Free Notice (Other) - District Court of Federal Claims - federal

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 1 of 59

IN THE UNITED STATES COURT OF FEDERAL CLAIMS ____________ No. 95-758T (Judge Nancy B. Firestone)

National Westminster Bank PLC, Plaintiff v. THE UNITED STATES, Defendant. _____________

EXPERT REPORT OF JAMES READ ON THE INTEREST RATE ISSUE

Prepared for the U.S. Department of Justice

The Brattle Group 44 Brattle Street Cambridge, MA 02138-3736 617.864.7900 voice 617.864.1576 fax March 4, 2005

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 2 of 59

INTRODUCTION 1. My name is James Read. I am a Principal and Director of The Brattle Group. My office is located at 44 Brattle Street in Cambridge, Massachusetts. 2. The Brattle Group is an economic, environmental, and management consulting firm. My consulting practice is focused on valuation, risk management, and capital budgeting. Before joining The Brattle Group I was a Principal and Director of Incentives Research Inc. and prior to that was Director of Financial Consulting at Charles River Associates. I hold a B.A. in Economics from Princeton University and an M.S. (Finance) from the Sloan School of Management at the Massachusetts Institute of Technology. A detailed description of my qualifications, including publications and recent testimony, is included as Appendix A. 3. I have been asked by the United States Department of Justice ("DOJ") to render an opinion as to whether the interest income and interest expense reported by National Westminster Bank PLC ("NatWest") with respect to transactions among its six U.S. branches and affiliates during the tax years 1981 through 1987 reflect arm's-length interest rates. I have also been asked, if I conclude that reported interest income and interest expense do not reflect arm's-length interest rates, to quantify, if possible, the adjustments that would be required to reported income and expenses so that they do reflect arm's-length interest rates. 4. A list of documents I have considered in the course of forming my opinion on this matter is included as Appendix B. 5. For my work in connection with this report and for my time involved with testimony at deposition and trial in this action, The Brattle Group bills at the customary rate of $360 per hour.

SUMMARY AND PRELIMINARY CONCLUSIONS 6. My examination has focused on transactions among NatWest U.S. branches and related parties. 7. NatWest has represented to the DOJ that all interest-bearing related-party transactions of its U.S. branches, other than capital loans, fall into two categories: current account transactions and money market transactions. 8. My understanding is that NatWest is liable under Article 7(2) of the United States­United Kingdom Income Tax Convention (the "U.S.-U.K. Tax Treaty" or the "Treaty") for tax on the profits that a permanent establishment in the United States "might be expected to make if it were a distinct and separate enterprise engaged in the same or similar activities under the same or similar conditions and dealing wholly independently with the enterprise of which it is a permanent establishment."

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 3 of 59

9. The U.S. branches did not pay interest on current accounts with third parties, whereas they did pay interest on current accounts with at least two related parties. If a U.S. branch had been "dealing wholly independently with the enterprise of which it is a permanent establishment", it would not have paid interest on current accounts with related parties. Therefore, it appears to be a violation of the principles set forth in the U.S.-U.K. Tax Treaty for NatWest to deduct interest expense from related-party current accounts on its U.S. income tax returns. I have identified approximately $33 million of interest the U.S. branches paid on current accounts with these two related parties during the tax years in question. 10. So far as I have been able to determine, the U.S. branches did not pay interest on other current accounts with related parties or third parties. However, the U.S. branches computed notional interest on these current account balances, which they deducted from commission fees the accounts owed. This fee structure was equivalent to paying interest, since the higher the interest rate used to compute notional interest, the lower the net commissions the customer would be charged, up to the point that commissions net of credit interest was zero. The fee structure applied to related parties as well as third parties, but the interest rate used to compute notional interest was higher for many related-party current accounts than it was for third-party current accounts. If a U.S. branch had been "dealing wholly independently" with related parties, it would have computed notional interest on all of its current accounts at the same rate it computed notional interest on third-party accounts. Had NatWest reported its income with respect to its U.S. branches in accord with the principles of the U.S.-U.K. Tax Treaty, its income for tax purposes in the years in question would have been greater. 11. The U.S. branches extended overdraft privileges to current account customers and charged interest on overdraft (debit) balances. The interest rate charged to many related parties on debit balances was lower than the rate charged to third parties. As a result, the income reported by NatWest with respect to its U.S. branches in the tax years in question was lower than it would have been had the U.S. branches charged third-party rates on related-party accounts. This too appears to be a violation of the principles set forth in the U.S.-U.K. Tax Treaty. 12. Aside from interest paid by the U.S. branches to the two related parties, I have not been able to calculate how much greater NatWest's income with respect to its U.S. branches' current account transactions would have been had NatWest reported its income from related-party transactions in accord with the principles set forth in the Treaty. The required calculations would be very time consuming. My best estimate at this time is that this would require roughly one man-year of effort. 13. With respect to money market transactions, I have found evidence that during the tax years in question, the interest rates the U.S. branches received from related parties for overnight placings were less than the rates they received from third parties. I have also found evidence that the interest rates the U.S. branches paid to related parties for overnight deposits were greater than the rates they paid to third parties. These findings suggest that NatWest may have overstated its interest expense and understated its interest income with respect to its money market transactions at U.S. branches in the tax years in question. However, the

-2-

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 4 of 59

findings are based on very small samples of transactions and therefore are not sufficient to reach a conclusion. 14. I am unable at this time to render an opinion as to whether the interest income and interest expense reported by NatWest with respect to money market transactions by its U.S. branches with related parties reflect arm's-length interest rates. The impediments to making such a determination are formidable: a. The U.S. branch records that NatWest has made available to the DOJ are not in a computer-readable format. In order to use data from these records, they must first be entered into electronic files, and the data entry process must be audited. Furthermore, some of the relevant records are illegible and others are missing altogether. b. I estimate that the U.S. branches had over 250,000 money market transactions during the tax years in question. Testing the arm's-length nature of interest rates on each relatedparty transaction would therefore be impractical. On the other hand, in order to rely on a subset of the transactions to test the arm's-length nature of the interest rates used by NatWest to compute interest income and expenses for its U.S. branches, one would need to develop and implement an appropriate statistical sampling methodology. Developing a satisfactory statistical sampling methodology would entail substantial time and effort. I have not undertaken to develop a sampling methodology for the U. S. branches' money market transactions in the short time since the close of discovery. 15. Because I have not been able to test whether U.S. branch money market transactions with related parties reflect arm's-length interest rates, I have of course not been in a position to calculate what NatWest's interest income and interest expense would have been with respect to its U.S. branches if the U.S. branches' income and expenses had been maintained in a manner consistent with the U.S.-U.K. Tax Treaty.

BACKGROUND 16. The six U.S. branches of NatWest are New York, New York International Banking Facility ("IBF"), Nassau, Chicago, San Francisco, and Cayman Islands. 17. The principal businesses of the U.S. branches of NatWest during the years in question were wholesale banking--in particular, lending money to large corporations and money market trading. 18. Due to the nature of its business, NatWest's income with respect to its U.S. branches consisted chiefly of interest income and its expenses consisted chiefly of interest expense. The interest rates on borrowing and lending by the U.S. branches are key determinants of NatWest's interest income and interest expense with respect to the U.S. branches for the tax years in question.

-3-

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 5 of 59

Legal Standards 19. My understanding is that the determination of NatWest's taxable income with respect to the operations of its six U.S. branches is governed by Article 7(2) of the 1975 United States­ United Kingdom Income Tax Convention (the "U.S.-U.K. Tax Treaty" or the "Treaty"). My further understanding is that NatWest is liable under Article 7(2) of the Treaty for tax on the profits that a permanent establishment (i.e., a branch) in the United States "might be expected to make if it were a distinct and separate enterprise engaged in the same or similar activities under the same or similar conditions and dealing wholly independently with the enterprise of which it is a permanent establishment." 20. The application of the Treaty to NatWest's branch operations in the United States has been interpreted in two U.S. Court of Federal Claims decisions, one by Judge Turner in July 1999 and the other by Judge Firestone in November 2003. a. Judge Turner held that the profits of a U.S. branch that are subject to tax "should be based on the books of account of such branch maintained as if the branch were a distinct and separate enterprise dealing wholly independently with the remainder of the foreign corporation, provided that the financial records of the branch, especially those reflecting intra-corporate lending transactions, are subject to adjustment as may be necessary for imputation of adequate capital to the branch and to insure use of market rates in computing interest expense." 44 Fed. Cl. 120, 128 (1999). b. Judge Firestone held that the Treaty "requires the government to use the properly maintained books of the branch to determine each element affecting the profits ... and may only allot additional capital to the branch, if, in fact, capital allotted to the branch was not properly noted on its books as `capital'." 58 Fed. Cl. 491, 492 (2003). Although the books may be adjusted if they do not properly identify capital infusions that are "in fact" made, Treaty Article 7 does not permit adjustments to reflect "hypothetical" infusions "based upon banking and market requirements that do not apply to the branch." 58 Fed. Cl. at 498. c. Judge Firestone continued and upheld NatWest's view that branch books may be adjusted in three circumstances: "(1) an interest expense was deducted for advances to the branch that were not used in the ordinary course of its banking business; (2) an interest expense was deducted on amounts designated as capital on its books or on amounts that were in fact allotted to it for capital purposes, such as funding capital infrastructure; and (3) interest paid on inter-branch borrowings was not at arms' length." 58 Fed. Cl. at 505. d. Finally, Judge Firestone ruled that the historic position of the United Kingdom, reflected in the Inland Revenue Manual, "will provide an adequate basis for the parties to resolve the `capital' accounts of the branch and the interest deduction allowed by the branch." 58 Fed. Cl. at 506. 21. As a result, the assessment of interest expense for tax purposes is divided into two issues: 1) How much capital did the branches have? 2) What were arm's-length interest rates on its

-4-

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 6 of 59

interest bearing obligations? This evaluation of interest expense using two segregated issues is problematic, because the interest rates a firm pays depend on its capital. The two issues are not independent. For example, the more capital the firm has outstanding, the lower the interest rate it will have to pay, other things being equal. Nevertheless, I am constrained by the Court's legal opinions to confine my attention to the question of whether the interest rates used to compute the interest income and interest expense reported by NatWest with respect to its U.S. branches in the tax years at issue are consistent with arm's-length transactions. I am aware that the U.S. branches were not separately incorporated and thus that the arm's-length rates at which the branches could borrow reflected the capital and credit quality of NatWest as a whole.

Determinants of Arm's-Length Interest Rates 22. An "arm's-length" transaction is a transaction between unrelated parties, each of whom is acting in its own self interest, neither being under any compulsion to deal with the other. The interest rate on an interest bearing debt obligation, therefore, can be characterized as an arm's-length interest rate if the rate is that which would prevail between a borrower and lender who are unrelated and are acting in their respective self interests. In the balance of my report, I will use the term "loan" as shorthand for "interest bearing debt obligation". 23. Interest rates prevailing in competitive money and capital markets are arm's-length interest rates. 24. Many factors can influence the arm's-length interest rate on a loan, including (but not necessarily limited to) the following: a. The currency in which the loan is denominated. For example, interest rates on U.S. dollar-denominated loans will in general differ from interest rates on U.K. sterlingdenominated but otherwise comparable loans. b. The origination date of the loan. Market rates, especially on short-term loans, vary from day to day and even within a day. c. The delivery date of the loan. The time funds are made available to the borrower often lags the origination date of a loan. d. Maturity date of loan. The interest rate on a loan extended for one month will differ from the interest rate on a loan extended for one year, for example. e. Repayment schedule. The interest rates on two otherwise identical loans will differ if one loan will be repaid in installments whereas the other will be repaid in a lump sum at maturity. 25. In addition to these timing attributes, arm's-length interest rates depend on the prospects that the borrower will default and, in the event of default, how much the lender will recover.

-5-

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 7 of 59

a. The credit quality of the borrower is closely related to the amount and quality of assets the borrower holds and the amount and quality of the other liabilities the borrower has outstanding. b. The seniority of the debt obligation. The term "seniority" refers to the rank of a debt obligation in the priority of claims. Other things being the same, the interest rate on a senior debt issue will be lower than the rate on a subordinated debt issue. c. The security, if any, pledged for the loan. Some loans are secured by specified assets (e.g., real estate for mortgage debt). The interest rate on a secured loan is usually lower than the rate on an otherwise equivalent unsecured loan to the same borrower. d. Most loans are extended at interest rates that are fixed at origination. However, in some cases the rate is adjusted from time to time according to a formula. The formula is another factor that influences competitive rates at the inception and over the life of a loan. The interest rate might be determined based on a specified index (one-month LIBOR plus a fixed spread) and adjusted with a specified frequency (e.g., monthly). 26. Other factors, too, can affect arm's-length interest rates. Among these are, e.g., the borrower's rights to prepay the loan and associated prepayment penalties, if any; and the lender's rights, if any, to demand payment of principal and interest prior to the maturity of the loan. Interest rates may be affected by other terms and conditions of a loan, such as requirements that certain financial ratios be maintained. 27. Finally, competitive interest rates are affected by costs related to marketing, credit evaluation, negotiation, and administering loans. A lender needs to charge an all-in interest rate (or interest rate plus additional fees) that cover(s) its costs of doing business. These related costs might be reflected, e.g., in interest rates that differ by size of loan.

ANALYSIS OF NATWEST U.S. BRANCH TRANSACTIONS 28. My efforts have been focused on assessing whether the interest rates on loans among the U.S. branches and related parties were at arm's length. I distinguished related parties from third parties using the A220 reports. Specifically, I treated counterparties identified on the A220 reports as "NWB/IWB Group"--and other counterparties identifiable as such but for which no A220 entry was found--as U.S. branch related parties. I treated all other counterparties as third parties. 29. My understanding is that Plaintiff has represented that all relevant interest-bearing related party transactions of the U. S. branches during the tax years at issue are reported in either the A220 or M262 reports.1 The A220 reports describe activity in "current accounts". Current accounts consist of demand deposits, i.e., funds that customers were entitled to withdraw with no notice. The M262 reports describe "money market" transactions. Money market
1

Plaintiff's response to Interrogatory No. 8.16.

-6-

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 8 of 59

transactions are short-term, usually with maturities of less than one year. Money market transactions amongst the U.S. branches and related parties included "placings" (loans to banks or other financial companies) and deposits. Most of these were very short term, often only one day.

NatWest Records of the Transactions of the U.S. Branches 30. Records made available by NatWest include A220 reports, M262 (or M070) reports, B118 reports, and B126 reports for each of the six U.S. branches. A220 reports are monthly records of current account transactions. M262 reports are daily records of money market transactions. (M070 reports contain the same information as M262 reports but are organized differently.) B118 reports are profit and loss (P/L) reports. B126 reports are balance sheets. Not all of these documents are legible. Some of the reports prepared during the tax years at issue are missing. 31. NatWest has also made available some manually prepared P/L reports and balance sheets, which were prepared on a quarterly basis. My understanding is that the manually prepared reports corrected some errors in the computer-generated reports, but that these manually prepared reports were prepared on a short deadline and would have been corrected again at the head office of NatWest.2 32. The A220, M262, B118, and B126 reports are available in hard copy or microfiche only. They are not available in a computer-readable format. Therefore, data processing and auditing are required before analyzing data contained in these reports. 33. A220 reports are monthly summaries of current account transactions prepared at each of the six U. S. branches. They show for each customer account daily debits and credits, balance, interest rates, interest credited or debited (in the case of an overdraft), and commissions. 34. M262 reports are daily summaries of money market transactions. They identify the counterparty, type of instrument, currency, amount of transaction, interest rate, maturity date, and value date of new, outstanding, and matured positions. Lending included loans to banks ("placings"), loans to non-bank corporate customers, and sales of federal funds. Borrowing included certificates of deposit, commercial paper, purchases of federal funds, and customer deposits. 35. The P/L reports were not structured to break out related party income and expense or balances by type of transaction. As a result, one cannot tell how much of the reported income and expenses for a given related party, or all related parties, is from current account transactions and how much from money market transactions, or within money market transactions how much is from different types of transactions. It also is not possible to confirm NatWest's representation that all of the related party interest income and expense (except interest expense on capital loans) is attributable to current accounts and money market transactions without re-computing interest from balances and interest rates.
2

Deposition of Harrylall Samaroo, January 23, 2002, p. 82.

-7-

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 9 of 59

36. In addition to data available in the aforementioned NatWest reports, evidence about arm'slength interest rates is available in the form of historical money market interest rates, including for example London Interbank Offer Rates (LIBOR) and federal funds rates. These data are available for each business day.

Current Account Transactions 37. During the tax years in question, the U.S. branches 1) paid interest on certain related-party current accounts but did not pay interest on third-party current accounts; 2) computed notional interest expense on the credit balances of certain other related-party current accounts using a higher interest rate than the one used to compute interest expense on third-party current accounts; and 3) computed interest income on the debit balances of certain relatedparty current accounts using a lower interest rate than the one used to compute interest on third-party current accounts. Income and expenses reported by NatWest with respect to its U.S. branches reflect these practices. This appears to be a violation of the arm's length standard and the U.S.-U.K. Tax Treaty, which states that the U.S. branch of a U.K. bank shall be assessed taxes on profits the U.S. branch "might be expected to make if it were a distinct and separate enterprise engaged in the same or similar activities under the same or similar conditions and dealing wholly independently with the enterprise of which it is a permanent establishment." NatWest's profits would have been higher had it computed interest for all related-party current accounts using the same rates it used to compute interest for third-party current accounts. Interest Paid to NWB London and IWB London 38. So far as I have been able to determine, the U.S. branches did not pay interest on current accounts with third parties. In contrast, the U.S. branches did pay interest on at least two related-party current accounts: NWB London Branch and IWB London Branch. 39. According to NatWest's deposition testimony, presented through Mr. Roger Walmsley, the NatWest New York Branch paid interest on current accounts with NWB London Branch.3 Through my review of P/L reports for the tax years in question and an interest rate analysis prepared by counsel for the Plaintiff for this litigation, I have identified approximately $30 million of interest payments by the New York Branch to NWB London. I identified interest payments for the 1983 through 1987 tax years by examining the corresponding year-end hand-written P/L reports for the New York Branch. Interest payments for the 1981 and 1982 tax years were obtained from Plaintiff's interest rate analysis submitted on June 8, 2001.4 The payments are shown in Exhibit A. 40. I have found evidence that the U.S. branches of NatWest also paid interest on current accounts with IWB London Branch. Specifically, by letter written in April 1981, IWB
3 4

Deposition of Roger Walmsley, February 1, 2005, pp. 141-42. Plaintiff's interest rate analysis was attached to a letter from Michael C. Moetell to Steven I. Frahm dated June 8, 2001.

-8-

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 10 of 59

London Branch informed NWB New York Branch that IWB London believed that it, like NWB London, should receive interest on current accounts.5 This position was supported in a letter from NWB New York to IWB London written in the same month.6 It is confirmed in a December 1983 note describing particulars of the interest rate calculations.7 I have been able to identify over $2 million in interest payments on current accounts paid to IWB London Branch in the 1983 to 1987 tax years by examining the New York Branch year-end P/L reports for the corresponding years. These interest payments too are reported in Exhibit A. I have not identified interest payments to IWB London in 1981 and 1982 because the P/L reports prepared for those years are for the New York and New York IBF Branches combined. Plaintiff's interest rate analysis of June 8, 2001 did not identify interest payments to IWB London. 41. The total interest payments I have identified from the U.S. branches to NWB London and IWB London are shown in Exhibit A. The total is approximately $33 million for the tax years at issue. However, given that Mr. Walmsley was unaware that the U.S. branches paid interest on current accounts with IWB London, I cannot conclude with confidence that the U.S. branches did not also pay interest on current accounts with other related parties. Notional Interest Deducted from Commission Fees 42. So far as I have been able to determine, the U.S. branches of NatWest did not pay interest on current accounts with related parties, other than NWB London and IWB London. However, the U.S. branches had in place a fee structure for other current accounts that was under certain circumstances equivalent to paying interest. Specifically, the U.S. branches established a fee ("commission") for current account transactions, but the charge for commissions each month was offset by a notional interest on current account balances.8 If the commissions on a current account exceeded the notional interest on credit balances in a given month, then the U.S. branches charged the commissions minus the notional interest. If the notional interest exceeded commissions, then the U.S. branches charged the customer nothing. 43. This fee structure applied both to related-party customers and third-party customers. However, the interest rates that the U.S. branches used to compute notional interest on many of its related-party accounts were higher than the interest rates it used to compute notional interest on third-party accounts. Specifically, roughly half of the accounts with related customers were credited interest at the federal funds rate, whereas the other related-party accounts and all third-party accounts were credited interest at a rate substantially less than the federal funds rate.9 The consequence of this policy of computing notional interest on certain
5

6 7 8

9

Letter from A. J. Welch, Chief Manager, IWB London Branch to J. D. S. Crossman, Manager, NWB New York Branch, dated April 3, 1981 (NWP 0000353-354). Letter to A. J. Welch from J. D. S. Crossman dated April 20, 1981 (NWP0000357-359). Note from B. T. Sharpe to H. P. Kilfoyle dated December 15, 1983 (NWP 0000360). This fee structure is described by Roger Walmsley in his deposition testimony of February 1, 2005 at pp. 149-52. The New York Branch A220 report for December 31, 1986 shows that notional interest on current account credit balances was computed using the federal funds rate on roughly half of the related-party accounts.

-9-

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 11 of 59

related-party current accounts with a higher interest rate than third-party accounts is that the net commissions the U.S. branches charged these related-party current accounts were lower than the net commissions that would have been charged to a third party under identical circumstances. 44. The crediting of certain related-party accounts with interest at a higher rate than the one used to credit third-party accounts reduced the profits NatWest reported for income tax purposes below what it would have reported had it computed interest instead at the third-party rates. This appears to be a violation of the arm's-length principles of the U.S.-U.K. Tax Treaty. Interest on Overdraft Balances 45. The U.S. branches extended overdraft privileges to current account customers. I have found that the U.S. branches charged interest on overdraft (debit) balances and that the interest rates charged to many related-party accounts were less than the rates charged to third-party accounts. Specifically, about half of the related-party current accounts were charged interest on overdraft balances at the federal funds rates, whereas third parties were charged a higher rate of interest. Had the business relationship between the U.S. branches and the related parties been independent, the related parties would have paid interest at the same, higher rates as third parties. Therefore, the U.S. branches' books underreported the profits that would be expected under the principles set forth in Article 7(2) of the U.S.-U.K. Tax Treaty, that is, had interest income been reported using third-party interest rates. Adjusting NatWest's Profits to Reflect Third-Party Interest Rates 46. To see how the U.S. branches' use of higher interest rates to compute notional interest expense on credit balances and lower interest rates to compute interest income on debit balances of certain related parties vis a vis third parties, consider an example taken from the current account records of the New York Branch. Exhibit B shows NatWest Singapore's current account balances in December 1986. NatWest Singapore had a credit balance on some days in December and a debit balance on other days, as shown in Column 1. Column 3 shows the federal funds rate that NatWest applied each day to compute interest on these balances and Column 4 shows the resulting interest. The bottom of Column 4 shows the net credit interest of $460, gross commissions of $5,388, and net commissions of ($4,928) that result. (Negative net commissions represent a charge to the account.) Columns 5 through 7 re-compute notional interest based on the credit and debit interest rates the New York Branch applied to third-party current accounts that month. The bottom of Column 7 shows the consequences. Net credit interest of ($518), i.e., an interest charge to NWB Singapore, and thus net commissions of ($5,906) are affected.10 In other words, in the case of NatWest Singapore in December 1986, applying third-party interest rates instead of the more favorable related-party rates would result in an increase in profits of $977. 47. Exhibit B also conveys something of the level of effort that would be involved to calculate the amount by which all the branches understated their profits with respect to their related party current account customers as a result of using a higher interest rate to compute notional
10

Here and in Exhibit B I follow the accounting convention of stating negative amounts in parentheses.

- 10 -

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 12 of 59

interest on credit balances and a lower rate to compute interest on debit balances of certain related parties. One would have to 1) identify the related party current accounts that received favorable interest rate treatment; and 2) for each of these current accounts in each month during the seven tax years in question, re-compute credit and debit interest at the third-party rates and then re-compute net commissions. 48. In December 1986, there were 56 related-party current accounts at the New York Branch, on 29 of which credit and debit interest were computed at the federal funds rate rather than the third-party rate. There are 84 months in the seven tax years at issue. If there were 29 related-party current accounts that needed adjustments over the entire seven-year period at issue--there may have been more or less--then there would be roughly 2,400 monthly accounts that would have to be re-computed. Based in part on the effort involved in creating Exhibit B, I estimate that it would require roughly a man-year of effort to calculate the adjustments to the relevant related-party current accounts. Of course, given the limited time available since the close of discovery, it has been impossible to perform all of the required calculations.

Money Market Transactions 49. In principle, one could perform two types of tests of whether interest rates on interest bearing money market obligations between the U.S. branches and related parties reflected arm'slength interest rates: 1) compare interest rates on related-party transactions with interest rates observed in money markets; 2) compare interest rates on related-party transactions with interest rates on third-party transactions of the U.S. branches. 50. To test whether the interest rate on an interest bearing obligation with a related party was an arm's-length rate, one would need to identify comparable third-party obligations, e.g., obligations of the same type and credit quality that were originated at the same time and extended for the same term. 51. The U.S. branches executed a very large number of money market transactions during the tax years in question. Based on my examination of M262 reports for a limited number of days, I would estimate that there were approximately 150 new money market transactions on average each business day during the seven tax years at issue. That works out to over 35,000 new transactions a year and over 250,000 transactions over seven years. Therefore, testing whether the interest rate on each transaction was an arm's-length rate is not practical. In order to perform a satisfactory test of whether the interest rates on loans among the U.S. branches and related parties were consistent with arm's-length interest rates, one would have to develop and implement a statistical sampling methodology. 52. Developing a statistical sampling methodology suited to the problem at hand would itself entail considerable time and effort. To begin, one would need to identify all of the attributes that render the money market transactions at issue "comparable". It is clear that transaction type (e.g., deposits), transaction date, and term (e.g., overnight) are important, but other attributes, such as whether transaction involved a broker, may be important too. For each

- 11 -

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 13 of 59

category of transactions, the sampling methodology would need to take into account such factors as the sizes of the populations of related-party and third-party transactions and the variances of the underlying interest rates. I have not undertaken to develop a sampling methodology for the U. S. branches' money market transactions in the short time since the close of discovery. NatWest's Test of Related Party Interest Rates 53. DOJ has provided me with an analysis prepared by NatWest's counsel, which purports to show that on a particular day--June 2, 1986--the interest rates the U.S. branches' paid on deposits (borrowings) from related parties were essentially the same as the interest rates the U.S. branches paid on deposits from third parties, as well as published money market rates.11 NatWest provided results by maturity, including overnight, three-week, one-month, twomonth, and three-month deposits. The average rate the U.S. branches paid on overnight deposits from third parties was 6.90 percent whereas the average rate it paid on overnight deposits from related parties was 6.85 percent, i.e., the U.S. branches paid related parties about 5 basis points less than third parties.12 54. I performed several interest rate calculations to evaluate the analysis provided for June 2, 1986. I worked with overnight transactions because they represent the bulk of the U.S. branches' money market activity. First, I computed the average interest rates the U.S. branches received from related and third parties on overnight placings (lending). I found that the average interest rate on U.S. branches' overnight placings was 6.959 percent for related parties and 6.966 percent for third parties, i.e., the U.S. branches received about 1 basis point less from related parties than they did from third parties. Thus, based on the June 2, 1986 records, the U.S. branches appear to have paid to and received from related parties and third parties interest at comparable rates for overnight money market transactions. 55. I then applied this methodology to four additional days during the tax years in question. Specifically, I identified all overnight deposits with (borrowing) and placings from (lending) the U.S. branches on December 15, 1983, December 31, 1984, January 17, 1985, and December 31, 1987; determined whether each transaction was from or to a related party or a third party; and then computed the weighted average interest rates for related parties and for third parties. The results for the four additional days, which are summarized in Exhibit C, are very different than for June 2, 1986--the date NatWest selected. a. On December 15, 1983, the U.S. branches paid on average a rate of 9.794 percent on related party deposits, whereas they paid 9.697 percent on third party deposits, i.e., about 10 basis points more interest (in contrast to 5 basis points less interest on June 2, 1986) to related parties than to third parties. On December 31, 1984, the U.S. branches paid on average about 14 basis points more interest to related parties than to third parties. On January 17, 1985 they paid related parties 9 basis points more than third parties. And on December 31, 1987 the U.S. branches paid related parties about 12 basis points more than
11 12

Schedule 1 attached to a letter from M. Carr Ferguson to Steven I. Frahm, September 19, 2000. A basis point is one one-hundredth of a percentage point. In other words, one percentage point is equal to 100 basis points.

- 12 -

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 14 of 59

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 15 of 59

Exhibit A NetWest New York Branch's Interest Expenses Paid on Current Accounts ($) NWB/IWB Unit NWB London Overseas IWB London Total 1981 5,161,815 5,161,815 1982 3,439,695 3,439,695 1983 2,729,147 123,335 2,852,482 1984 4,037,739 227,782 4,265,521 1985 3,251,504 1,370,596 4,622,100 1986 3,189,170 124,928 3,314,098 1987 8,772,930 328,316 9,101,245 Total 30,582,000 2,174,956 32,756,956

Sources and Notes: (1) The amounts for 1983 to 1987 are based on NWB New York branch's year-end manually prepared P/L reports. (2) The interest paid to NWB London Overseas for 1981 and 1982 were obtained from Plaintiff's interest rate analysis submitted on June 8, 2001. [Attachments to letter dated June 8, 2001 from: Michael C. Moetell of Shaw Pittman, to: Steven I. Frahm, Esq. (DOJ) Enclosures include 'Plaintiff's FRE 1006 Summaries With Respect To Interbranch Interest Expense 1981, 1982, 1983, 1984, 1985, and 1987 Tax Years.' Values obtained from: (1) Section 2(A) '1981 New York and IBF Borrowings From London Overseas'; (2) Attachment 1981-A; (3) Section 3(A) '1982 New York and IBF Branch Borrowing From London Overseas'; (4) Attachment 1982-A.] (3) The interest expenses paid to IWB London for 1981 and 1982 are assumed to be zero.

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 16 of 59

Exhibit B Analysis of New York Branch's A220 Interest Calculation: Related Party versus Third Party NatWest Singapore -- December 1986 True Balance ($) [1] 330,473 (3,070) (641,045) (94,810) 21,696 (45,481) 348,666 258,533 (42,957) (84,687) 904,303 (2,257,695) 730,132 1,107,752 1,192,083 269,375 (405,867) (1,168,634) (39,558) (40,547) 98,483 15,628 49,183 No. of Days [2] 1 1 1 1 1 3 1 1 1 1 3 1 1 1 1 3 1 1 2 3 1 1 0 Related Party Rate Interest Rate Interest (%) ($) [3] [4] 6.2500 6.2500 6.6250 6.7500 6.2500 6.0000 5.8750 5.9375 5.7500 5.8750 6.0000 6.5000 6.3750 7.6250 6.5000 6.1250 6.2500 6.3125 6.1250 6.2500 7.8125 14.000 8.0000 57 (1) (118) (18) 4 (23) 57 43 (7) (14) 452 (408) 129 235 215 137 (70) (205) (13) (21) 21 6 460 1,796 3 5,388 Third Party Rates Debit Rate Credit Rate Interest (%) (%) ($) [5] [6] [7] 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 9.0000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 3.5000 32 (1) (160) (24) 2 (34) 34 25 (11) (21) 264 (564) 71 108 116 79 (101) (292) (20) (30) 10 2 (518) 1,796 3 5,388 (5,906) (5,906) 977 Difference ($) [8] 25 0 42 6 2 11 23 18 4 7 188 157 58 127 99 59 31 87 6 9 12 5 977

Date

11/30/1986 12/1/1986 12/2/1986 12/3/1986 12/4/1986 12/5/1986 12/8/1986 12/9/1986 12/10/1986 12/11/1986 12/12/1986 12/15/1986 12/16/1986 12/17/1986 12/18/1986 12/19/1986 12/22/1986 12/23/1986 12/24/1986 12/26/1986 12/29/1986 12/30/1986 12/31/1986

Total Net Credit Interest ($) Number of Entries Commission per entry ($) Commisions ($)

0

Net Credit Interest minus Commissions ($) (4,928) Net Charge to customer ($, less of Net Credit Interest minus Commissions and 0) (4,928) Sources and Notes: [1]-[3]: New York Branch A220 report for December 1986 [INT537-001097-99]. [4] = [1] x ([2]/360) x ([3]/100). [5]-[6]: New York Branch A220 report for December 1986 [INT537-000106-110]. [7]: If [1] < 0: [7] = [1] x ([2]/360) x ([5]/100); - or If [1] >= 0: [7] = [1] x ([2]/360) x ([6]/100). [8] = [4] - [7].

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 17 of 59

Exhibit C Analysis of NatWest U.S. Branches' M262 Average Interest Rates for New Transactions: Related Party versus Third Party Overnight Deposits Third Party Difference 9.697% 8.798% 8.094% 6.90% 6.400% 0.096% 0.137% 0.090% -0.050% 0.123% Overnight Placements Third Party Difference 9.921% 10.000% 6.966% 6.991% -0.167% -0.888% -0.007% -0.235%

Overnight Fed Funds 12/15/1983 12/31/1984 1/17/1985 6/2/1986 12/31/1987 9.940% 8.740% 8.270% 7.050% 6.890%

Related Party 9.794% 8.935% 8.184% 6.85% 6.523%

Related Party 9.754% 9.112% 8.190% 6.959% 6.756%

Source: NatWest M262 reports for all branches available for the dates specified. Notes: [1]: Boxed cells indicate analysis prepared by plaintiff. [3]: Transactions identified as "new" but initiated prior to date of M262 report, and transactions with zero interest are excluded from this analysis.

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 18 of 59

APPENDIX A QUALIFICATIONS OF JAMES READ

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 19 of 59

JAMES A. READ, JR.

Principal

James Read is an expert in asset valuation, risk management and capital budgeting. His consulting practice focuses on the electric power and natural gas industries. Mr. Read makes extensive use of option pricing methods to calculate the value and risk of real assets and non-financial contracts as well as derivatives. He has given many presentations on this and related topics at industry conferences and professional seminars. Prior to joining The Brattle Group, Mr. Read was a Principal with Incentives Research Inc., and before that was Director of Financial Consulting with Charles River Associates. He holds a B.A. in economics from Princeton University and an M.S. (concentration in finance) from the Sloan School of Management at Massachusetts Institute of Technology.

CONSULTING EXPERIENCE $ Worked with an electric utility to develop custom methodology for measuring the risk of its supply portfolio and evaluating candidate power contracts. Advised several clients in the electric utility industry in connection with the design, pricing and risk management of "provider of last resort" and similar retail transition services created as part of industry restructuring. Analyzed the impact of credit risk on the pricing of energy contracts. Analysis was performed in the context of a regulatory review of energy procurement decisions. Provided legal counsel with economic analysis of structured finance transactions in a litigation matter involving companies in the energy and financial services industries. Advised power companies in connection with the valuation of generation assets. Projects entailed development and use of options-based valuation tools as well as development of long-term forward price curves and volatility term structures. Developed derivatives-based methodology for estimating the cost of capital for investments in merchant power generation. Developed methodology for allocating capital requirements to lines of business in multiple-line insurance companies.

$

$

$

$

$

$

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 20 of 59

$

Prepared testimony on behalf of New York State Electric & Gas in connection with its Price Protection Program (New York Public Service Commission, Case 01-E-0359). Testimony addressed risk and cost of capital for an electric utility that has sold its generation but continues to have retail service obligations. For the Electric Power Research Institute, directed development of the Energy Book System (EBS), a suite of software tools for valuation and management of energy resources. This includes tools for portfolio risk management, valuation and pricing of wholesale contracts and retail service agreements, and valuation and management of thermal and nuclear generation (with project evaluation and capital budgeting capabilities). The EBS exploits concepts and methods of derivative asset valuation to facilitate integration of valuation, risk analysis and asset management. Developed and taught professional training courses on the application of derivatives methods for understanding the value and risk of commodity contracts and physical assets. EPRI-sponsored courses currently on offer include Value & Risk in Energy Markets, Applied Valuation & Risk Management, and Generation Asset Valuation. Used option pricing methods to estimate the premium over cost required to compensate investors for the long-term irreversibility of investments in railroad assets. Analysis was used in a revenue adequacy proceeding before the Surface Transportation Board. Designed methodology for pricing a new product in the natural gas pipeline industry that would allow shippers to purchase options on pipeline capacity expansion. Developed valuation algorithm for a novel retail electric service that allows supplier to buy back electric energy when wholesale market conditions are tight. Advised the developer of a new energy storage technology regarding marketing to the power industry. Developed a valuation and decision-making model of nuclear power plants that explicitly incorporates flexibility to shut down prior to expiration of operating license. Advised Tennessee Valley Authority and other companies in connection with their evaluations of bids received in response to power purchase RFPs. These "paper" power plants represented an alternative to conventional "bricks and mortar" capacity expansion. Engagements involved development of models for evaluating option-type Read A-2

$

$

$

$

$

$

$

$

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 21 of 59

JAMES A. READ, JR. Principal bids, formulation of bid selection strategy, and development of forward price and volatility curves. $ For EPRI, prepared a report that describes how the theory and methods of option pricing can be exploited to help evaluate investment projects and contracts. Helped an electric utility renegotiate several long-term power purchase agreements. The key aspect of this engagement was the valuation of option terms. Mr. Read is a principal author of the Utility Capital Budgeting Notebook, which integrates previous EPRI studies in finance and project evaluation into a single text. Part of a project team that advised an investor-owned utility that wanted to spin off an electric power station as a stand-alone generating company. A key issue in this case is the risk sharing defined by a novel power purchase contract. Helped an electric utility with surplus generating capacity identify opportunities for selling electric power outside its traditional service territory. Also evaluated the threat posed by competitive power producers. For an interstate natural gas pipeline, was part of a team working to implement a form of the "gas inventory charge" (GIC) propounded by the FERC in Order No. 500. This project involved estimating the costs associated with maintaining a reliable supply of gas and evaluating alternative pricing mechanisms for cost recovery. For a client in the natural gas industry, was part of a team working to identify and evaluate new products for the gas market. Helped a client in the natural gas industry evaluate alternative capital recovery schemes for new and existing pipeline assets. Mr. Read prepared an affidavit on behalf of a natural gas pipeline company that challenged a FERC order on the allowed rate of return. The principal issue in this case concerned the use of ex post data to infer the cost of capital. Advised an electric utility engaged in a major capacity planning decision in which the central issue was whether to delay or abandon a nuclear construction program. Read A-3

3

$

$

$

$

$

$

$

$

$

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 22 of 59

JAMES A. READ, JR. Principal $ In a study for EPRI, identified a conceptual problem that arises in the application of the revenue requirements method when utility ratemaking procedures are inflexible. The study pointed out that there is feedback between demand and rates, which may undermine the logic for cost-based evaluation of projects. In another study for EPRI, developed a rigorous procedure for calculating the cost of holding fuel and other commodity inventories. The procedure exploits information in commodity futures and money markets. Part of a study team that developed new methods for evaluating natural gas contracts. These procedures, which are derived from the theory of option pricing, explicitly incorporate uncertainty about future spot prices. They have been applied, for example, to estimate the cost of standby rights on gas pipelines. In a project for a client in the petroleum industry, Mr. Read worked with an academic expert in finance to develop testimony concerning the value of expropriated oil fields in the Persian Gulf. This testimony was submitted to the Iran-U.S. Claims Tribunal in the Hague. In a study for EPRI, developed a methodology for selecting project-specific discount rates. The methodology is based on the idea that cash flows can be partitioned into risk classes, and hence that the value of an investment project can be found by adding up the values of the parts. In another study for EPRI, prepared an exposition of the revenue requirements method. Among other findings, the report concluded that the appropriate riskadjusted discount rate for calculating the present value of revenue requirements may differ from the discount rate used to calculate net present value. It also identified the logical errors involved in the use of customer discount rates for calculating the present value of revenue requirements. For the U.S. Department of Energy, Mr. Read was part of a team studying the problem of earnings attrition in the electric utility industry. The project used case studies, statistical analysis, and financial simulations to determine the sources and extent of attrition. In a project for the California Public Utilities Commission, Mr. Read was a key member of a team to prepare a guide to alternative methods of estimating the cost of Read A-4

4

$

$

$

$

$

$

$

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 23 of 59

JAMES A. READ, JR. Principal equity capital. All of the major methods now in use were evaluated, including the discounted cash flow (DCF) model, the comparable earnings test, the capital asset pricing model (CAPM), and market to book method. Each method was evaluated on the basis of theoretical, empirical, and practical criteria. $ For the Edison Electric Institute (EEI), prepared a discussion paper of electric utility shocks and power plant phase-ins. The paper identified the sources of rate shocks, outlined the major proposals for plant phase-ins, and developed criteria for choosing among alternative phase-in mechanisms. Project manager in a study for the U.S. Department of Energy to assess the cost of capital for public and private investments in petroleum stockpiles. The objective of the research was to assess the investment value of private oil stocks and thereby determine the effectiveness of government policies aimed at stimulating private stockpile formation. A major focus of this research was the unusual risk profile of petroleum prices and its implications for investment value. For a client in the telecommunications industry, investigated the effects of diversification into a high-risk line of business on the cost of capital for a public utility. The analysis distinguished the effects on the cost of debt, the cost of equity, and the overall cost of capital. It also assessed how the proposed capital structure affected the conclusions. In a study for EPRI, was part of study team that developed theoretical and empirical analysis of a bias that exists in conventional measures of market risk when applied to the shares of public utility companies. It explained why a bias is likely to arise, provided empirical confirmation of the bias, and devised corrected measures of market risk. Advised legal counsel on issues related to the acquisition of a public utility by a closely held corporation. Evaluated the implications of a proposed financing plan on the ability to attract capital for future utility investments. For the U.S. Department of Energy, part of a team analyzing markets for natural gas imported from Canada. His work focused on the availability and cost of pipeline capacity among domestics markets, domestic sources, and Canadian sources.

5

$

$

$

$

$

Read A-5

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 24 of 59

JAMES A. READ, JR. Principal $ Prepared a statistical analysis of gold and silver markets for a client in the minerals industry. This work included an analysis of the time series properties of gold and silver prices, an analysis of the gold-silver price ratio, and analyses of the correlation between gold and silver price movements and the stock market, credit markets, foreign exchange and other macroeconomic factors. In a project for a major motor carrier, Mr. Read was part of a team exploring a new business concept in the truckload freight market. This work included concept development, market research, and an implementation plan. For a private client engaged in a rate of return proceeding, Mr. Read evaluated the impact of deregulation on business risk. His work considered two aspects of the problem: the theoretical basis for a linkage between regulation and business risk, and an empirical investigation of the investment risk of industries recently subject to deregulation. In a major antitrust case, Mr. Read worked with an academic expert in finance to critique the damages study submitted by the plaintiff. Among other issues, the critique identified a fundamental conceptual error: the damages were calculated using a mixture of foresight and hindsight. In the Penn Central bankruptcy, Mr. Read worked on the valuation of the Intercity Freight Lines. He was responsible for extensive analysis of U.S. capital market data to determine the cost of capital in the railroad industry. In addition, he worked with academic experts in finance theory to evaluate alternative approaches to valuation. Project manager for a study to analyze the competitive impact of differences in national health and safety regulations. The OSHA standard for occupational exposure to lead was used as a case study. Cost differentials arising directly from workplace standards and indirectly from government tax and investment policy were assessed using a financial analysis framework. For the U.S. Department of the Interior, prepared a report describing methodologies for using corporate financial information to assess the state of metals and minerals industries. The report focused on publicly available data, but it also offered recommendations for industries dominated by privately held companies.

6

$

$

$

$

$

$

Read A-6

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 25 of 59

JAMES A. READ, JR. Principal $ For a private client involved in antitrust litigation, Mr. Read was responsible for critiquing the testimony of an opposing witness who had used corporate financial data to draw inferences about risk and profitability. The critique identified two major errors, one computational and one conceptual, which, when corrected, reversed the conclusions of the original analysis. Helped prepare testimony on rates of compensation for owners of private rail cars. This testimony addressed the cost of capital for rail cars, the concept of "vintage year" interest, and the appropriate treatment of certain tax provisions. It also addressed whether compensation should be based on the original cost or replacement cost of rail cars. For a gas distribution company, developed a model for selecting a portfolio of natural gas contracts in a way that minimized costs. Contract size, contract price, and "take or pay" provisions were incorporated in the model. Conducted a study for the U.S. Department of Transportation to compare the response of the for-hire trucking industry during the recessions of 1980 and 1974-75. This research considered such factors as financial performance, prices, output and entry. The results were used as background for congressional oversight hearings on trucking deregulation.

7

$

$

$

OTHER EXPERIENCE Financial Analyst, Corporate Financial Staff, General Motors Corporation. Mr. Read worked in forward product programs and corporate transfer pricing. Staff Economist, Mail Classification Research Division, United States Postal Service. Mr. Read's responsibilities included writing statements of work, technical evaluation of analytical study proposals, and directing contractors in the Postal Service=s Long Range Classification Research Program. Staff Economist, Office of Rates, United States Postal Service. Mr. Read was engaged in the preparation of testimony filed with the Postal Rate Commission in support of requests for changes in rates. His responsibilities included cost analysis, revenue forecasting, econometric analysis of postal markets, and rate design.

Read A-7

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 26 of 59

JAMES A. READ, JR. Principal PUBLICATIONS Retail Risk Management: A Primer (with R. Goldberg), EPRI, Palo Alto: 2003. 1002225.

8

Analytic Approximations for Generation Option Values (with R. Goldberg), EPRI, Palo Alto: 2003. 1002209. Portfolio Optimization: Concepts and Challenges, EPRI, Palo Alto: 2002. 1001567. "Capital Allocation for Insurance Companies" (with Stewart C. Myers), Journal of Risk and Insurance, December 2001. (Selected by the Casualty Actuarial Society as the most valuable paper published by the American Risk and Insurance Association in 2001. Winner of the Robert C. Witt Research Award for outstanding feature article in the Journal of Risk and Insurance in 2001.) Optimization and Valuation of Natural Gas Storage (with R. Goldberg), EPRI, Palo Alto: 2001. 1005947. Describing Commodity Prices in the Energy Book System (with R. Goldberg), EPRI 1001170, Palo Alto: EPRI, 2000. "Energy Derivatives and Price Risk Management" (with A. Altman and R. Goldberg), in Pricing in Competitive Electricity Markets, A. Faruqui and K. Eakin (eds.), Kluwer Academic Publishers, 2000. Residual Obligations Following Electric Utility Restructuring (with F. Graves), Edison Electric Institute, May 2000. "Dealing With a Price Spike World" (with R. Goldberg), Energy & Power Risk, May 2000. Valuation and Management of Nuclear Assets, EPRI TR 107541, Palo Alto: Electric Power Research Institute, 1998. "Capacity Prices in a Competitive Power Market" (with F. Graves), in The Virtual Utility, Kluwer Academic Publishers, 1997. Option Pricing for Project Evaluation: An Introduction, EPRI TR-104755, Palo Alto: Electric Power Research Institute, 1995. The Utility Capital Budgeting Notebook (with others), EPRI TR-104369, Palo Alto: Electric Power Research Institute, 1994. "It=s All Downstream From Here" (with S. Thomas), Energy Risk, June 1994. "Analysis for Changing Minds" (with S. Thomas), Energy Risk, April 1994. Read A-8

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 27 of 59

JAMES A. READ, JR. Principal Project-Specific Discount Rates, report prepared for Electric Power Research Institute, 1992. "Rates of Return that Include New Gas Industry Risks," Natural Gas, November 1989.

9

"Estimating the Cost of Switching Rights on Natural Gas Pipelines" (with F. Graves and P. Carpenter), Energy Journal, October 1989. Holding Costs for Fuel Inventories, EPRI P-6184, Palo Alto: Electric Power Research Institute, 1989. "Option Pricing: A New Approach to Mine Valuation" (with S. Palm and N. Pearson), in Selected Readings in Mineral Economics, Pergamon Press, 1987. Capital Budgeting for Utilities: The Revenue Requirements Method, EPRI EA-4879, Palo Alto: Electric Power Research Institute, 1986. "Determining the Cost of Capital for Utility Investments" (with A.L. Kolbe and R. Lincoln), in Energy Markets in the Longer Term: Planning Under Uncertainty, Ed. A.S. Kydes and D.M. Geraghty, Elsevier Science Publishers, 1985. The Cost of Capital: Estimating the Rate of Return for Public Utilities (with A.L. Kolbe and G. Hall), Cambridge: MIT Press, 1984. Rate Shock and Power Plant Phase-Ins (with A.L. Kolbe), Charles River Associates, Washington, DC: Edison Electric Institute, 1984. Critique of Conventional Betas as Risk Indicators for Electric Utilities (with A.L. Kolbe), EPRI EA-3392, Palo Alto: Electric Power Research Institute, 1984.

PRESENTATIONS "Fundamentals of Portfolio Risk Management" (with R. Goldberg), Tutorial presented to Portfolio Optimization, Infocast, Houston, November 14-16, 2001. "Retail Transition Services in Electric Utility Restructuring," Presentation to Illinois Energy Leadership, Chicago, October 29-30, 2001. "Provider of Last Resort: Retrospect & Prospect," Presentation to Staff Subcommittee on Accounting and Finance, National Association of Regulatory Utility Commissioners, Portland, Maine, October 1, 2001. "Theory & Methods of Portfolio Risk Management" (with T. Parkinson), Tutorial presented to Portfolio Risk Analysis & Management, Infocast, Houston, February 16-18, 2000 and Chicago, October 2-4, 2000. Read A-9

Case 1:95-cv-00758-NBF

Document 280-3

Filed 03/04/2005

Page 28 of 59

JAMES A. READ, JR. Principal

10

"Using Option Pricing Formulas," Presentation to Pricing Wholesale Energy Products & Services, Infocast, Houston, November 28-30, 2000. "The Effect of Volatility Modeling on Management Decisions," Presentation to Market Price Volatility, Infocast, October 30 - November 1, 2000. "Option Pricing in a Price Spike World," Presentation to International Energy Pricing Conference, EPRI, Washington, D.C., July 26-28, 2000. "Applications of Portfolio Techniques to Fuel Decisions," Presentation to Fuel & Power Supply Seminar, EPRI, Cleveland, November 9-11, 1998. "Managing Nuclear Generation Assets," Presentation to Generation Asset Management: Opportunities and Challenges in the Electric Marketplace, EPRI, Baltimore, July 13-15, 1998. "Managing the Risks of Generation Assets," Presentation to Integrating Risk Management for Fuel Supply & Power Sales, Center for Business Intelligence, Houston, February 5-6, 1998. "Tactics for Matching Strategy & Market Opportunity through Hedging," Presentation to Fuel Management: Innovative Fuel Strategies for a Price-Competitive Power Market, Center for Business Intelligence, Colorado Springs, August 14-15, 1997. "Implementing Risk Management in Electric Power," Presentation to European Electricity Trading, ICM Marketing Ltd., London, January 29-30, 1997. "Integrating Fuel & Power Price Risk Management," Presentation to Managing Fuel Risk, Cent