Derivatives activity by U.S. banks: impact on risk and profitability.

Sundaram, Sridhar ; Willey, Thomas

Abstract

The primary objective of this study is to analyze and assess the impact of derivatives activity by U.S. Banks on their corresponding profitability and risk. Using quarterly data for the period of 1999 to 2006, our analysis focused on the twenty-five largest commercial U.S. banks. Our results indicate a positive contribution to trading revenue; however, this increase does not lead to an increase in profitability. Further, as derivative use increased, the risk of the banks in our sample increased.

I. INTRODUCTION

The stability and integrity of the U.S. banking system is crucially important to a sound and healthy U.S. and world economy. The financial crisis in the third and fourth quarter of 2008 has reinforced this critical role of financial institutions. Many market participants suggest the catalyst for these events can be traced back to sub-prime mortgage lending practices. These risky lending decisions, and subsequent defaults, began a cascading effect for both small and large intermediaries. The declines in market values and financial viability for Lehman Brothers, Merrill Lynch and AIG serve as examples of this downward spiral. However, increased leverage and risk exposure is not unique to the most recent economic period. In the 1990s, the increased use of derivatives by the banking industry and a spate of derivative disasters (bankruptcy of Orange County, collapse of Barings PLC, Bankers Trust and Procter and Gamble, derivative related losses at Mettallgesellschaft AG and NatWest Markets, to name a few) drew the attention of the regulators and investors (see Kuprianov (1995) for a discussion of derivative debacles).

Bank derivative usage increased from $32.5 trillion (notional value) in the first quarter 1999 to $122.5 trillion in the fourth quarter of 2006 (see Tables 1 and 2). This represents an annual compounded growth of over 18%. This is a much higher than the growth rate of total assets for these banks. In recent years, total assets are only 5-6% of the total notional value of the derivative contracts. Of these contracts, majority of the derivative positions today represent interest rate contracts (over 85%; OCC bank derivatives report, see Table 3). Further, the majority of the derivative contracts are traded in the OTC market leading to a higher credit risk. On average, 90% of the contracts are OTC traded and only 10% are exchange traded (Table 3). Given this significant growth in the use of derivatives by banks, a reason for concern by regulators is the lack of knowledge concerning the impact of these activities on the risk and profitability of the banking industry.

II. BANK PROFITABILITY

Derivatives are financial instruments that derive their value from the performance of an underlying financial asset, such as interest rates, currency exchange rates or indexes. Banks involve themselves with derivative instruments in two ways, as dealers (about 95%) and as end-users (about 5%). For the fourth quarter 2006, the total derivative notional value in commercial bank portfolio was $122.5 trillion, of which end user notional value and dealer notional value were $2.8 trillion (2%) and $119.6 trillion (98%), respectively (see Table 1 and Figure 1).

[FIGURE 1 OMITTED]

By acting as derivative dealers, banks are able to generate fee income from primarily corporate clients looking to reduce their risk exposure. Bank source of revenue from dealing derivatives takes three forms: transaction fees, bid-offer spreads, and trading profits. Banks reported trading revenue of about $19 billion in 2006 (see Table 4). This represents on average about 5% of gross revenue for the largest banks with significant derivative activity.

Studies examining the impact of derivatives on bank profitability include Holland (1988), who examined the relation between return on equity and return on assets to the ratio of derivatives to total assets held by banks. In both cases, Holland found no significant relationship between the ratio of derivatives to total assets and the profitability of the bank. He also performed a mean test between the largest twenty-five banks and the smallest twenty-five banks in his population (the one hundred largest U. S. banks). The conclusion from that test was that banks with relatively high levels of derivatives are generally less profitable than banks with relatively low levels of such activities. Hindi and Sundaram (2002) examined the impact of off-balance sheet activities by banks and found that a significant positive relation exists between the notional value of the off-balance sheet positions (OBSAs) and ROA, ROE and non-interest income of the banks. They also find that a significant negative exposure exists between the credit exposure of these OBSAs and bank profitability.

III. BANK RISK: RISK FROM HEDGING WITH DERIVATIVES

As end-users, banks use derivatives largely for hedging purposes. Banks build up positions in derivatives as a consequence of market-making activities. These positions leave the banks exposed to financial market risks. To hedge against these risks, banks engage in offsetting trades. Banks' traditional activities can also result in exposure to financial market risks. For example, mortgage lending increases a bank's exposure to interest rate risk. As a result, banks may engage in derivative activity to hedge this risk. Participation in derivatives for hedging purposes is only a small part of the overall derivative activity. Gibson (2007) presents a discussion of the usefulness of derivatives for hedging the risk exposure of a commercial bank, an investment bank and a hedge fund investor. Table 2 indicates that on average only a relatively small percentage of all derivatives held by banks are for non-trading purposes.

The primary risk of hedging lies in the relationship between the value of the hedged asset and the value of the derivatives contract. This relationship is referred to as correlation. A higher correlation results in lower risk from the hedging activity. One measure of the effectiveness of derivative hedging activity for a bank is its effect on the net operating income. Brewer, Jackson and Moser (1996) find that equity return volatility is negatively related to S&L's participation in derivatives.

There are a number of recent articles on the effects of derivatives instruments on the risk of financial intermediaries. Kim and Koppenhaver (1993) investigated the use of interest rate swaps by commercial banks. The banks were partitioned into two distinct groups: users and suppliers. The suppliers, either dealers and/or market makers, were more likely to hold higher amounts of the swaps than the user banks. In addition, their findings supported the positive relationship between interest rate sensitivity and propensity to employ swaps. Jagtiani, Saunders and Udell (1997) explored the off-balance sheet instruments of derivatives and standby letters of credit for the period of 1984 to 1991 for commercial banks. Although their findings indicate significant differences on a cross-sectional basis for banks and their activities, no relationship was indicated for the bank specific characteristics of size, leverage, creditworthiness and derivative use. Koski and Pontiff (1999) examine the use of derivatives instruments in the mutual fund industry. They find similar risk exposure and performance between derivatives users and non-users. In a study of large commercial banks, Carter and Sinkey (1998) find that banks with assets of more than $100 million, but less than $1 billion, use interest-rate derivatives to hedge their exposure to interest-rate risk. In a subsequent paper, however, Sinkey and Carter (20005 p. 446) conclude "that user banks, compared to nonuser banks, are associated with riskier capital structures (less equity capital, but more notes and debentures), larger maturity mismatches between on-balance sheet assets and liabilities, greater net loan charge-offs, and lower net interest margins". Although they acknowledge data limitations in their work, Gorton and Rosen (1995, p. 32) find that "banks seem to have hedged most of the risk" associated with interest rate swaps and therefore, there is little net interest-rate risk for the banking system in aggregate. Guay (1999) examined firms that initiate derivatives use and found that firms use derivatives to hedge and not to increase, entity risk. Adkins, Carter and Simpson (2007) explored the role of compensation and ownership on the use of foreign-exchange derivatives by U.S. bank holding companies. Their empirical analysis focuses on derivatives used for hedging, as opposed to trading, and the authors found a positive relationship between the use of derivatives and managerial ownership. In addition, hedging activities were influenced by the degree of ownership of large institution stockholders. Minton, Stulz and Williamson (2008) examined the use of credit derivatives by large U.S. bank holding companies during the period of 1999 to 2005 and found that only 5.8% of the firms used credit derivatives. In addition, the authors found that some of the factors that lead to use of derivatives were a lower capital ratio, higher foreign loan originations, less agricultural loans and more commercial and industrial loans.

IV. BANK RISK: RISK FROM DEALING DERIVATIVES

Banks participating in the derivative markets as dealers are exposed to three major risks. These risks are market risk, credit risk and operational risk. Becketti (1993) and Puwalski (2003) provide an excellent discussion on derivatives risk in commercial banking.

Market Risk: The market risk for a dealer bank may result from fluctuations in the value of a derivative contract, reductions in market liquidity for derivative security, uncertainty over settlement and vulnerability to cross-market fluctuations. The level of the market risk is largely influenced by the extent to which these banks engage in matched trading, market making or positioning. Matched trading will result in the lowest level of market risk, while extensive positioning will result in the highest amount of market risk.

Credit Risk: The credit risk for a dealer bank results primarily from two sources: the risk of default by a counterparty and the risk from changes in credit exposure. The risk of default is an important risk for banks, since their participation is heavily in OTC derivatives (see Table 3). The evaluation of future credit exposure in derivative positions is a challenge for many banks. But, the use of netting agreements by banks has significantly reduced this credit exposure (see Figure 2).

Operational Risk: The largest losses to date from derivatives activities have resulted from operational failures. Operational risk refers to risks associated with monitoring and controlling risk-taking of bank employees. From the collapse of the Barings Bank in 1995 to the recent derivative debacle at National Australia Bank in 2004, derivative traders who circumvented risk-management controls and ultimately exposed the institutions to significant losses from derivative positions fall into this category. Another important risk is the accurate valuation of a derivative position. NatWest Markets in 1997 suffered large losses stemming from the mispricing of interestrate options by a bank employee. Others operations risks include legal risk (for example, Procter and Gamble and Bankers Trust) and regulatory risk (for example, Basle II).

The potential for financial institutions to use derivatives instruments for speculative purposes has been the subject of several recent papers. Peek and Rosengren (1996) observe that a relatively high percentage of large banks active in derivatives markets were "troubled banks", ones possibly seeking to place speculative "second bets". Simons (1995) also found that larger banks with weaker asset quality tended to use derivatives more, than larger banks with stronger asset quality. Brewer, Jackson and Moser (1996, p. 495), however, conclude that "despite the existence of extreme moral hazard incentives during the last half of the 1980s few, if any, of our sampled S&Ls used derivatives to shift risk onto the deposit insurance fund". Instefjord (2005) identifies the potential benefits of derivatives in hedging and risk sharing and the potential costs of derivatives by increasing a bank's risk exposure due to increased speculation.

[FIGURE 2 OMITTED]

V. RESEARCH OBJECTIVE

In large part because of the relatively recent and rapid increase in the use of derivatives by banks, the overall effects of derivatives usage in the banking industry are still not completely known. Two particularly important issues that remain open are the effects of derivatives on bank risk and its profitability.

The primary objective of this research is to analyze and assess the potential impact of derivatives activity by U.S. banks on its risk and profitability. Specifically, does the dominant form of derivative contracts outstanding, the interest rate agreement, generate a majority of the profits for commercial banks? The study examines the twenty-five U.S. commercial banks with the largest derivative positions for each quarter. The holdings of derivatives are concentrated in the largest banks. The top twenty-five banks account for 99 percent of the total notional amount of derivatives in the U.S. commercial banking system (OCC Bank Derivatives Report). Hence, an examination of the twentyfive largest banks will provide a comprehensive overview of the impact of derivatives on bank profitability and risk.

VI. DATA AND METHODOLOGY

The Federal Financial Institutions Examination Council (FFIEC), which coordinates reporting requirements across the bank regulatory agencies, approved an expansion of the Call Reports to include more detail about financial derivatives, beginning in March 1995. The new reports have made a finer distinction among various types of derivative contracts, including separate categories for over-the-counter and exchange traded derivatives (varying credit risk), for all banks. Additionally, these banks must distinguish between positive and negative amounts of each type of derivative contract, report fair values and the purpose of the contracts. Finally, banks will report the income they earned from derivative contracts, making a distinction between trading contracts and hedging contracts.

The above information, provided in the Call Reports since 1995, will allow us to study the relation between the use of derivatives by U.S banks and their risk and profitability in a comprehensive fashion. The sample period used in the study is from first quarter 1.999 to fourth quarter 2006. Data on credit derivatives were reported beginning in 1998. Given the significant increase in the use of credit derivatives by banks in recent years, we felt it would be important to examine their impact on banks. The call report data are provided by the Chicago Fed. SAS data set files for each quarter with the complete call report data were downloaded from the Chicago Fed website: (http://www.chicagofed.org/economic_research_and_data/ commercial_bank_data.cfm).

We use cross sectional data for the twenty-five U.S. commercial banks with the largest derivatives position for each quarter. Data is collected for twenty-four quarters starting with first quarter 1999 and ending with fourth quarter 2006. Regression analysis is performed to examine the relation between the derivative usage by these twenty-five banks and their profitability and risk Profitability is examined by looking at the impact of the derivative usage on trading revenue. Impact on the risk of the banks is examined by testing the relation between credit losses from derivatives and the use of derivatives.

VII. EMPIRICAL RESULTS

Bank profitability is examined in Table 4. Results indicate that in the first quarter of 1999 trading revenue amounted to about $3 billion, which represents about 10% of gross income. Both, total gross income and trading revenue continued to increase in recent years. But, toward the end of 2006 trading revenue as a percentage of total gross income declined to only about 5% (half of what it was in 1999). Another interesting fact that reflected in Table 4 is the contribution of trading revenue by the various types of contracts. Traditionally, interest rate and foreign exchange contracts provided most of the trading revenue (about 80-85%). While most of the derivative contracts are interest rate contracts, it does not provide the majority of the trading revenue. Foreign exchange contracts appear to be more profitable by providing nearly half the trading revenue, even though they represent only 10% of the derivative contracts participated by these banks. Also, trading revenue from equity and index contracts and commodity contracts have increased significantly in 2006.

Regression analysis (Table 5) shows that usage (notional amounts of derivatives) of derivatives is negatively related to the ROA and ROE. But, it is significantly and positively related to trading revenue generated by these banks. These contrasting relationships may highlight an interesting fact that while dealing in derivatives directly contribute to trading revenue, other financial characteristics of these banks appear to result in lower return to these banks. These results support the findings by Sinkey and Carter (2000) that banks using derivatives are associated with riskier capital structures, larger maturity mismatches and greater net loan charge-offs. These results warrant further investigation to examine the impact of the derivative usage on the overall profitability of these banks.

Table 6 contains our results for bank risk. The notional amount of' derivatives are not an accurate measure of the risk exposure of banks due to bilateral netting agreements between the banks and their corporate customers. Figure 2 indicates that the netting benefit is over 80% in recent years. Given the netting benefit, the total credit exposure for banks is measured as the bilaterally netted current exposure and the future exposure. Table 6 shows that total credit exposure has increased as a percentage of total assets while decreasing as a percentage of risk-based capital during 2002-05 but shows a significant increase in 2006. This reflects the increase in risk-based capital maintained by these banks while highlighting the increase in credit exposure from the derivative positions held by these banks.

Banks use derivatives for hedging purposes. The impact of these derivatives on interest income, interest expense and other allocations are reported in Table 7. The net impact of these derivatives has been consistently positive, supporting the use of derivatives for hedging purposes. Credit losses from banks are reported in Table 7. Testing the relation between credit losses and the use of derivatives we find a significant relation between them. Table 8 confirms a significant positive relation between the level of derivative usage and the amount of credit losses for these banks. Further examination reveals that majority of the credit losses can be attributed to interest rate contracts held for trading purposes by these banks. Table 9 reports that the only variable significant in the regression analysis testing the impact of the various types of derivative contracts on credit losses by banks is the interest rate contracts.

VIII. CONCLUSIONS

The primary objective of this study is to analyze and assess the impact of derivatives activity by U.S. banks on its risk and profitability. The study examines the twenty-five U.S. commercial banks with the largest derivative positions for each quarter since derivatives are concentrated in the largest banks. The basic results can be summarized as follows:

Profitability: The dealing in derivatives by U.S banks contribute directly to its trading revenue, but it does not result in the increase of the overall bank profitability. Derivative activities for hedging purposes resulted in a net positive impact on the financial performance of these banks.

Risk: The use of derivatives by U.S. commercial banks resulted in an increase in the credit exposure to these banks. But, the credit exposure was substantially reduced due to netting arrangements.

Our initial analysis shows that the derivative usage by U.S. commercial banks has a significant impact on their risk and profitability. This warrants further analysis to examine the negative impact of the derivatives on bank ROA and ROE. Further, analysis of the risk needs to examine the impact of these derivatives on the market risk and operational risk of these banks.

References

Adkins, L. C., D. A. Carter and W. G. Simpson (2007), "Managerial Incentives and the Use of Foreign-Exchange Derivatives", Journal of Financial Research, Vol. 30, No. 3, 399-413.

Becketti, S. (1993), "Are Derivatives Too Risky for Banks?", Economic Review, 27-42.

Brewer III, E., W. E. Jackson and J. T. Moser (1996), "Alligators in the Swamp: The Impact of Derivatives on the Financial Performance of Depository Institutions", Journal of Money, Credit and Banking, Vol. 28, 482-497.

Carter, D. A. and J. F. Sinkey, Jr. (1998), "The Use of Interest Rate Derivatives by End-users: The Case of Large Community Banks", Journal of Financial Services Research, Vol. 14, 17-34.

Gibson, M. S. (2007), "Credit Derivatives and Risk Management", FEDS Working Paper No. 2007-47, October.

Gorton, G. and R. Rosen (1995), "Banks and Derivatives", National Bureau of Economic Research, Working Paper No. 5100.

Guay, W. R. (1999), "The Impact of Derivatives on Firm Risk: An Empirical Examination of New Derivative Users", Journal of Accounting and Economics, Vol. 26, 319-351.

Hindi, N. and S. Sundaram (2002), "The Relation Between Off-Balance Sheet Activities and U.S. Bank Profitability", Journal of Accounting and Finance Research, Vol. 10, No. 3, 15-25.

Holland, D. S. (1988), "An Examination of the Off-Balance Sheet Activities of U.S. Banks", DBA Dissertation, George Washington University.

Instefjord, N. (2005), "Risk and Hedging: Do Credit Derivatives Increase Bank Risk?", Journal of Banking and Finance, Vol. 29, No. 2, 333-345.

Jagtiani, J., A .Saunders and G. Udell (1997), "Financial Innovations and the Growth of Bank Derivative Activities', in Schacher, Barry ed. Derivatives, Regulation and Banking, (Elsevier: Amsterdam), 7-40.

Kim, S. H. and G. D. Koppenhaver (1993), "An Empirical Analysis of Bank Interest Rate Swaps", Journal of Financial Services Research, Vol. 7, No. 1, February, 57-72.

Koski, J. L. and J. Pontiff(1999), "How Are Derivatives Used? Evidence from the Mutual Fund Industry", The Journal of Finance, Vol. 54, 791-816.

Kuprianov, A. (1995), "Derivative Debacles: Case Studies of Large Losses in Derivatives Markets", Economic Quarterly, Vol. 81, 1-39.

Minton, B. A., R. M. Stulz and R. G. Williamson (2008), "How Much Do Banks Use Credit Derivatives to Hedge Loans?", Fisher College of Business, Working Paper No. 2008-03-001, January.

Peek, J. and E. S. Rosengren (1997), "Derivatives Activity at Troubled Banks", Journal of Financial Services Research, Vol. 12,287-302.

Puwalski, A. C. (2003), "Derivatives Risk in Commercial Banking", FDIC: FYI. http://www.fdic.gov/bank/analytical/fyi/2003/032603fyi.html

Simons, K. (1995), "Interest Rate Derivatives and Asset-Liability Management by Commercial Banks", New England Economic Review, Federal Reserve Bank of Boston, 17-28.

Sinkey, J. F., Jr. and D. A. Carter (2000), "Evidence on the Financial Characteristics of Banks that do and do not use Derivatives", The Quarterly Review of Economics and Finance, Vol. 40,431-449.

SRIDHAR SUNDARAM AND THOMAS WILLEY

Seidman College of Business Grand Valley State University

Table 1
Total Notional Value of Derivatives for All U.S.
Insured Commercial Banks: By Type of Use
(Dollars are in Billions)

Year   Quarter   End--User    Dealer       Total
                 Nationals   Nationals   Nationals

1999     Q1         1.4        31.0        32.4
1999     Q2         1.5        31.3        32.8
1999     Q3         1.5        33.9        35.4
1999     Q4         1.6        33.0        34.6
2000     Q1         1.6        35.7        37.3
2000     Q2         1.7        37.3          39
2000     Q3         1.5        36.5          38
2000     Q4         1.2        38.9        40.1
2001     Q1         1.2        42.4        43.6
2001     Q2         1.2        46.2        47.4
2001     Q3         1.3        49.6        50.9
2001     Q4         1.8        43.2          45
2002     Q1         1.9        43.9        45.8
2002     Q2         2.0        47.5        49.5
2002     Q3         2.4        50.2        52.6
2002     Q4         2.1        53.3        55.4
2003     Q1         2.4        58.3        60.7
2003     Q2         2.6        62.4          65
2003     Q3         2.5        63.7        66.2
2003     Q4         2.4        67.7        70.1
2004     Q1         2.5        72.8        75.3
2004     Q2         2.5        76.9        79.4
2004     Q3         2.6        79.7        82.3
2004     Q4         2.6        82.9        85.5
2005     Q1         2.5        85.5          88
2005     Q2         2.5        89.6        92.1
2005     Q3         2.6        91.1        93.7
2005     Q4         2.6        93.0        95.6
2006     Q1         2.6       102.1       104.7
2006     Q2         2.6       110.1       112.7
2006     Q3         3.0       115.3       118.3
2006     Q4         2.8       119.6       122.4

Source: OCC Bank Derivatives Reports.

Table 2
Derivative Notional Value for the 25 U.S.
Commercial Banks with the Most Off-Balance
Sheet Derivative Contracts
(Dollars are in Millions)

                               National
Year   Quarter     Total       Value of
                  Assets      Derivatives

1999     Q1      2,524,790    32,315,259
1999     Q2      2,527,800    32,655,874
1999     Q3      2,534,928    35,351,775
1999     Q4      2,694,488    34,541,266
2000     Q1      2,805,873    37,358,124
2000     Q2      2,867,338    39,026,989
2000     Q3      2,922,981    38,023,636
2000     Q4      3,020,272    40,295,768
2001     Q1      3,166,526    43,664,443
2001     Q2      3,167,858    47,558,456
2001     Q3      3,403,911    50,987,118
2001     Q4      3,385,647    45,110,369
2002     Q1      3,320,093    46,067,544
2002     Q2      3,463,806    49,816,197
2002     Q3      3,579,995    52,890,512
2002     Q4      3,704,749    55,786,207
2003     Q1      3,759,506    61,098,701
2003     Q2      3,987,177    65,505,944
2003     Q3      3,915,971    66,773,883
2003     Q4      3,915,971    66,773,883
2004     Q1      4,303,204    76,139,086
2004     Q2      4,407,754    80,632,014
2004     Q3      4,488,259    83,801,980
2004     Q4      4,741,381    87,503,917
2005     Q1      4,850,847    90,721,979
2005     Q2      5,055,874    95,810,044
2005     Q3      5,154,634    98,380,614
2005     Q4      5,196,408   101,048,641
2006     Q1      5,405,674   109,751,560
2006     Q2      5,615,214   118,769,408
2006     Q3      5,851,822   125,722,807
2006     Q4      6,172,456   131,043,649

        Total
       Assets to  % Held        % Not
Year   National     for          for
        Value     Trading      Trading

1999    7.8%       96.3%        3.4%
1999    7.7%       96.2%        3.8%
1999    7.2%       96.4%        3.6%
1999    7.8%        96%          4%
2000    7.5%       96.1%        3.9%
2000    7.3%       96.2%        3.8%
2000    7.7%       96.7%        3.3%
2000    7.5%       97.5%        2.5%
2001    7.3%       97.7%        2.3%
2001    6.7%       97.9%        2.1%
2001    6.7%       97.7%        2.3%
2001    7.5%       96.4%        3.6%
2002    7.2%       96.1%        3.9%
2002    7.0%       96.2%        3.8%
2002    6.8%       95.8%        4.2%
2002    6.6%       96.5%        3.5%
2003    6.2%       96.4%        3.6%
2003    6.1%       96.3%        3.7%
2003    5.9%       96.6%        3.4%
2003    5.9%       96.9%        3.1%
2004    5.7%       97.0%        3.0%
2004    5.5%       97.1%        2.9%
2004    5.4%       97.2%        2.8%
2004    5.4%       97.2%        2.8%
2005    5.3%       97.4%        2.6%
2005    5.3%       97.6%        2.4%
2005    5.2%       97.5%        2.5%
2005    5.1%       97.5%        2.5%
2006    4.9%       97.8%        2.2%
2006    4.7%       97.9%        2.1%
2006    4.6%       97.7%        2.3%
2006    4.7%       97.9%        2.1%

Source: OCC Bank Derivatives Reports.

Table 3
Derivative Contracts for the 25 U.S. Commercial Banks with the Most Off-Balance
Sheet Derivative Contracts: By Exchange and Type of Contracts
(Dollars are in Millions)

                                 Exchange         OTC     Interest
                   National        Traded    Contracts        Rate
                   Value of    Contracts %       % of    Contracts
Year   Quarter   Derivatives     of Total       Total    % of Total

1999       Q1     32,315,259         12.0%       87.0%        75.8%
1999       Q2     32,655,874         10.9%       88.0%        77.0%
1999       Q3     35,351,775          9.5%       89.6%        78.4%
1999       Q4     34,541,266          9.2%       90.0%        79.0%
2000       Q1     37,358,124          9.2%       90.1%        79.5%
2000       Q2     39,026,989          9.0%       90.2%        79.2%
2000       Q3     38,023,636          8.3%       90.9%        79.9%
2000       Q4     40,295,768          9.8%       89.6%        80.7%
2001       Q1     43,664,443          8.8%       90.6%        80.8%
2001       Q2     47,558,456         11.8%       87.6%        82.4%
2001       Q3     50,987,118         13.3%       86.1%        83.7%
2001       Q4     45,110,369         11.4%       88.0%        83.9%
2002       Q1     46,067,544         11.4%       88.1%        84.3%
2002       Q2     49,816,197          9.9%       89.6%        84.8%
2002       Q3     52,890,512         12.1%       87.4%        85.5%
2002       Q4     55,786,207         11.0%       88.5%        85.8%
2003       Q1     61,098,701         11.6%       87.8%        86.6%
2003       Q2     65,505,944         12.3%       87.2%        86.0%
2003       Q3     66,773,883          9.9%       89.6%        86.4%
2003       Q4     66,773,883          9.8%       89.8%        86.6%
2004       Q1     76,139,086          8.5%       91.0%        86.0%
2004       Q2     80,632,014         10.0%       89.5%        86.7%
2004       Q3     83,801,980          8.1%       91.4%        86.3%
2004       Q4     87,503,917          7.4%       92.2%        85.6%
2005       Q1     90,721,979          8.8%       90.8%        85.2%
2005       Q2     95,810,044          8.3%       86.0%        80.7%
2005       Q3     98,380,614          9.1%       90.5%        83.6%
2005       Q4    101,048,641          7.2%       92.4%        82.9%
2006       Q1    109,751,560          8.6%       91.0%        83.4%
2006       Q2    118,769,408          8.4%       91.2%        82.5%
2006       Q3    125,722,807          8.7%       91.0%        81.5%
2006       Q4    131,043,649          7.4%       92.3%        81.4%

          Foreign
         Exchange
        Contracts       Other        Credit
       Derivatives   Contracts   Derivatives
        Contracts        % of          % of
Year   % of Total       Total         Total

1999         20.3%        2.3%          0.6%
1999         19.0%        2.3%          0.6%
1999         17.8%        2.3%          0.7%
1999         16.9%        2.4%          0.8%
2000         16.2%        2.7%          0.8%
2000         16.4%        2.7%          0.9%
2000         15.7%        2.7%          1.0%
2000         15.0%        2.7%          1.0%
2001         15.3%        2.5%          0.8%
2001         13.9%        2.4%          0.7%
2001         12.8%        2.2%          0.7%
2001         12.5%        2.1%          0.9%
2002         12.1%        2.1%          0.9%
2002         11.5%        2.1%          1.0%
2002         10.9%        2.0%          1.1%
2002         10.8%        1.8%          1.1%
2003         10.1%        1.7%          1.2%
2003         10.7%        1.5%          1.2%
2003         10.3%        1.6%          1.3%
2003         10.1%        1.5%          1.4%
2004         10.4%        1.5%          1.6%
2004          9.5%        1.4%          1.8%
2004          9.4%        1.6%          2.3%
2004          9.8%        1.6%          2.7%
2005          9.3%        1.6%          3.4%
2005          8.8%        1.5%          3.2%
2005          9.0%        1.9%          5.2%
2005          9.1%        1.8%          5.7%
2006          9.3%        1.9%          5.0%
2006          9.4%        2.2%          5.5%
2006          8.9%        3.0%          6.3%
2006            9%        2.4%          6.9%

Source: Bank Call Reports, Economic Research and Data, Federal
Reserve Bank of Chicago website.

Table 4
Trading Revenue from Derivative Contracts for the Top Five U.S. Commercial
Banks with the Most Off-Balance Sheet Derivative Contracts:
By Type of Contracts (Dollars are in Millions)

Year   Quarter   Gross      Trading     Total    Interest
                 Income     Revenue    Trading       Rate
                          % of Gross   Revenue   Contracts
                             Income              % Trading
                                                      Rev

1999   Q1        31,302         9.6%     3,005       41.5%
1999   Q2        29 155         5.8%     1,691       35.1%
1999   Q3        30 316         5.7%     1,728       40.4%
1999   Q4        38 407         5.4%     2,074       34.0%
2000   Q1        39 711         8.3%     3,296       46.3%
2000   Q2        41 274         6.2%     2,559       33.1%
2000   Q3        41 545         5.5%     2,285       39.4%
2000   Q4        45 200         5.0%     2,260       40.9%
2001   Q1        45 465         7.1%     3,228       46.2%
2001   Q2        39 000         5.8%     2,262       53.4%
2001   Q3        38 614         7.0%     2,703       45.7%
2001   Q4        32 761         6.7%     2,195       51.4%
2002   Q1        34 763         7.9%     2,642       48.0%
2002   Q2 *      59,479         7.6%     2,855       48.9%
2002   Q3        48,684         4.8%     1,850       56.0%
2002   Q4        34,158         3.8%     1,298       36.1%
2003   Q1        35,439         6.6%     2,339       29.8%
2003   Q2        36,108         6.5%     2,347       41.5%
2003   Q3        35,044         6.8%     2,383       43.2%
2003   Q4        36,428         4.2%     1,530       30.1%
2004   Q1        37,086         8.1%     3,004       48.1%
2004   Q2        40,618         5.5%     2,234       13.0%
2004   Q3        36,700         2.0%       734     -207.8%
2004   Q4        43,513         3.7%     1,610      -28.4%
2005   Q1        48,883         7.7%     3,764       36.0%
2005   Q2        51,400         3.0%    1,542        13.7%
2005   Q3        57,310         7.1%     4,069       47.8%
2005   Q4        57,767         4.3%     2,484       28.4%
2006   Q1        87,517         5.6%     4,901       22.6%
2006   Q2        67,227         6.6%     4,437       30.3%
2006   Q3        70,259         5.4%     3,794        9.6%
2006   Q4        66,739         4.6%     3,070       36.9%

Year    Foreign      Equity     Commodity
       Exchange    and Index    and other
       Contracts   Contracts   Derivatives
       % Trading   % Trading    % Trading
            Rev         Rev           Rev

1999       41.3%        9.4%          7.8%
1999       47.1%       14.8%          3.0%
1999       45.5%       11.2%          2.9%
1999       34.8%       21.7%          9.5%
2000       29.9%       18.7%          5.1%
2000       39.8%       19.8%          7.3%
2000       34.7%       22.5%          3.4%
2000       41.4%       14.4%          3.3%
2001       30.8%       20.4%          2.6%
2001       23.9%       18.0%          4.7%
2001       40.8%       10.2%          3.3%
2001       31.6%       18.4%          1.4%
2002       37.8%       13.9%          0.3%
2002       35.7%       16.1%          0.7%
2002       35.0%        6.1%         15.1%
2002       69.3%        7.2%          1.8%
2003       47.3%       20.0%          2.9%
2003       49.6%       14.5%          5.6%
2003       44.7%        8.6%          3.5%
2003       55.0%       12.3%          2.6%
2004       29.9%       19.2%          2.8%
2004       54.0%       14.7%         18.3%
2004      109.7%       59.2%        138.7%
2004      101.4%       20.0%          7.0%
2005       35.9%       22.5%          5.6%
2005       69.9%        5.6%         10.8%
2005       20.4%       19.6%         12.2%
2005       51.0%       31.2%         10.6%
2006       36.7%       34.5%          6.2%
2006       41.6%       21.7%          6.4%
2006       29.4%       40.4%         20.6%
2006       37.6%       29.8%          4.3%

Source: Bank Call Reports, Economic Research and Data, Federal
Reserve Bank of Chicago website.

Table 5
Estimates of Slope Coefficients from Regressing ROA, ROE, FEE
(Non-interest Income/Net Income), and TRADREV (Trading
Revenue/Net Income) on National Amounts of All Derivatives

ROA = [alpha] + [[beta].sub.1] (DERIVATIVE/TA) + [[epsilon].sub.i]
ROE = [alpha] + [[beta].sub.1] (DERIVATIVE/TA) + [[epsilon].sub.i]
FEE = [alpha] + [[beta].sub.1] (DERIVATIVE/TA) + [[epsilon].sub.i]
TRADREV = [alpha] + [[beta].sub.1] (DERIVATIVE/TA) + [[epsilon].sub.i]

          Regression
Measure   Coefficient    F-Value    Adj R-Sq

ROA         -0.0001      12.24      0.0274
(n=400)     (0.0005)     (0.0005)
ROE         -0.0008       6.16      0.0128
(n=400)     (0.0135)     (0.0135)
FEE          0.0688       1.56      0.0014
(n=400)     (0.2119)     (0.21.19)
TRADREV      0.03235     73.15      0.1531
(n=400)     (0.0001)     (0.0001)

* The probabilities are presented in parentheses below
the regression coefficient and the F-value statistic.

ROA = Net Income / TA
ROE = Net Income/Total Equity
FEE = Non Interest Income/Net Income
TRADREV = Trading Revenue / Net Income

Table 6
Credit Exposure for the 25 Commercial
Banks with the Most Off-Balance Sheet
Derivative Contracts (Dollars are in Millions)

Year   Quarter     Notional    Bilaterally     Future       Total
                   Value of        Netted    Exposure      Credit
                 Derivatives      Current                Exposure
                                 Exposure                from All
                                                         Contracts

1999       Q1     32,315,259       170,438     208,719    379,157
1999       Q2     32,655,874       147,327     211,453    358,780
1999       Q3     35,351,775       151,638     231,387    383,025
1999       Q4     34,541,266       162,567     229,508    392,075
2000       Q1     37,358,124       170,046     243,601    413,647
2000       Q2     39,026,989       144,027     270,003    414,030
2000       Q3     38,023,636       138,634     256,811    395,445
2000       Q4     40,295,768       153,763     274,390    428,153
2001       Q1     43,664,443       175,269     315,463    490,733
2001       Q2     47,558,456       161,530     341,011    502,542
2001       Q3     50,987,118       183,440     369,397    552,838
2001       Q4     45,110,369       165,483     311,491    476,973
2002       Q1     46,067,544       130,894     310,093    440,987
2002       Q2     49,816,197       181,195     337,925    519,120
2002       Q3     52,890,512       214,754     347,483    562,237
2002       Q4     55,786,207       213,040     373,603    586,644
2003       Q1     61,098,701       221,503     433,566    655,069
2003       Q2     65,505,944       231,098     484,326    715,424
2003       Q3     66,773,883       202,091     507,660    709,751
2003       Q4     66,773,883       212,524     535,148    747,673
2004       Q1     76,139,086       201,157     570,016    771,174
2004       Q2     80,632,014       168,475     577,007    745,482
2004       Q3     83,801,980       179,146     617,596    796,743
2004       Q4     87,503,917       216,691     677,655    894,346
2005       Q1     90,721,979       194,696     716,985    911,682
2005       Q2     95,810,044       196,775     765,871    962 646
2005       Q3     98,380,614       209,034     841,948  1,050,982
2005       Q4    101,048,641       187,383     911,057  1,098,439
2006       Q1    109,751,560       185,805   1,120,455  1,306,260
2006       Q2    118,769,408       195,006   1,196,441  1,391,447
2006       Q3    125,722,807       171,955   1,320,044  1,491,999
2006       Q4    131,043,649       180,993   1,404,241  1,585,235

Year     % Total        % Total
          Credit         Credit
        Exposure    Exposure to
       to Capital         Total
           Ratio         Assets

1999       102.7%            15%
1999        98.9%          14.2%
1999        92.8%          15.1%
1999        89.9%          14.5%
2000        86.8%          14.7%
2000       84.63%          14.4%
2000       81.49%          13.5%
2000       86.54%          14.2%
2001       104.0%          15.5%
2001       102.6%          15.9%
2001       112.4%          16.2%
2001        66.3%          14.1%
2002        58.1%          13.3%
2002        71.5%          15.0%
2002        77.9%          15.7%
2002        80.0%          15.8%
2003        85.6%          17.4%
2003        93.2%          17.9%
2003        92.6%          18.1%
2003        96.6%          19.1%
2004        95.1%          17.9%
2004        86.5%          16.9%
2004        90.5%          17.8%
2004        83.3%          18.9%
2005        82.7%          18.8%
2005        86.2%          19.0%
2005        91.0%          20.4%
2005        91.2%          21.1%
2006       103.9%          24.2%
2006       103.4%          24.8%
2006       105.2%          25.5%
2006       109.6%          25.7%

Source: OCC Bank Derivatives Reports.

Table 7
Impact of Derivatives Held for Purposes other than Trading:
For the 25 U.S. Commercial Banks with the Most Off-Balance
Sheet Derivative Contracts and Credit Losses on Derivatives
(Dollars are in Millions)

Year   Quarter     Total     Total    Net Increase   Net Increase
                    Non-    Trading   to Interest    to Interest
                 interest   Revenue        Income        Expense
                  Income

2001       Q1      20,693     3,788            171           -146
2001       Q2      39,588     6,607            526           -330
2001       Q3      58,931     9,879          1,471           -442
2001       Q4      81,178    12,510          2,928           -909
2002       Q1      20,320     3,112          1,685           -760
2002       Q2      43,861     6,265          3,255         -1,099
2002       Q3      67,164     8,454          4,601         -2,273
2002       Q4      90,446    10,481          6,236         -3,300
2003       Q1      22,787     2,918          2,040           -852
2003       Q2      48,658     5,989          3,793         -1,820
2003       Q3      72,886     8,928          5,846         -2,794
2003       Q4      98,144    10,581          8,643         -3,762
2004       Q1      27,177     3,683          1,534         -1,343
2004       Q2      53,162     6,164          3,416         -2,803
2004       Q3      74,895     7,208          4,799         -3,870
2004       Q4     105,377     9,339          5,774         -4,717

Year   Other Non-        Credit
         interest     Losses On
       Allocations   Derivatives

2001           445             2
2001           418             1
2001         1,112           107
2001         1,355           477
2002           762            67
2002         1,931           102
2002         4,120           160
2002         4,979           236
2003         1,301            25
2003         2,113            52
2003         1,387            83
2003         1,735            96
2004         1,240           120
2004           680            79
2004           988           170
2004         1,657           175

Source: Bank Call Reports, Economic Research
and Data, Federal Reserve Bank of Chicago website.

Table 8
Estimates of Slope Coefficients from Regressing Credit
Losses on Derivative (Credit Losses/Net Income) on
National Amounts of All Derivatives

CRDTLOSS = [alpha] + [[beta].sub.1] (DERIVATIVE/TA)
+ [[epsilon].sub.i]

Measure     Regression    F-Value    Adj R-Sq
            Coefficient

CRDTLOSS       0.0003       43.20     0.0956
(n = 400)     (0.0001)    (0.0001)

* The probabilities are presented in parentheses below
the regression coefficient and the F-value statistic.
CRDTLOSS = Credit Loss on Derivatives/Net Income

Table 9
Estimates of Slope Coefficients from
Regressing Credit Losses on Derivative
(Credit Losses/Net Income) on Notional
Amounts of All Interest Rate Derivatives,
Foreign Exchange Derivatives, Equity
Derivatives and Commodity Derivatives. Each
of these derivatives are separated based upon
if they are held for trading Or non trading
purposes.

CRDTLOSS = [alpha] + [[beta].sub.1] (IRC--trading/Derivative) +
[[beta].sub.2] (IRC-non trading/Derivative) + [[epsilon].sub.1]

CRDTLOSS = [alpha] + [[beta].sub.1] (FEC--trading/Derivative) +
[[beta].sub.2] (FEC-non trading/Derivative) + [[epsilon].sub.1]

CRDTLOSS = [alpha] + [[beta].sub.1]  (EQC--trading/Derivative) +
[[beta].sub.2] (EQC-non trading/Derivative) + [[epsilon].sub.1]

CRDTLOSS = [alpha] + [[beta].sub.1] (COC--trading/Derivative) +
[[beta].sub.2] (COC-non trading/Derivative) +  [[epsilon].sub.1]

Measure     Regression     Measure    Regression
            Coefficient               Coefficient

IRCTRAD          0.0092   IRCNTRAD         0.0007
(n = 400)     (@.0001)    (n = 400)     (0.7255)
FECTRAD         -0.0027   FECNTRAD        -0.0315
(n = 400)     (0.2250)    (n = 400)     (0.3649)
EQCTRAD         -0.0039   EQCNTRAD        -0.1983
(n = 400)     (0.5985)    (n = 400)     (0.2628)
COCTRAD         -0.0152   COCNTRAD       -70.6546
(n = 400)     (0.5635)    (n = 400)     (0.6182)

Measure        F-Value         Adj
                              R-Sq

IRCTRAD           14.68    0.06.42
(n = 400)     (0.0001)
FECTRAD            1.05      0.0003
(n = 400)     (0.3493)
EQCTRAD            0.73     -0.0013
(n = 400)     (0.4803)
COCTRAD            0.28     -0.0036
(n = 400)     (0.7533)

* The probabilities are presented in parentheses below
the regression coefficient and the F-value statistic.

IRC = Interest Rate Contracts

FEC = Foreign Exchange Contracts

EQC = Equity Contracts

COC = Commodity Contracts