changed little over the year, the average rate earned on assets edged slightly lower as the distribution of bank assets shifted toward those that carry lower interest rates. On the liability side, the net interest margin has been squeezed by the need to fund rapid asset growth with managed liabilities, on which the average rate paid substantially exceeds that paid on core deposits.
[Chart 14 ILLUSTRATION OMITTED]
The net interest margin has been drifting lower since 1993 but remains high relative to the levels of the late 1980s. Some reports in the financial press in the early 1990s attributed the rise in bank net interest margins at that time to the concurrent rapid decline in short-term market interest rates and to the steepening of the yield curve that accompanied that decline. Underlying this explanation is the assumption that rates on liabilities adjust more frequently than rates on assets at many banks. The validity of the assumption is hard to assess directly because of the difficulty of measuring the repricing frequency of many bank assets and liabilities. However, this assumption is not consistent with past movements in the net Interest margin and the slope of the yield curve, which do not suggest a tight link between them; nor is such a link evident between net interest margins and changes in the slope of the yield curve. For example, since the early 1990s the yield curve has flattened considerably, but the net Interest margin, while trending lower, has remained fairly wide. Similarly, during periods of very steep (or steeping) yield curves in the 1980s, the net Interest margin showed little if any response.
Rather than being a response to a very steep yield curve, the sharp widening of the net interest margin in the early 1990s likely reflected two other factors. First, margins had been compressed in the late 1980s by competition among banks for loans and funding sources as well as by the elevated rates that some troubled banks and thrift institutions were paying for funds. Second, a number of banks may not have had the capital levels they needed to meet the risk-based capital rules phased in between 1990 and 1992. With bank equity prices depressed at that time, capital was expensive to raise, and so these banks were under pressure to limit balance sheet expansion and boost profits. Their consequently less aggressive efforts to bid for deposits and make loans likely led to a widening of spreads between loan and deposit rates. During this time, competitive pressures on margins may also have eased as troubled Institutions were recapitalized or closed.
Since 1993, banks' increasingly competitive stance in loan markets has contributed to some narrowing of the net interest margin. However, the resulting squeeze on banks' margins has been mitigated by three other factors. First, margins were supported until last year by the shift of bank assets away from securities, which generally yield relatively low returns, toward loans, especially loans to households. In addition, respondents to the November 1997 BLPS indicated that the average rate earned on business loans had been boosted over the previous year by an increase in their average risk, which in turn primarily reflected an increase in loans used to finance mergers and acquisitions.
A second factor supporting the net interest margin has been the relatively low level of rates paid on retail deposits as gauged by the difference between deposit rates and market interest rates in earlier years. Although the lower level of rates has increased banks' reliance on relatively expensive managed liabilities, it has kept down the cost of core deposits, which continue to account for more than half of bank liabilities. Finally, compared with the early 1990s, banks have been funding a significantly larger fraction of assets with capital, and the returns paid on capital are not included in interest expense. More broadly, to the extent that banks must pay higher returns on equity than on borrowed money, the rise in capital ratios gives banks a strong incentive to boost net Interest margins to raise the return on assets and thereby keep the return on equity from deteriorating.
Noninterest income Increased 5 basis points as a percentage of assets last year. The types of noninterest income that expanded most were earnings from fiduciary activities and the "other fee income" component of the broad category "other noninterest income," which includes, among other things, credit card fees, mortgage servicing fees, fees from the sale and servicing of mutual funds and annuities, ATM surcharges, and fee income from securitzed loans. In particular, fee income from securitized credit card loans likely increased last year because of the high volume of securitization noted earlier. Through the first three quarters of 1997, higher trading revenue also buoyed noninterest income, but trading results were depressed in the fourth quarter by the effects of the economic problems in Asia (discussed below). On balance, trading revenues over the year were about unchanged as a share of assets.
Taking a longer perspective, a shift by banks away from traditional intermediation and toward fee-based income sources has been enlarging the share of noninterest income in bank revenue for more than a decade. Since the mid-1980s, noninterest income has Increased from about 26 percent to about 38 .percent of total bank revenue (defined as net interest income plus noninterest Income, chart 15). Since the early 1990 the bulk of the increase has come from "other fee income," which has risen from about 12 percent to more than 15 percent of revenue since 1991. The second largest contributor to the rise is the nonfee component of other noninterest income, which includes revenue from the provision of data processing services, income from unconsolidated subsidiaries, and gains from sales of assets other than securities and trading assets (including bank premises, other real estate owned by banks, and loans). Before 1991 data on these two income components were not reported separately; the share of revenue contributed by the two combined increased roughly 4 1/2 percentage points between 1985 and 1990.
[Chart 15 ILLUSTRATION OMITTED]
Banks also benefited last year from a reduction in noninterest expense relative to both assets and revenues (chart 16). The bulk of the improvement was produced by a decline in "other noninterest expense," a broad category that accounts for nearly half of noninterest expense and includes deposit Insurance premiums, losses on the sale of assets other than securities and trading assets, amortization of intangible assets, expenditures for information processing services provided by others, advertising, and merger restructuring charges. In part, last year's improvement reflected a temporary rise in expenses in 1996 owing to a large special charge for merger-related costs and a one-time assessment to recapitalize the Savings Association Insurance Fund, which was paid by banks that had acquired the deposits of thrift institutions.
[Chart 16 ILLUSTRATION OMITTED]
Labor costs and occupancy costs, the other components of noninterest expense, grew more slowly than industry revenue last year but expanded rapidly m comparison with earlier in the decade. Labor costs increased 6 3/4 percent, the largest rise since 1986. Industry employment expanded 2 percent, after several years of essentially no growth, and labor costs per employee continued to rise at about the same rate seen in recent years. Similarly, occupancy costs increased roughly 5 3/4 percent, just below the year-earlier pace but considerably faster than over the previous several years. The number of bank offices rose 2 3/4 percent last year, the largest advance since 1994 and the third largest since 1981.
Despite the pickup in these expense categories last year, the banking industry has restrained the growth In labor and occupancy costs since the mid-1980s. Since 1985, after adjusting for inflation, consolidated assets increased nearly 30 percent and revenues expanded about 60 percent. By contrast, employment declined 2 percent and the number of bank offices increased less than 20 percent. Thus, average revenue generated per employee increased more than 60 percent, while revenue per office rose more than 30 percent. Furthermore, over the same period, the inflation-adjusted occupancy cost per bank office fell 3 percent, a decline influenced perhaps by a shift of some banks toward smaller branches in supermarkets and other nontraditional locations.
By contrast, other noninterest expense increased substantially as a share of revenue in the late 1980s and early 1990s, and only a part of that rise has been reversed since 1991. The earlier rise likely resulted, at least in part, from collection costs and legal expenses generated by the high level of problem loans at that time. With these expenses presumably down considerably since then, noninterest expense has probably been supported more recently by increases in servicing and administrative costs generated by the rapid growth in consumer loans, particularly credit card loans, as well as by the costs associated with the growing volume of off-balance-sheet and fee-based activities.
Loss Providing and Loan Quality
Provisioning for loan and lease losses as a percentage of assets edged higher last year. Nonetheless, with charge-offs remaining relatively low, provisioning as a share of loans has risen only a little from its 1994 trough (chart 17). The low level of charge-offs, in turn, reflects the excellent overall performance of bank loan portfolios thus far in this expansion. This overall outcome, however, masks substantial differences between the results for loans to businesses and those for loans to households. Delinquency and charge-off rates on loans to businesses declined sharply earlier in the decade and have remained very low (chart 18, top panels), whereas those on loans to households, and especially on credit card loans, have increased substantially since late 1994 (chart 18, bottom panels). Consumer delinquency rates flattened out early last year, however, and by late in the year, charge-offs showed signs of stabilizing.
[Chart 17 ILLUSTRATION OMITTED]
The flattening of loss rates on loans to households last year was reflected in the results for credit card banks.(10) Profitability at these institutions has been much higher than for the industry as a whole for several years, as strong noninterest income and the high spread on credit card loans have more than compensated for the relatively high level of noninterest expense and loan losses. However, credit card banks' earnings deteriorated considerably between mid-1995 and early 1997 before rebounding in the second half of last year. For the year as a whole, the return on equity for credit card banks averaged nearly 18 percent, considerably below the 25 percent to 30 percent returns posted between 1988 and 1995 but only about 1 1/2 percentage points lower than in 1996.
The apparent stabilization in measures of consumer loan quality was mirrored in a flattening of the trajectory of household bankruptcy filings in the second half of last year after two years of double-digit annual increases. Two factors have likely contributed importantly to the plateauing of these measures of financial distress. First, as noted above, some banks have selectively tightened lending standards in an effort to reduce loan losses. Second, the household debt burden (interest payments and required principal payments as a percentage of disposable income) has changed little recently after increasing steadily between 1994 and 1996 (chart 19). This stability reflects the slowing of consumer loan growth and the lower interest rates paid by households, which in turn resulted from mortgage refinancing and the substitution of mortgage credit for consumer loans.
[Chart 19 ILLUSTRATION OMITTED]
In contrast, the low and declining burden of business debts likely contributed to the low delinquency and charge-off rates on loans to businesses in recent years. The business debt burden (nonfinancial corporate interest payments as a percentage of cash flow) has declined since its peak in 1990 for three reasons: the reduction in the general level of interest rates, significant declines in corporate leverage in the early 1990s, and strong growth in profits. However, the debt burden of the nonfinancial business sector leveled out recently as profit growth moderated while debt growth remained strong.
With total charge-offs about matching loss provisioning in each of the past several years, banks' reserves have been about flat, and the rapid pace of loan growth has unwound about half of the 1980s increase in the ratio of reserves to loans (chart 20). Although reserves have been declining relative to charge-offs since 1994, they remain relatively high by historical standards, as one would expect with aggregate loan losses near their cyclical lows as a percentage of loans and the economy performing exceptionally well.
[Chart 20 ILLUSTRATION OMITTED]
Effects of the Economic Difficulties in Asia
Profits at several large U.S. banks were reduced by the effects of economic problems In some of the industrializing economies in Asia. These problems emerged last summer when the Thai baht dropped sharply following a decision by the Thai authorities to no longer defend the baht's peg. Subsequently, other East Asian economies experienced downward pressure on their currencies and equity prices and upward pressure on interest rates. The turbulence spread to Taiwan and Hong Kong in the fall. In Taiwan, the authorities allowed some downward adjustment of the Taiwan dollar, whereas in Hong Kong the peg to the dollar has been maintained at the cost of somewhat elevated interest rate levels. Near the end of the year, the crisis spread to Korea, where the condition of the financial system had been strained by bankruptcies of a number of major industrial conglomerates in 1997.
In response, authorities in Thailand, Indonesia., and Korea negotiated international support packages with the International Monetary Fund and other international financial institutions, as well as bilateral assistance programs with other countries. Markets in these countries were kept turbulent into 1998, however, by concerns about the magnitude of the countries' financial problems and in some cases about the willingness or ability of their governments to undertake difficult reforms. On balance, the currencies of these countries depreciated significantly relative to the U.S. dollar in 1997, with the Indonesian rupiah dropping the most (about 58 percent), followed by the Korean won (44 percent) and the Thai baht (42 percent).
The effects of the financial crisis in Asia on the earnings of U.S. banking organizations were concentrated on a fairly small number of large institutions with relatively large exposures to the region (see box "International Operations of U.S. Banks"). For the three most troubled Asian economies (Indonesia, Korea, and Thailand), the total exposure of reporting banking organizations amounted to roughly 13 percent of capital (table 3). Most of this exposure was at six large money center organizations (which include five of the largest ten banks either directly or through a parent bank holding company), which had exposures totaling about 30 percent of capital.
3. Exposure of U.S. banking organizations to troubled Asian economics, relative to capital, year-end 1997
Percent Country All reporting Money center Super regional Indonesia 2.6 6.2 1.4 Korea 7.4 17.2 3.0 Thailand 2.7 6.8 .6 Total 12.7 30.2 5.0 Selected other economies(1) 19.9 50.1 6.7 MEMO: Country Other Total Exposure (billions of dollars) Indonesia .3 9.0 Korea 1.4 25.3 Thailand .4 9.4 Total 2.1 43.6 Selected other economies(1) 1.4 68.1
NOTE. Exposures includes the institutions' lending and derivatives exposures for cross-border as well as local-office operations. Respondents may file information on one bank or on the bank holding company as a whole. Capital is defined as equity, subordinated debt, and loan loss reserves.
(1.) Mainland China, Hong Kong, Taiwan, Malaysia, the Philippines, and Singapore.
SOURCE. Federal Financial Institutions Examination Council, Country Exposure Report.
The effect of the problems in Asia showed up primarily in the trading income of the top ten banks, which averaged $1.9 billion per quarter over the first three quarters of the year but fell to just $810 million in the fourth quarter (table 4). Trading income related to foreign exchange positions was strong in the fourth quarter, suggesting that some U.S. banks benefited either from wider spreads on Asian currency contracts or an increased volume of trades in such contracts. By contrast, gains on interest rate positions fell more than half, and substantial losses on equity, commodity, and other exposures reversed all of the gains attained on such contracts over the first three quarters of the year. Reportedly, these losses reflected those on positions not only in Asia but also in other emerging markets, including those in Latin America and Eastern Europe, that suffered from the Asian downdraft.
4. Trading revenue at the ten largest U.S. banks, by type of exposure, 1995-97
Millions of dollars Year Total Interest Foreign Equity and rate exchange other 1995 4,830 2,632 1,772 426 1996 6,213 3,621 1,973 618 1997 6,570 3,549 3,039 -18 Q1 2,052 1,221 505 326 Q2 1,609 822 698 88 Q3 2,099 1,081 813 205 Q4 810 425 1,023 -637
Despite the Asia-related troubles in the fourth quarter, however, net trading revenues for the year as a whole were nearly 6 percent higher than in 1996 because of the strong results in the first three quarters of the year. Also, the largest U.S. banks continued to report strong total earnings in the fourth quarter, thanks to extraordinarily robust domestic earnings and higher-than-usual realized gains on investment account securities.
DEVELOPMENTS IN 1998
During the first quarter of 1998, assets at the domestic offices of U.S. commercial banks expanded somewhat more rapidly than they did last year. Growth in commercial and industrial loans picked up a bit further from its already robust 1997 pace, and the surge in refinancing activity that followed the decline in interest rates late last year and early this year supported growth in real estate loans. By contrast, the value of consumer loans on banks' books declined over the quarter, as a moderate increase in bank originated loans outstanding was more than offset by increased securitization. The pace of securities acquisitions slowed a bit from its rapid pace late last year but remained quite strong.
Stock prices of the largest banking companies have, on balance, increased sharply this year, although they have remained volatile. In part, the rise likely reflected the market's belief that the economic situation in Asia might be stabilizing and the consequent fading of concerns about the effects of the Asian crisis on future earnings. In addition, the effects of both anticipated and announced mergers involving large banking organizations substantially boosted the stock prices of some of the affected companies, at least for a time. Finally, investors pushed the broader equity markets sharply higher, as incoming economic data were seen by the markets as increasing the likelihood of continued healthy growth with low inflation. Over the first four months of 1998, stocks of money center banks advanced 18 1/2 percent, and regional bank stocks rose 12 percent. By contrast, the broader market, as measured by the S&P 500, rose 14 3/4 percent.
Initial reports of first-quarter profits of bank holding companies generally showed a continuation of last year's trends, with gains in noninterest income about offsetting weaker net interest income. A few large banks reported costs relating to problems in Asia, but trading income rebounded from the poor results posted in the fourth quarter of 1997.
(1.) Except where otherwise indicated, data in this article are from the quarterly Reports of Condition and Income (Call Reports) for insured domestic commercial banks and nondeposit trust companies (hereafter, banks). The data consolidate information from foreign and domestic offices and have been adjusted to take account of mergers (see appendix). Size categories, based on assets at the start of each quarter, are as follows: the 10 largest banks, large banks (those ranked 11 through 100 by size), medium-sized banks (those ranked 101 through 1,000 by size), and small banks (those not among the largest 1,000 banks). At the start of the fourth quarter of 1997, the approximate asset size of the banks in those groups were as follows: the 10 largest banks, more than $70 billion; large banks, $6 billion to $70 billion; medium-sized banks, $300 million to $6 billion; small banks, less than $300 million. Many of the data series reported here begin in 1985 because the Call Reports were significantly revised at the start of that year. Data from before 1985 are taken from Federal Deposit Insurance Corporation, Statistics on Banking (FDIC, 1997). The data are also available on the World Wide Web site of the FDIC (http://www.fdic.gov/databank/sob/). Data shown may not match data published in earlier years because of revisions and corrections. In the tables, components may not sum to totals because of rounding.
(2.) The growth rates have been adjusted to remove the effects of an accounting change that lowered measured growth in loans and increased measured growth in federal funds sold. Before 1997, sales of federal funds by foreign offices were classified as loans. Starting in 1997, they are classified as federal funds sold.
(3.) This explanation presumes that the increase in the expected return on equity that occurs when capital is allocated to riskier assets increases the value of the bank's stock, but this need not be true. Just as selling $100 of safe stock and buying $100 of risky stock leaves ones net worth unchanged, replacing low-risk assets with high-risk assets should leave the value of the bank's stock essentially unaffected so long as the bank must pay appropriately higher rates on its liabilities. However, the rates banks pay on insured deposits are insensitive to a broad range of riskiness in bank assets, and the sensitivity of many other bank liabilities at the largest banks may be muted by the perception that regulators might be unwilling to allow such institutions to fail because of the damage to the financial system that could result.
(4.) For information on the securitization of credit card loans, see William R. Nelson and Brian K. Reid. "Profits and Balance Sheet Developments at U.S. Commercial Banks in 1995," Federal Reserve Bulletin, vol. 82 (June 1996), p. 488.
(5.) Core deposits consist of demand deposits, NOW accounts, savings and money market deposit accounts, and small (that is, less than $100,000) time deposits.
(6.) Goodwill is the difference between the acquisition price and the net fair value of the identifiable assets and liabilities acquired. Unrealized gains on available-for-sale investment account securities are the difference between the fair value of the securities and their amortized cost.
(7.) The tier 1 ratio is the ratio of tier 1 capital to risk-weighted assets, and the total ratio is the ratio of the sum of tier 1 and tier 2 capital to risk-weighted assets. Tier 1 capital consists mainly of common equity (excluding intangible assets such as goodwill and excluding net unrealized gains on investment account securities classified as available for sale) and certain perpetual preferred stock. Tier 2 capital consists primarily of subordinated debt, non-tier-1 preferred stock, and loan-loss reserves. Risk-weighted assets are calculated by multiplying the amount of assets and the credit-equivalent amount of off-balance-sheet items (an estimate of the potential credit exposure posed by the item) by the risk weight for each category, where the risk weights rise from zero to 1 as the credit risk of the assets increases. The leverage ratio is the ratio of tier 1 capital to average tangible assets. Tangible assets are equal to total assets less assets excluded from common equity in the calculation of tier 1 capital.
(8.) Well-capitalized banks are those with a total capital ratio greater than 10, a tier 1 ratio greater than 6, a leverage ratio greater than 5, and a composite CAMELS rating of 1 or 2.
(9.) Over the past two five-year periods, the return on assets improved even more than the return on equity. The increasing importance of off-balance-sheet activities in recent years, however, makes comparisons of return on assets over long periods of time potentially misleading. Nevertheless, a large fraction of banking is still tied to traditional on-balance-sheet items, and in interpreting changes in net income over shorter periods, assets remain a useful scaling factor for separating the effects of growth from those of improved profitability. By contrast, return on equity should not be affected by changes in the relative importance of off-balance-sheet activity because investors expect to receive an appropriate return on their investment regardless of whether activities are on or off the balance sheet. Returns on equity may, however, have been affected at least temporarily by the substantial increases in capital-to-asset ratios in recent years, which have in part been a response to regulatory changes.
(10.) Credit card banks are defined as banks among the top 1,000 for which credit card loans are more than half of assets. Primarily as a result of consolidation in this market segment, the number of credit card banks dropped from more than 40 at the end of 1995 to just 29 at the end of last year. See William R. Nelson and Ann L. Owen, "Profits and Balance Sheet Developments at U.S. Commercial Banks in 1996," Federal Reserve Bulletin, vol. 83 (June 1997), pp. 476-77, for a discussion of the profitability of credit card banks.
(11.) When the ownership of a bank changes substantially (for example, when it is bought by a holding company but retains its separate corporate existence), its assets and liabilities may be revalued according to the price paid by the acquiring firm for some or all of its shares. (In most cases revaluation is required.) Income items subsequently reported on the Call Report include earnings only since the date of the revaluation. This change in accounting basis is called push-down accounting because the revaluation adjustments made in the purchase by the acquiring firm are "pushed down" to the books of the acquired firm. Data on the banks applying push-down accounting are available only since 1995.
APPENDIX: ADJUSTMENTS TO THE REPORTED BANK INCOME DATA
Income and expense items are reported on quarterly Call Reports on a year-to-date basis. Complete industry income for a given year cannot, however, be collected from the year-end Call Reports because a number of factors can lead to discrepancies between income in a year and income reported at year-end. The data used in this article have been adjusted to eliminate, as far as possible, such discrepancies.
The most common problem is bank mergers handled under "purchase accounting." Under that method, the balance sheet items of the acquired bank are marked to market and then combined with those of the acquiring bank; the difference between the purchase price of the bank and the balance sheet value of identifiable assets and liabilities is reported as the intangible asset item "goodwill." The year-to-date flow of income and expense of the acquired bank as of the date of the merger goes unreported by the acquiring institution subsequent to the merger.
In contrast, "pooling of interest accounting" combines the balance sheets and income statements of the merging banks; the income statement of the successor institution for the year of the merger includes the income earned by each entity before the merger.
Beginning in 1995, data exist on the accounting method used for each bank merger. To calculate the adjustment required for mergers before 1995, we use the income data reported by the individual banks involved in the transaction to evaluate which of the accounting methods was the more likely to have been employed.
The income data in this article include an estimate of the income earned by banks acquired under purchase-accounting rules during the part of the year preceding the date of the merger. The estimate is based on the income reported by the acquired bank for those quarters preceding the merger and includes an estimate of the income earned in the quarter of the merger.
Two other situations that lead to discrepancies between actual industry income for a year and that reported on the fourth-quarter Call Reports are bank closures and the adoption by banks of "push down" accounting during the year.(11) Methods similar to those used for purchase mergers are used to estimate the income earned by such banks that is not reported at year-end.
In recent years the cumulative effects of the adjustments made to reported data have boosted industry net income about 1/2 percent relative to the aggregate income reported on fourth-quarter Call Reports. This increase in net income raised the average return on assets about 1/2 basis point. The effects were considerably larger in some earlier years.
RELATED ARTICLE: Off-Balance-Sheet Activity
Off-balance-sheet activities are of three general types. The first involves a promise by the bank to provide funds on demand (for example, a loan commitment) or as a guarantee (as with certain letters of credit). The obligation does not appear on the balance sheet because the funds have not been extended. The meaning of "off-balance-sheet" is somewhat more obscure when applied to the second type of activity, derivatives, because the value of most derivatives is reported on the balance sheet--as an asset if the value is positive or as a liability if the value is negative. Derivatives are assets whose payments are derived from the performance of other assets; it is the underlying assets that are off the balance sheet. The third type of activity, loan securitization, is generally spoken of as an off-balance-sheet activity because the securitized loans typically are moved off of banks' balance sheets. The table provides the year-end amounts of selected off-balance-sheet items in dollars and as a percentage of assets.
Selected off-balance-sheet items, year-end
1997 Percentage of assets Item (billions of dollars 1990 1997 Unused commitments 3,040.7 33.0 61.2 Letters of credit 29.2 .9 Commercial .6 Standby Financial 185.9 3.7 3.7 Performance 44.4 1.7 .9 Derivatives (excluding credit derivatives) Interest rate Notional amount 17,176.1 98.1 345.5 Fair value Positive 162.8 n.a. 3.3 Negative 161.3 n.a. 3.2 Exchange rate Notional amount 7,832.5 104.3 157.6 Fair value Positive 192.2 n.a. 3.9 Negative 185.5 n.a. 3.7 Other Notional amount 493.7 2.4 9.9 Fair value Positive 22.9 n.a. .5 Negative 22.7 n.a. .6 Credit derivatives (notional amount) Guarantor 33.4 n.a. .7 Beneficiary 63.7 n.a. 1.3 Assets transferred with recourse 230.6 n.a. 4.6
n.a. Not available
Unused commitments equaled $3 trillion at the end of 1997, nearly two-thirds of assets. The rise in unused lines as a share of assets since 1990 is almost entirely attributable to the growth of credit card lines, which tripled as a percentage of assets over the period and accounted for more than half of unused commitments last year. Unused residential and commercial real estate lines and unused lines for securities underwriting summed to less than 10 percent of unused commitments last year; other unused lines, primarily commercial and industrial, accounted for the remainder.
Letters of Credit
Banks issue commercial and standby letters of credit. Commercial letters of credit are issued specifically to facilitate payment for goods. They are arranged by the buyer to guarantee payment to the seller of the goods, who receives funds from the bank only when the terms of the purchase are fulfilled. Commercial letters of credit equaled less than 1 percent of bank assets last year.
A standby letter of credit is a promise by the issuing bank to pay a specific sum to a third party if the issuing bank's customer fails to fulfill specific commitments; the customer is then obligated to repay the funds to the bank. If the customer's commitments are financial, such as repaying holders of commercial paper, the letter is called a financial standby letter of credit. If the commitments are not financial, such as the delivery of merchandise or the completion of a construction project, the letter is called a performance standby letter of credit. Banks' potential obligations under financial standby letters equaled 3 3/4 percent of their assets last year; potential obligations under performance standby letters totaled about 1 percent of assets.
Derivatives can be roughly classified into two types: forwards and options. Forwards are agreements to buy or sell something for a specific price at a designated future date; options give the holder the opportunity, but not the obligation, to buy or sell something at a specific price, typically during an agreed-upon interval.(1) Swaps, in which the income streams from two assets are exchanged at specified future dates, are essentially a combination of several forward contracts. Most derivatives contracts held by banks are based on interest rates or exchange rates. At banks, the most common forms of interest rate contracts are swaps; exchange-rate contracts are most commonly forwards; and other derivatives are most often options.
Derivatives contracts are reported for accounting purposes in terms of their fair value, which is the price at which the contract could be replaced, and their notional value, which is generally the value of the underlying asset used in the computation of the payment streams. For example, an interest rate swap is commonly written so that its initial fair value is zero, that is, so that the present values of the bank's obligation to its counterparty and the counterparty's obligation to the bank are equal, even if the notional value--the reference amount used for calculating the income stream being swapped--is in the millions of dollars.
The difference between notional and fair values given in the example is reflected in the aggregate values: The total notional amount of banks' holdings of derivatives (excluding credit derivatives, which are discussed below) at the end of last year equaled $25 1/2 trillion, while the gross fair value of the derivatives (positive and negative) was about $750 billion. Notional amounts can be useful as one indicator of the change in the amount of derivatives activity over time. However, because notional amounts are so far removed from the actual value of derivatives, they vastly overstate the exposure of the institutions.
Derivatives holdings are concentrated at the largest banks. At the end of last year, more than 99 percent of derivatives, measured either by notional amount or gross fair value, were held at the top 100 banks. Furthermore, about 90 percent of derivative banks.
More than 90 percent of derivatives (by notional amount) were held in trading accounts last year. Such holdings frequently arise either as financial institutions trade among themselves or when a nonbank customer wishes to purchase a derivative and the bank acts as a counterparty. Given the volatility of these assets, banks rarely allow a position to be unmatched for very long. As a result, at any given time, the positive and negative fair values of banks' derivatives holdings tend to be about equal.
Another type of derivative contract, an option, is a "credit derivative," which allows parties to transfer the credit risk of an underlying asset. Generally, the derivatives are structured so that the seller (guarantor) will pay the buyer (beneficiary) if an asset held by the beneficiary defaults, thus allowing the beneficiary to hold the asset without being exposed to some or all of the credit risk of the asset. At the end of last year, the notional value of credit derivatives on which banks were the guarantors equaled 3/4 percent of bank assets, and the notional value of credit derivatives on which banks were the beneficiary equaled 1 1/4 percent of assets.
Although loan securitization is often spoken of as an off-balance-sheet activity, securitized assets are reported as an off-balance-sheet item only if the assets have been transferred with recourse; that is, if the bank has removed the asset from its balance of loss posed by the asset. When residential mortgages are securitized through one of the federal housing agencies, for example, the originating bank has no responsibility for the repayment of the loan (although it may service the mortgages for a fee), and thus the loan is not an off-balance-sheet item. In contrast, credit card securitizations are typically structured so that if the repayment performance of the underlying accounts deteriorates sufficiently, the originating bank is obliged to repurchase the remaining securitized loans over a fairly short period. Most of the loans that are reported as off-balance-sheet items on the Call Report, which equaled 4 1/2 percent of bank assets last year, were credit card loans.
(1.) For additional information on the use and holdings of derivatives by banks, see "Derivatives Disclosures by Major U.S. Banks, 1995," Federal Reserve Bulletin, vol. 82 (September 1996), pp. 791-805.
RELATED ARTICLE: International Operations of U.S. Banks
The share of U.S. bank assets that were booked at foreign offices increased about one-fourth, from 12 percent to 15 percent, between the end of 1993 and the end of 1997 (table).(1) The share of bank profits earned at foreign offices peaked at more than 16 percent in 1993 and was roughly 12 percent over the 1994 to 1996 period; the share dipped further in 1997 because foreign office results suffered in the second half of the year from the economic problems in Asia.
Share of U.S. bank assets and net income booked at foreign offices, 1993-97
Percent Year Assets Net income 1993 12.2 16.3 1994 13.2 11.9 1995 13.6 11.6 1996 14.8 12.0 1997 15.1 10.2 Q1 14.9 16.2 Q2 15.2 14.6 Q3 15.5 8.4 Q4 15.1 2.3
Responses to the Federal Reserve's Quarterly Report of Assets and Liabilities of Large Foreign Offices of U.S. Banks provide data on the geographical distribution of the assets and liabilities of major foreign branches and subsidiaries of U.S. banks. As has been the case for some time, about half of the assets reported on the survey at the end of 1997 were booked in European branches and subsidiaries. The bulk of the European assets were booked in the United Kingdom, a share reflecting, at least in part, the importance of London's financial markets. Nearly one-fourth of the reported assets were booked in Asian branches and subsidiaries, with the largest volumes in Hong Kong and Singapore. Large shares were also booked in the Caribbean (primarily the Cayman Islands and the Bahamas), with considerably smaller volumes in Latin America and elsewhere. The location in which an asset is booked is often a strong indicator of the nationality of the customer or the nature of the asset, but the interactions between U.S. and foreign regulations or tax laws can also influence the booking site.
Not surprisingly, banks with by far the largest share of assets and earnings at foreign offices were the largest banks (those with assets of more than $150 billion) (chart). Among these five banks, however, the scope of international operations varied considerably. Two of the banks held roughly three-fourths of their assets, and booked more than three fourths of their income, at foreign offices. On the other hand, at one of the banks, a "super regional" institution, foreign operations accounted for less than 4 percent of assets and an even smaller share of income.
[Chart ILLUSTRATION OMITTED]
(1.) Foreign offices include Edge Act and agreement subsidiaries and international banking facilities (IBFs). Edge Act and agreement subsidiaries are federally or state-chartered corporations, respectively, that are domiciled in the United States but engage in international banking activities. An IBF is a set of asset and liability accounts that cover selected international transactions of the U.S. offices of the bank. For more detail on the structure of foreign operations of U.S. banks, see James V. Houpt, International Trends for US. Banks and Banking Markets, Staff Studies 156 (Board of Governors of the Federal Reserve System, 1988).
|Printer friendly Cite/link Email Feedback|
|Author:||Nelson, William R.|
|Publication:||Federal Reserve Bulletin|
|Article Type:||Industry Overview|
|Date:||Jun 1, 1998|
|Previous Article:||Statements to Congress.|
|Next Article:||Treasury and Federal Reserve foreign exchange operations.|