2756 1 UNITED STATES OF AMERICA Before the 2 OFFICE OF THRIFT SUPERVISION DEPARTMENT OF THE TREASURY 3 In the Matter of: ) 4 ) UNITED SAVING ASSOCIATION OF ) 5 TEXAS, Houston, Texas, and ) ) 6 UNITED FINANCIAL GROUP, INC., ) Houston, Texas, a Savings ) 7 and Loan Holding Company ) ) OTS Order 8 MAXXAM, INC., Houston, Texas, ) No. AP 95-40 a Diversified Savings and ) Date: 9 Loan Holding Company ) Dec. 26, 1995 ) 10 FEDERATED DEVELOPMENT CO., ) a New York Business Trust, ) 11 ) CHARLES E. HURWITZ, ) 12 Institution-Affiliated Party ) and Present and Former Director ) 13 of United Savings Association ) of Texas, United Financial Group,) 14 and/or MAXXAM, Inc.; and ) ) 15 BARRY A. MUNITZ, JENARD M. GROSS,) ARTHUR S. BERNER, RONALD HUEBSCH,) 16 and MICHAEL CROW, Present and ) Former Directors and/or Officers ) 17 of United Savings Association of ) Texas, United Financial Group, ) 18 and/or MAXXAM, Inc., ) ) 19 Respondents. ) 20 TRIAL PROCEEDINGS FOR 10-14-97 21 22 2757 1 A-P-P-E-A-R-A-N-C-E-S 2 ON BEHALF OF THE AGENCY: 3 KENNETH J. GUIDO, Esquire Special Enforcement Counsel 4 BRUCE RINALDI, Esquire RICHARD STEARNS, Esquire 5 and BRYAN VEIS, Esquire of: Office of Thrift Supervision 6 Department of the Treasury 1700 G Street, N.W. 7 Washington, D.C. 20552 (202) 906-7395 8 ON BEHALF OF RESPONDENT MAXXAM, INC.: 9 FRANK J. EISENHART, Esquire 10 of: Dechert, Price & Rhoads 1500 K Street, N.W. 11 Washington, D.C. 20005-1208 (202) 626-3306 16 12 DALE A. HEAD (in-house) 13 Managing Counsel MAXXAM, Inc. 14 5847 San Felipe, Suite 2600 Houston, Texas 77057 15 (713) 267-3668 16 ON BEHALF OF RESPONDENT FEDERATED DEVELOPMENT CO. AND CHARLES HURWITZ: 17 RICHARD P. KEETON, Esquire 18 of: Mayor, Day, Caldwell & Keeton 1900 NationsBank Center, 700 Louisiana 19 Houston, Texas 77002 (713) 225-7013 3 20 21 22 2758 1 ON BEHALF OF RESPONDENT FEDERATED DEVELOPMENT CO., CHARLES HURWITZ, AND MAXXAM, INC.: 2 JACKS C. NICKENS, Esquire 3 of: Clements, O'Neill, Pierce & Nickens 1000 Louisiana Street, Suite 1800 4 Houston, Texas 77002 (713) 654-7608 5 ON BEHALF OF JENARD M. GROSS: 6 PAUL BLANKENSTEIN, Esquire 7 MARK A. PERRY, Esquire of: Gibson, Dunn & Crutcher 8 1050 Connecticut Avenue, N.W. Washington, D.C. 20036-5303 9 (202) 955-8500 10 ON BEHALF OF BERNER, CROW, MUNITZ AND HUEBSCH: 11 JOHN K. VILLA, Esquire MARY CLARK, Esquire 12 PAUL DUEFFERT, Esquire of: Williams & Connolly 13 725 Twelfth Street, N.W. Washington, D.C. 20005 14 (202) 434-5000 15 OTS COURT: 16 HONORABLE ARTHUR L. SHIPE Administrative Law Judge 17 Office of Financial Institutions Adjudication 1700 G Street, N.W., 6th Floor 18 Washington, D.C. 20552 Jerry Langdon, Judge Shipe's Clerk 19 REPORTED BY: 20 Ms. Marcy Clark, CSR 21 Ms. Shauna Foreman, CSR 22 2759 1 2 EXAMINATION INDEX 3 Page 4 TERENCE C. SMITH Examination by Mr. Guido.................2760 5 Cross-Examination by Mr. Nickens.........2827 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 2760 1 P-R-O-C-E-E-D-I-N-G-S 2 (10:00 a.m.) 3 THE COURT: The hearing will come to 4 order. Are there any procedural matters that we 5 should consider before we start? 6 MR. GUIDO: Not at this time, Your 7 Honor. 8 THE COURT: All right. Proceed. The 9 OTS, are you ready to call a witness? 10 MR. GUIDO: The OTS calls Mr. Terry 11 Smith, Your Honor. 12 13 TERENCE C. SMITH, 14 15 called as a witness and having been first duly 16 sworn, testified as follows: 17 18 EXAMINATION 19 20 21 THE COURT: Be seated, please. 22 Q. (BY MR. GUIDO) Would you please state 2761 1 your full name for the record. 2 A. Terence C. Smith. 3 Q. Where are you employed? 4 A. Federal Home Loan Bank of Dallas. 5 Q. How long have you been employed by the 6 Federal Home Loan Bank of Dallas? 7 A. Just over 11 years. 8 Q. And when did you first join the staff 9 of Federal Home Loan Bank of Dallas? 10 A. January 1, 1986. 11 Q. And what was your position there? 12 A. I was an assistant vice president of 13 financial management. 14 Q. And prior to 1986, where were you 15 employed? 16 A. At Southern Methodist University. I 17 was an assistant professor of finance in the 18 Edwin L. Cox School of Business there. 19 Q. And what is your academic background? 20 A. Undergraduate degrees from SMU in 21 economics, math, and history and completed my 22 Ph.D. studies at the University of Chicago for 2762 1 economics. 2 Q. Now, you -- when you first joined the 3 Federal Home Loan Bank of Dallas, did you do any 4 work with regard to mortgage-backed securities -- 5 A. Yes. 6 Q. -- portfolios? 7 A. Yes, I did. 8 Q. What is it that you did? 9 A. I wrote several articles for the Home 10 Loan Bank of Dallas quarterly publication 11 describing mortgage securities referred to as 12 risk-controlled arbitrages. 13 Q. Where did you get the information for 14 those articles? 15 A. Primarily from investment bank 16 publications. Morgan Stanley, First Boston, 17 Salomon Brothers, Goldman Sachs, all published 18 risk-controlled arbitrage pamphlets. Also, we 19 wrote -- I wrote an in-house computer program just 20 to try to understand what was going on with 21 mortgage securities. And then my financial 22 background and training allowed me to understand a 2763 1 fair amount about the prepayment option behavior 2 in mortgage securities. 3 MR. GUIDO: At this time, Your Honor, I 4 would like to move the admission of the four 5 articles that are being referred to. One is 6 entitled "The ABCs of Risk-controlled Arbitrage," 7 Exhibit B1509. That is the document that 8 Mr. Nickens referred to on pages 213 and 214 of 9 the transcript in his opening argument. 10 MR. NICKENS: No objection, Your Honor. 11 THE COURT: All right. Will you state 12 the numbers again? 13 MR. GUIDO: Yes. The title of the 14 article is "The ABCs of RCAs" and it's 15 Exhibit B1509 and it has Bates stamp W100880 16 through W100884, Your Honor. 17 THE COURT: All right. B1509 is 18 received. 19 MR. GUIDO: The next, Your Honor, in 20 the series is Exhibit B1649. It's "The Target 21 Value of Prepayment Options" and then it has a 22 subheading "Risk-controlled Arbitrage Considers 2764 1 the Effect of Market Volatility on Current 2 Coupons." 3 MR. NICKENS: No objection to B1649, 4 Your Honor. 5 THE COURT: Received. 6 MR. GUIDO: The third, Your Honor -- 7 I'm not so sure that these are in the correct 8 order, but it's Exhibit B1737. The title is 9 "Rebalancing: The Final Leap in Risk-Controlled 10 Arbitrage." 11 MR. NICKENS: No objection to 12 Exhibit B1737. 13 THE COURT: Received. 14 MR. GUIDO: The next, Your Honor, is 15 B2055; and the heading is "Prepayment Risk Can 16 Throw an Arbitrage Off Track." 17 MR. NICKENS: No objection to B2055, 18 Your Honor. 19 THE COURT: Received. 20 MR. GUIDO: We need one more copy of 21 that. 22 Q. (BY MR. GUIDO) Now, in the article 2765 1 entitled "Risk-Controlled" -- "The ABCs of RCAs," 2 the first article, you refer to the 3 risk-controlled arbitrage as "The grandiose titles 4 refer to programs developed by Wall Street's best 5 and brightest to determine how to finance the 6 purchase of mortgage-backed securities." 7 Do you recall that? 8 A. Yes, I do. 9 Q. Can you tell the Court what it is that 10 you're referring to when you refer to the 11 risk-controlled arbitrage of mortgage-backed 12 securities? 13 A. Risk-controlled arbitrage in the time 14 frame that I was writing was generally understood 15 to mean that mortgage assets were acquired by a 16 savings institution in this case. And the 17 financing of those mortgages generally came from 18 reverse repurchase agreements on Wall Street. And 19 there was generally a hedge involved which was a 20 synthetic lengthening of the liabilities to try to 21 mitigate some of the interest rate risk that's 22 inherent in the mortgage assets themselves. So, 2766 1 when you put all that together, you get an RCA. 2 Q. Now, look at Page 5 of the article. 3 There is a chart there or a graph. I guess it's 4 described as Figure 4, and it says "Federal Home 5 Loan Bank of Dallas risk-controlled arbitrage 6 analysis." And it has -- the first section is 7 called the trade structure. 8 Do you see that? 9 A. Correct. 10 Q. And it has under it "assets." Do you 11 see that? 12 A. Yes, I do. 13 Q. And it makes reference to Ginnie Mae 14 9 percents. Do you see that? 15 A. Correct. 16 Q. Now, can you tell us what Ginnie Maes 17 are or generically what a mortgage-backed security 18 is? 19 A. When institutions, banks, savings 20 institutions, or whatever originate home 21 mortgages, the mortgage that you would take out to 22 finance your house, those mortgages become assets 2767 1 of the savings institution or the commercial bank. 2 When the commercial bank or savings institution 3 has seven, eight, 900 of these mortgages, the size 4 obviously will get large enough that they may want 5 to worry about the credit risk. They will treat 6 those -- that bundle of mortgages as a pool of 7 mortgages that need to be hedged, need to be 8 worried about from credit concerns. In I guess 9 the late Seventies, early Eighties, securitization 10 was born, which meant that those mortgages, that 11 pool, the seven or 800 -- or it could be as few as 12 a hundred mortgages -- could be pooled together 13 and sold essentially to Ginnie Mae, Fannie Mae, or 14 ultimately Freddie Mac when it was created later 15 in the Eighties. And in return for delivering 16 those mortgages to Freddie Mac or Fannie Mae or 17 Ginnie Mae, the institution received a few 18 benefits. One of the benefits was that the 19 mortgages were credit guaranteed. A fee was paid 20 to one of the securitization agencies like Ginnie 21 Mae to guarantee the credit in that portfolio so 22 the ultimate investor that purchased the 2768 1 securities supported by those mortgages was 2 guaranteed to receive his principal and interest 3 payments whether or not the underlying mortgages 4 actually performed. If people actually defaulted 5 on the mortgages underneath those mortgage 6 securities, Ginnie Mae, Fannie Mae, Freddie Mac 7 would make the payments to the ultimate investors 8 of those mortgages; so, there was a credit support 9 that was provided by the agencies. 10 The other thing that occurred through 11 securitization is those -- that group of mortgages 12 became more liquid when they were securitized. 13 And by "liquid," what that means is they were more 14 easily tradable. If you have a Ginnie Mae 15 9 percent or -- I think that's what's here. A 16 Ginnie Mae 9 percent -- there was an assumption 17 that almost all Ginnie Mae 9 percent pools were 18 going to behave roughly the same. They were going 19 to have the same credit. They were going to have 20 roughly the same prepayment behavior and that all 21 of the mortgages that were made that supported 22 that Ginnie Mae 9 were properly documented; that 2769 1 if there was a foreclosure, there would not be a 2 problem in moving to foreclose -- I'm sorry. If 3 there was a default, there would be no problem in 4 moving to foreclose on the underlying properties. 5 So, because of that, people were 6 willing to trade Ginnie Mae 9 percents without 7 doing due diligence on the underlying loans which 8 made them much more liquid because you could talk 9 about Ginnie Maes -- let's talk about the -- 10 instead of talking about the characteristics of 11 all the loans in the pool. 12 Q. So, in other words, you could sell them 13 more quickly? 14 A. Right. And they were also more 15 available for financing. 16 Q. Now, in your hypothetical, your typical 17 risk-controlled arbitrage that you have here in 18 Figure No. 4, how were the purchase of the 19 mortgage-backed securities to be financed? 20 A. In this case -- I'm having a hard time 21 reading the titles. I believe it was on a reverse 22 repurchase agreement. 2770 1 Q. Okay. Now, what was a reverse 2 repurchase agreement? 3 A. When -- if you owned a Ginnie Mae 4 9 percent security, you could give that security 5 to Wall Street as collateral against a short-term 6 borrowing. The security was typically 7 hair cutted, which means that you wouldn't get a 8 hundred percent of its value in borrowing. 9 So, Wall Street may be willing to lend 10 you up to 95 percent of the value of that security 11 for a 30-, 60-, 90-day period; and you would pay a 12 rate of return on that money, an interest rate on 13 the cash that they gave you. 14 Q. Now, what was the advantage to a thrift 15 in using reverse repurchase agreements to purchase 16 mortgage-backed securities? 17 A. Reverse repurchases were a more 18 efficient way to fund than growing the deposit 19 base. At least, that was the thought at that 20 time. If you had to raise deposits in the market 21 in order to finance your mortgage purchases, odds 22 are you were going to have to raise the rate that 2771 1 you paid for deposits which was going to 2 ultimately affect your interest spread. 3 By using reverse repurchase agreements, 4 you would be putting those securities into a much 5 broader market, the national capital markets, to 6 raise that financing. And you could continue to 7 lever that financing. 8 Q. Did you get greater leverage by doing 9 so than you would by using deposits? 10 A. You could get much greater leverage, 11 yes. 12 Q. Okay. And what was the typical 13 leverage in a risk-controlled arbitrage portfolio? 14 A. The expectation was that you could get 15 up to 20 to 1 leverage. If there was a 5 percent 16 haircut on the security, then if you borrowed -- 17 if you provided $100 worth of collateral, you 18 could get $95 back. You could then take that $95 19 to buy securities, pledge them again against 20 repurchase agreements, and get 95 percent of 95. 21 And if you work that out, it's about 20 to 1 22 leverage. 2772 1 Q. So, that's on top of whatever cash that 2 you put in? 3 A. The capital that you initially put in. 4 Q. Now, you then talk about serialized 5 interest rate swaps in this diagram. What were 6 they intended to be used for in your hypothetical 7 transaction here? 8 A. Interest rate swaps were the hedge 9 vehicle that you tried to use to mitigate the 10 interest rate risk and then mortgages themselves. 11 The interest rate swaps, a financial contract was 12 entered into where one counter-party agrees to pay 13 a fixed rate for a time period. 14 For example, the institution that's 15 actually doing the risk-controlled arbitrage would 16 agree to pay a fixed rate of interest, perhaps 17 7 percent for five years, and they would pay that 18 to a Salomon or a Goldman or someone that was a 19 dealer in swaps. 20 And in return for paying 7 percent 21 fixed for five years, they would receive a payment 22 from their counter-party that was a short-term 2773 1 interest rate. They would receive three-month 2 LIBOR or three-month treasury bills reset every 3 three months over those five years. 4 So, they were able to fix their rate of 5 interest that they paid, receive a short-term 6 rate, and then the receipt of that short-term rate 7 of interest on the interest rate swap provided 8 them with the proceeds to repay the reverse 9 repurchase agreements. 10 So, as long as the sun and moon were 11 properly aligned, you were able to take 12 three-month repurchase agreements and turn them 13 into five-year fixed rate liabilities through 14 interest rate swaps. That's essentially what a 15 swap is. And by "serializing," what I meant is 16 you would not do a single five-year swap against 17 the entire asset position. You may do some 18 one-year swaps, three-year swaps, five-year swaps. 19 Q. Now, with regard to this swap 20 agreement, under this agreement, did the holder of 21 the risk-controlled arbitrage receive a variable 22 rate interest? 2774 1 A. They received a variable rate from the 2 interest rate swaps. 3 Q. And was the expectation that that would 4 offset the short-term interest rates that they had 5 to pay in the future? 6 A. Correct. 7 Q. So that no matter how rates moved, the 8 short-term rates would match each other? 9 A. That was the expectation. 10 Q. The expectation. And then on the other 11 side, there was a fixed rate. Right? 12 A. Correct. 13 Q. And that that fixed rate would lock in 14 the spread between the mortgage-backed securities 15 and the hedged instrument? 16 A. Correct. 17 Q. And that was the concept? 18 A. (Witness nods head affirmatively.) 19 Q. Now, you indicated that you used 20 staggered or serialized interest rate swaps as 21 opposed to just anticipating what the life of the 22 mortgage-backed security portfolio would be or the 2775 1 average life and just picking one date? 2 A. Correct. 3 Q. Why the serialized? 4 A. There are really two reasons. One is 5 that by -- the mortgages were going to amortize. 6 If no one ever repaid a mortgage early or 7 refinanced early, there was an expected 8 amortization in the mortgages over time. So if 9 you started off a with a hundred million dollars 10 of securities, at the end of five years, you knew 11 you were not going to have a hundred million 12 dollars. And what you would try to do by 13 serializing or staggering the swaps would be to 14 have a group of one-year swaps that would pay-down 15 and allow the remaining swaps to more closely 16 match the balance of the remaining assets in the 17 risk-controlled arbitrage. 18 So, by trying to fit the expected 19 repayments of principal on the mortgage assets to 20 how you set the swaps up allowed you to more 21 closely balance the transaction over time. The 22 other thing it allowed you to do is take advantage 2776 1 of the old curve because short rate -- the fixed 2 rate that you paid on short-term swaps was lower 3 generally than the fixed rate you would pay on 4 five-year swaps. So it was a little more 5 efficient way to finance the transaction. 6 Q. So, it lowered the cost of the hedge? 7 A. Correct. 8 Q. Now, why was risk-controlled arbitrage 9 being recommended to thrifts as opposed to just 10 purchasing mortgage-backed securities leveraged 11 with reverse repos? It seems that there would be 12 a much greater profit that way. 13 A. There would be significantly more 14 profit. There is also significantly more risk. 15 Q. Why is that? 16 A. Because if you're funding what's 17 essentially a 30-year mortgage, fixed rate 18 mortgage with a liability that would reset once a 19 month or once every three months as rates moved up 20 and down, you would significantly affect the 21 market value of that transaction and the income 22 that you earned from that transaction. 2777 1 One way to think about it is if your 2 mortgage asset has a 10 percent yield and you 3 initially fund it at 4 percent, you're making a 4 6 percent spread. Over time, however, if rates go 5 from 6 percent to 12 percent, you've now got 6 12 percent short-term financing funding an asset 7 that's yielding 10 percent. So you're actually 8 losing 2 percent on the transaction. 9 So, the issue is that if you just did 10 leveraged mortgages, you would have more interest 11 rate risk in the transaction and potentially 12 because of that, stand a much greater risk of 13 losing your initial capital investment plus more. 14 Q. So, if you have a portfolio of 15 mortgage-backed securities leveraged 20 to 1 as 16 you're describing, and interest rates moved 17 adversely, that could have a dramatic effect on 18 the value of the asset? 19 A. That's correct. 20 Q. Now -- and so, the risk-controlled 21 arbitrage was being recommended as a way of 22 avoiding that risk? 2778 1 A. You were trying to mitigate that 2 interest rate risk by locking in some amount of 3 financing at a fixed rate for some longer 4 maturity. 5 Q. Now, did the risk-controlled arbitrage 6 create a risk that was specific to the 7 risk-controlled arbitrage? 8 A. Well, the risk-controlled arbitrage had 9 several risks embedded in it. One of the risks 10 was there was still interest rate risk in the 11 transaction even though it was hedged. If you 12 think about what you're trying to do in this 13 transaction, if you have the prior transaction we 14 discussed where you just have a 10 percent 15 mortgage financed every 90 days and the rate can 16 reset, if you look at that transaction, there is a 17 distribution of outcomes. Some are very positive. 18 Some are very negative. What you're trying to do 19 in the risk-controlled arbitrage is give up the 20 very, very positive outcomes in order to get rid 21 of the very, very negative outcomes. 22 So, you're trying to mutate the risk or 2779 1 change the shape of the distribution of outcomes 2 so that you don't get as many bad as you do -- or 3 you don't get these very, very bad outcomes that 4 can wipe out your capital. 5 So, that's what you're trying to do in 6 a risk-controlled arbitrage. However, you can't 7 mitigate all the risk and still have any positive 8 spread left. So, you still have some degree of 9 interest rate risk. You have prepayment risk in 10 the mortgages and what that means is as rates move 11 up and down, people are either going to refinance 12 their homes or they are not. So, if rates drop 13 significantly, the people who have the mortgages 14 underneath these mortgage securities will start to 15 refinance down to lower coupons and, as a result, 16 the mortgage security will pay off. When the 17 mortgage security pays off more rapidly than 18 expected, then you're still going to have high 19 cost liabilities in a low-cost world which can 20 affect the institution so that prepayment risk and 21 interest rate risk are still present in the 22 risk-controlled arbitrages. They have just been 2780 1 changed from the pure levering transaction. 2 Q. Now, why do you still have high cost 3 liabilities? 4 A. Because if I agree to pay a fixed rate 5 interest for five years thinking that the original 6 mortgage security was going to exist for five 7 years and after I initiate the trade, rates drop 8 precipitously, all of the homeowners are going to 9 say "why am I paying a 10 percent mortgage in a 10 world where I can refinance to a 7 percent 11 mortgage?" They are going to prepay their 12 10 percent mortgage. Those prepayments flow 13 through to the holders of the mortgage-backed 14 securities as reduction in principal. 15 So, now, instead of having a hundred 16 million in mortgage securities, they may only have 17 30. But they have still got a hundred million 18 dollars in high rate five-year swaps in place. 19 So, they are still paying a very, very high rate 20 interest on the interest rate swap which is their 21 liability cost and now they are trying to reinvest 22 the proceeds from these 10 percent mortgages in a 2781 1 much lower rate world which would generate a 2 negative spread. 3 Q. Now, what were the ways to ameliorate 4 that prepayment risk in selecting the assets to 5 put into the portfolio? In other words, the type 6 of mortgage-backed securities. 7 A. The higher -- the choice of the coupon 8 of mortgage asset that you purchase is going to 9 say a lot about how much inherent prepayment risk 10 it has. Securities whose coupons ten years ago 11 were somewhere around 125 to 150 basis points or 12 one and a quarter, one and a half percent above 13 what you could get a current mortgage at were the 14 mortgages that were most susceptible to 15 prepayments because at that time, it took about a 16 point and a quarter to a point and a half 17 difference in rates to initiate the homeowners to 18 go out and pay the closing costs and the 19 refinancing costs to get a lower rate. 20 So, if you were buying mortgages that 21 were 150 basis points or one and a half percent 22 above where the current mortgage market was, those 2782 1 securities were going to be very, very susceptible 2 to having large prepayment changes for small 3 changes in rates. If rates went up a little 4 bit -- if rates went down just a little bit more, 5 you would expect those mortgages to prepay very 6 rapidly which would potentially destroy the 7 integrity of the original hedge position. 8 Q. And what about those that were close to 9 the current market interest rate? 10 A. The ones that are close to the market 11 are going to have the less prepayment sensitivity 12 for small changes in interest rates because if 13 rates go -- if 9 percent is where the mortgage 14 market is right now and you do a 9 percent 15 mortgage and rates fall to 8.75, the change in 16 your payment stream is probably not going to be 17 enough to pay the cost to -- to bear the cost of 18 refinancing, certainly ten years ago. 19 So, the issue is is you get securities 20 whose coupons are very close to the current market 21 or at a discount to the current market, they are 22 going to have less prepayment sensitivity. You're 2783 1 going to have a greater expectation that the 2 security is going to behave normally. 3 Q. And what if you bought a coupon that 4 was 125 to 150 basis points below that 9 percent 5 current market? 6 A. Those would tend to look like treasury 7 bonds. There would be no incentive for anyone to 8 prepay those mortgages unless they were changing 9 jobs and having to leave Houston to move to New 10 York. So, if you were in a 9 percent market world 11 and you had securities with 8 percent coupons, 12 anyone that tried to refinance would be taking a 13 cost on themselves. So, you would see very little 14 refinancing in that market. 15 Q. So, you had a cushion of a hundred 16 basis points before you became similar to an 17 at-the-money coupon? 18 A. Right. 19 Q. Is that a fair way to characterize it? 20 A. (Witness nods head affirmatively.) 21 Q. Now, had you ever referred to the type 22 of coupon that goes into the mortgage-backed 2784 1 securities as a mountain top? 2 A. It's really the -- the coupon and the 3 hedges together can create mountain tops or what 4 we used to call table top or mountainmountain top 5 trades. And what I meant by that is the typical 6 analysis that was used to determine the risk and 7 the return -- the expected risk and return in 8 these transactions was something called scenario 9 analysis. You would say -- assume nothing changes 10 from today. Interest rates stay exactly where 11 they are. Prepayments speeds on the mortgages are 12 exactly where they are. The market price of all 13 the assets are exactly where they are. What do 14 you expect the market value of this trade to do 15 over the next X years, whatever your horizon is? 16 And what do you expect the book income and the 17 book values to do? And then you would raise 18 interest rates by a hundred basis points, 200 19 basis points, do the same analysis. You'd lower 20 rates by a hundred, lower them by 200. And what 21 you would get is a chart that plotted rate changes 22 across the horizontal axis and returns across the 2785 1 vertical axis and what you would find is for 2 coupons -- for very high coupons, hundred and 50 3 or so out of the money, that were very difficult 4 to hedge, you tended to get what we called 5 mountain top trades and what that meant is you get 6 a very high expected return but small changes in 7 interest rates would give you a high volatility. 8 You would lose money in both directions in a 9 sense. 10 What table top trades were were using 11 assets that were less volatile. You were better 12 able to hedge them and for fairly wide changes in 13 interest rates, you could assure yourself of a 14 fairly stable return. 15 So, the question that you had to -- one 16 of the questions you had to ask yourself in 17 starting these transactions is how much return in 18 and risk am I willing to bear? Am I willing to 19 put myself in a position where a small change in 20 rates can wipe out my capital or am I willing to 21 take less return and stabilize my expected 22 returns? 2786 1 Q. Now, after the portfolio was in place, 2 you mentioned rebalancing the portfolio of 3 interest if interest rates declined to ameliorate 4 the effect of prepayment rates. Can you tell us 5 what you meant by that? 6 A. When you initially do any kind of 7 securities transaction that's hedged, over time, 8 the hedge is going to degrade because the world 9 doesn't turn out to be exactly what you thought 10 the world was going to be. And in risk-controlled 11 arbitrages particularly, as interest rates would 12 move up or down, you would find that the matching 13 of the hedge to the asset would start to fall 14 apart. You would get gains in the asset and 15 offsetting losses in the hedges that may or may 16 not completely offset on a dollar basis. And in 17 those situations, you would have to stop and look 18 at the transaction and say "Do I still have the 19 same expected returns and risk profile that I did 20 when I started?" And if I don't, what can I do to 21 try to recover the original profile of expected 22 return? So, rebalancing is essentially looking at 2787 1 the transaction periodically, weekly, monthly, or 2 when rates moved by some amount, 25, 50 basis 3 points, to see if the behavior of the actual 4 transaction is what you expected it to be. And if 5 it isn't, then you need to start looking at 6 readjusting either the assets or the liabilities 7 to bring yourself back into alignment. 8 Q. Well, in reference to the roll-down 9 strategy, were you recommending that institutions 10 roll down to lower coupons in response to a 11 reduction in interest rates in order to ameliorate 12 the impact of the acceleration in prepayments? 13 A. Yes. If rates were to decline, then 14 rolling down the coupon is probably the easiest 15 transaction because the securities were very 16 liquid. You could go from a Ginnie Mae 9 to a 17 Ginnie Mae 8 and a half. You would also have a 18 gain from that transaction because rates had 19 declined. And then you would take the gain from 20 that transaction and roll it back into rebalancing 21 the hedges which were probably at a loss. 22 Q. Now -- so, one of the recommendations 2788 1 was to sell out the higher coupons, buy a lower 2 coupon at a profit, take -- reinvest the proceeds, 3 take the profit and buy down the swap? Is that 4 what you said? 5 A. Buy down the swap. 6 Q. And did you also recommend another 7 alternative to buying down the swaps? 8 A. There were -- there were several things 9 that were done. If you were looking at a very, 10 very small change in interest rates, a quarter of 11 a percent, and you rolled down the assets and 12 you're still very near the beginning of the 13 transaction within the first six or eight months, 14 then you could probably roll down the asset, take 15 the gain out of the transaction, and not have very 16 many rebalancing costs. And the reason for that 17 is a 25 basis point move in the market is probably 18 not going to generate tremendous changes in 19 prepayments as long as the original asset was 20 fairly close to current market. Additionally, if 21 you're very near the trade inception, then if you 22 have an old 30 year asset that you're -- that's 2789 1 now a 29 and a half year asset that you're rolling 2 out for another -- a new 30-year asset, the 3 durations, the cash flow behaviors of those 4 securities aren't going to be very different. So, 5 you're probably fairly okay. If you're three 6 years into the transaction or five years into the 7 transaction and you start rolling five-year-old 8 mortgages for new mortgages, then you're going to 9 have a problem again because the cash flow 10 behavior is going to be different. 11 Q. So, one of the options that you were 12 recommending that they take the proceeds, they 13 take the cash and buy down the hedges. The other 14 is if it was just -- if it was a recently acquired 15 mortgage-backed security and it was a short -- a 16 small drop, to take those profits and they didn't 17 necessarily. 18 Q. You would have to reinvest those 19 profits? 20 A. That's correct. 21 Q. But did you ever recommend that they 22 take the principal, the return on whatever was 2790 1 left in the mortgage-backed securities that had 2 been invested in the mortgage-backed securities, 3 and the profit and reinvest that -- that, as well, 4 into new mortgage-backed securities? 5 A. I don't recall that being a specific 6 recommendation of mine. There were Wall Street 7 recommendations running around which would 8 essentially try to outgrow your problem. 9 Q. And what do you mean by "outgrow the 10 problem"? 11 A. You would take the gains and buy more 12 mortgages and then lever them with those gains and 13 attempted to make them new -- the new transaction 14 big enough that you could get enough extra spread 15 out of the new stuff to offset the loss in the old 16 hedges. 17 Q. So, was the theory that you would have 18 a larger portfolio although you were earning less 19 money that you would still -- less money per 20 entity, your net result would be the same because 21 you were earning it on a bigger pool of 22 mortgage-backed securities? 2791 1 A. That's correct. 2 Q. Now, at what level of interest rate 3 change were you recommending that the portfolio be 4 rebalanced? 5 A. Between 25 and 50 basis points 6 generally the portfolios needed some sort of 7 rebalancing transaction. 8 Q. Now, in -- that was in 1987 that you 9 wrote these articles. Was that your understanding 10 that that was the generally accepted level at 11 which the portfolio needed to be rebalanced in a 12 declining interest rate environment? 13 A. Well, it certainly was in '87 and I 14 believe before that, given some of the articles I 15 was reading from Wall Street at that point in time 16 I think were indicating the same general levels. 17 Q. Now, I'd like you to look at the 18 article which was entitled "Prepayment Risk Can 19 Throw an arbitrage Off Track," which is 20 Exhibit B2055. 21 I'd like to direct your attention to 22 Page 73. In the far right-hand column, the second 2792 1 paragraph -- do you see that paragraph? 2 A. The results of a 200 basis point? 3 Q. Yes. 4 A. Okay. 5 Q. It says "The results of the 200 basis 6 point increase or decrease in rates will expose 7 critical thresholds of the transaction and 8 indicate appropriate types of rebalancing." And 9 then it says "If no rebalancing is completed by 10 the time rates have moved 200 basis points, the 11 chances are that any rebalancing would be too 12 little too late." 13 Do you see that? 14 A. That's correct. 15 Q. What did you mean by "too little too 16 late"? 17 A. If you wait until rates have moved 200 18 basis points, what's happened in that case is that 19 the gain in the mortgage assets if it's a rate 20 decline is typically not going to be great enough 21 to offset the loss in the hedges. And you're 22 going to have to reinvest the cash at a much lower 2793 1 rate of interest, which means that you're going to 2 have to rebalance into a negative spread 3 transaction or a negative market value 4 transaction. 5 Q. And so that by the time you got to a 6 hundred basis point decline in interest rates, say 7 from 9 to 7 percent, is it the losses that already 8 been embedded in that market value of that 9 mortgage-backed security? 10 A. That's what our analysis showed, that 11 the losses were embedded in the transaction. 12 Q. In the transaction? 13 A. Transaction. 14 Q. Which was the reverse repos, the hedge 15 swap instrument, and the mortgage-backed security 16 pool acting all in conjunction? 17 A. Correct. 18 Q. To create a net result? 19 A. Correct. 20 Q. Now, look at the rebalancing article 21 which is Exhibit B1737. And I direct your 22 attention to the far left-hand column, the last 2794 1 paragraph which says "Most simulations of RCA 2 performance focus on the behavior of book income. 3 However, focusing on book measures can lead to 4 missed opportunities and delayed rebalancing of 5 the structure. Market value provides a leading 6 indicator for future changes in book income 7 induced by changes in prepayment speeds." 8 What did you mean by that when you were 9 making reference to "book income"? 10 A. By book income, I meant generally 11 accounted -- generally accounting correct income, 12 GAAP income. And what I meant by that is if you 13 were to put a risk-controlled arbitrage on and use 14 traditional accounting measures to book earnings 15 off of that transaction, those earnings were going 16 to lag what was going on in the marketplace. And 17 there is a lot of reasons for that. But I think 18 the easiest one to understand is if rates decline 19 a hundred basis -- or let's say decline 200 basis 20 points to assure ourselves that the mortgage asset 21 you started with will be prepaying very rapidly. 22 Rates decline 200 basis points today. It's going 2795 1 to take all the homeowners two or three months to 2 start the refinancing process. And again, at this 3 time period ten years ago, another couple months 4 to actually refinance. 5 So, between the time rates dropped 200 6 and the time you start to see the prepayments on 7 your mortgages, you may be five or six months down 8 the road. When the prepayments on the mortgages 9 start to flow into the institution, then you'll 10 start to see book income respond because at that 11 point, accretions and amortizations of 12 principal -- of premiums and discount will start 13 to accrue on the assets. At that point, you'll 14 start to see your book income spreads degrade. 15 However, as soon as rates declined, if you look at 16 the market value, you know you have a problem 17 because if you go out and revalue and try to sell 18 the assets, terminate the hedges, get out of the 19 reverse repurchase agreements, you would recognize 20 at that point you had a problem, that the 21 transaction wasn't working. So, if you wait until 22 book income responds to a change in rates, you're 2796 1 going to be five or six months beyond the point 2 where your problem really occurred. And if rates 3 are sort of slowly trending down as opposed to 4 this big shock, if you're looking at book 5 earnings, it could be a very, very long time 6 before you actually started to see the degradation 7 of the transaction of book income. 8 Q. So, were you recommending not to rely 9 on book income because it was misleading? 10 A. The lags in recognition of accounting 11 earnings could give you very misleading 12 expectations about how your trades were doing, 13 yes. 14 Q. Now, was it -- what were the 15 expectations at the time by the Federal Home Loan 16 Bank Board on the competence of the institution to 17 manage the risk-controlled arbitrage? 18 A. I was a bank employee but not a 19 regulator, but I'll take a shot at answering. The 20 expectations certainly that we had on the banking 21 side were that the institution entering into this 22 transaction would have the ability to model and 2797 1 understand the potential risk in the transaction, 2 that they would have adequate management 3 understanding of what can go wrong and what can go 4 right in the transaction, and if they were using 5 an investment advisor to help them, that they 6 maintain the appropriate controls over that 7 investment advisor. 8 Q. What do you mean by "the appropriate 9 controls over the investment advisor"? 10 A. The investment advisor generally was 11 expected to not have trading authority over the 12 account. The institution would have to do the 13 transactions themselves. The investment advisor 14 preferably would not be paid on a commission 15 basis, meaning that they would be potentially 16 incented to churn the account to recommend more 17 and more trades just to generate commissions or to 18 generate gains trading for the investment advisor. 19 So, trying to find a way to pay them that was not 20 commission based was recommended. 21 Q. Was it also recommended that they were 22 not to rely on -- solely on the advice of the 2798 1 people that were selling them the mortgage-backed 2 security risk-controlled arbitrage? 3 A. Which comes back to the fact that the 4 institution was expected to have some degree of 5 modeling capability internally to at least double 6 check what the investment advisor was doing for 7 them. 8 Q. Now, you say that it had somebody 9 managing the portfolio that understood the risks. 10 What risks were you talking about the portfolio 11 manager understanding? 12 A. The interest rate in prepayment risk 13 that we've already discussed. Potentially there 14 is the liquidity risk on top of that. If the 15 institution's credit was perceived to decline, 16 they may not be able to roll the reverse 17 repurchase agreements at maturity which would 18 create a liquidity problem. They couldn't fund 19 their balance sheet anymore. Those were the 20 primary risks. But the -- in terms of the pure 21 risk-controlled arbitrage, the focus was on really 22 interest rate risk and prepayment risk. 2799 1 Q. And when you say that the person should 2 appreciate the prepayment risk and how to manage 3 that, are you talking about assessing the risk -- 4 assessing the impact of the prepayments based on 5 the percentage changes in interest rates that 6 you've just discussed? 7 A. Correct. There should have been some 8 ongoing monitoring of the transaction. 9 Q. And should there have been an 10 appreciation of the significance of an interest 11 rate move of 50 basis points, for example? 12 A. Yes. 13 Q. A hundred basis points, for example? 14 A. Yes. 15 Q. 200 basis points, for example? 16 A. Absolutely. 17 Q. Now, was it the position of the Federal 18 Home Loan Bank when it was discussing 19 risk-controlled ash with people that the portfolio 20 be substantially hedged against interest rate 21 risk? 22 A. I'm not sure that the Home Loan Bank 2800 1 had a specific view. I know that what John and I 2 in our papers were recommending was exactly that. 3 Q. Now, in the article that -- 4 Exhibit B2055, there is an insert that says "Be 5 Aware of Potential Risks in Arbitrage" by Edward 6 Hjerpe? 7 A. Hjerpe. 8 Q. And who was he? 9 A. Ed at that time was an economist at the 10 bank board. 11 Q. And did you work with him in preparing 12 his insert into the article that's included here? 13 A. We had some preliminary discussions, 14 but I cannot take authorship of it. 15 Q. Okay. But you had discussions with him 16 about that? 17 A. Yes. This article was put together by 18 the savings institution staff and they asked for 19 John and I to Wright the lead article and for 20 Ed -- and I believe there is another person, Larry 21 Kenney, also have inserts in this article. 22 So, we had an initial discussion to 2801 1 sort of carve out the territory that each of us 2 would discuss so there wouldn't be a lot of 3 overlap. 4 Q. Well, look at the right-hand column on 5 the third page which is about halfway down the 6 beige page. Do you see that insert, it says "The 7 lure of gamblimg"? 8 A. Correct. 9 Q. What did that refer to? 10 A. Well, he said -- direct quote is it 11 comes into play if the arbitrage is used 12 inappropriately as a highly-leveraged, one-time 13 bet on interest rates. 14 Q. What is a highly-leveraged, one-time 15 bet on interest rates? 16 A. That would be in its purest case buying 17 mortgages that are 125 to 150 basis points above 18 the current interest rate funding them with 19 reverse repos and levering it as high as you can. 20 That would be a one-time rate. If rates went 21 down, you would hit what they refer to as a home 22 run. If rates go up, you'd be crushed. 2802 1 Q. Now, could the gambling also be done 2 with the money coupons? 3 A. You can gamble with pretty much any 4 security you wish. 5 Q. So, is this reference to basically not 6 having any hedges? 7 A. That's correct. 8 Q. So that you have a portfolio -- you 9 leverage your portfolio up with reverse repos and 10 you don't put on the swap? 11 A. Correct. 12 Q. Or some other hedging instrument? 13 A. Correct. 14 MR. NICKENS: Your Honor, we want to 15 get through this, but I do object to these leading 16 questions. I don't believe the witness is 17 testifying anymore, and I object to them. 18 THE COURT: Yeah. It seems to be 19 falling into Mr. Guido making the statements and 20 the witness agreeing. 21 MR. GUIDO: I will rephrase my 22 questions, Your Honor. 2803 1 Q. (BY MR. GUIDO) What was the risk to 2 the lure of gambling that you and Mr. Hjerpe 3 discussed that's included in his article? 4 A. The real risk is you put the 5 transaction on. You -- the transaction begins to 6 make money and you start thinking -- you start 7 confusing what the futures traders call confusing 8 brains with the bull market, that as this 9 transaction starts to make money for yourself, you 10 begin to believe that you've got the hot hand and 11 the transactions that you start to enter into to 12 lever the transaction more or to potentially even 13 hedge the transaction are driven more from your 14 view of the market as opposed to doing the full 15 hedge analysis initially, the full risk analysis 16 of the hedging. 17 Q. Now -- and you were recommending 18 against that? 19 A. Correct. 20 Q. Now, with regard to the trading risk 21 that's referred to in this article, what was the 22 trading risk that's being referred to there? 2804 1 MR. NICKENS: Could I get the 2 reference, Your Honor? 3 MR. GUIDO: I'm sorry. It's the next 4 item down on the same page in Exhibit B2055. 5 A. What he said discussing trading risk is 6 that you begin to churn the assets. You begin to 7 trade in and out of different mortgage assets as 8 you see gains or losses in them and stop really 9 looking at the transaction as a whole which is the 10 assets, the liabilities, and the financing. 11 Q. (BY MR. GUIDO) So, are you saying 12 that the trading risk is not consistent with the 13 risk-controlled arbitrage strategy? 14 A. The trading risk is something that can 15 come out of the risk-controlled arbitrage 16 strategy. When you start to see gains in the 17 transaction, there is always an incentive -- no 18 matter who you are and how pure you are, there is 19 always an incentive to take the gain when you see 20 it. So, you take a gain and nothing bad happens 21 to you and you get another little gain and you 22 take it and nothing bad happens to you. And the 2805 1 next thing you know, you've started trading the 2 portfolio. You're trading for the gains and 3 losses as opposed to looking again at the risk 4 profile of the entire transaction and looking to 5 see how the entire transaction's behaving and 6 whether or not you've knocked the hedges out of 7 balance by trading the assets. 8 Q. Now, let's shift a little bit and talk 9 about your understanding of the accounting for 10 these transactions at the time. 11 Was it your understanding -- 12 MR. NICKENS: Your Honor, I object to 13 any expert testimony in accounting. This person 14 has not been identified as an expert witness on 15 accounting issues. I don't believe he has an 16 accounting background or degree but, more 17 importantly, he's never been identified as an 18 expert on any issues but particularly on 19 accounting. 20 MR. GUIDO: Your Honor, I can rephrase 21 the question. I'm not asking him for any expert 22 opinion. I'm just asking him what his 2806 1 understanding was as to how the institutions 2 accounted for the transaction. 3 MR. NICKENS: And on that basis, I have 4 no objection. But I do strongly object to this 5 being received as expert testimony on proper 6 accounting. 7 THE COURT: Well, let's hear the 8 question first. 9 Q. (BY MR. GUIDO) The -- is it -- can 10 you tell us what your understanding was of how the 11 participants in a risk-controlled arbitrage should 12 account for the swap transaction? 13 A. The swap transaction was accounted for 14 on an accrual basis that you would essentially 15 accrue whatever the fixed rate of interest was 16 that you were going to pay. You would also accrue 17 the interest that you were going to receive. 18 Q. And if there were sales of the 19 mortgages and the mortgage-backed securities, 20 what's your understanding of how those were 21 accounted for? 22 A. My understanding is that the gains 2807 1 flowed through to earnings. The gains or losses 2 would flow through to earnings. 3 Q. Now, the next item talks about 4 mismatched hedging. Do you see that item? 5 A. Uh-huh. (Witness nods head 6 affirmatively.) 7 Q. Now, what did that refer to? 8 A. That is, I believe, referencing the 9 re -- or the lack of rebalancing in a transaction, 10 that if you initiate a transaction and it moves 11 through time and the expected behavior of the 12 assets change and you haven't done anything to 13 rebalance, you're going to be in a mismatched 14 hedge position. 15 Another thing it could possibly mean or 16 another way you can think about mismatched hedging 17 is that at the inception of the hedge, you did not 18 properly structure the hedges because you were 19 inherently putting a rate view into the 20 transaction. So that opposed to a pure hedge 21 which was trying to stabilize returns in an up and 22 down rate environment, you were trying to tilt the 2808 1 transaction so you got a little more gain if 2 trades went one way than the other way. I believe 3 that's generally what the mismatched hedging 4 concerns are. 5 Q. Could you get a mismatch because of the 6 roll-down strategy that you implemented? 7 A. Depending on the amount of time that 8 went by, you could get a mismatch from rolling 9 down. 10 Q. If you rolled down after a hundred 11 basis point move and you took the proceeds and you 12 reinvested those in lower coupon mortgage-backed 13 securities and the profits you took and used for 14 other purposes, would that result in a mismatched 15 hedge? 16 A. In general, it would. There are some 17 specific cases where it may not. 18 Q. And what were the cases that it may 19 not? 20 A. If rates had declined a hundred basis 21 points and the security that you initially started 22 with was a very, very deep discount meaning that 2809 1 its coupon was significantly below current market 2 rates, that rate movement may not change the 3 prepayment characteristics of the asset that you 4 were holding or the asset that you rolled down 5 into as long as a significant amount of time 6 hadn't passed for that hundred basis point drop to 7 occur. 8 Q. Any other situations? 9 A. That's the only one I can identify. 10 Q. Now, have you ever had any discussions 11 with the people at USAT about their 12 risk-controlled arbitrage portfolio? 13 A. I have talked to them -- I have talked 14 to people at USAT about risk-controlled arbitrage. 15 Q. Okay. Do you recall whether or not you 16 had a meeting with the people at USAT with regard 17 to the termination of a firm called Smith Breeden 18 by USAT? 19 A. Yes, I do. 20 Q. Can you tell us who Smith Breeden was? 21 A. Smith Breeden is a -- at that time was 22 a consulting firm, an investment advisory firm, 2810 1 that was one of the initial firms to try to hedge 2 mortgage securities using their first pass was 3 futures and options. Doug Breeden is a finance 4 professor. Smith, whose name I don't remember, 5 his first name, was a broker, a futures broker in 6 Kansas City. And they put this firm together to 7 advise savings institutions primarily on hedging 8 mortgage security transactions. 9 Q. Now, why did you meet with the people 10 at USAT regarding the termination of Smith 11 Breeden? 12 A. I was requested to meet with the people 13 at USAT to determine if there was either a flaw in 14 the Smith Breeden program that they were 15 terminating them because of or if they were going 16 to attempt to run a Smith Breedan-like program 17 internally without the help of a firm like Smith 18 Breeden. 19 Q. Now, did you have any other meetings 20 with the people from USAT in that time frame in 21 1986, early '87? 22 A. I don't recall any related to the 2811 1 mortgage securities. 2 Q. Now, did you ever have any meetings 3 with people at USAT regarding the mountain top or 4 table top distinction? 5 A. I remember having a discussion with 6 people about mountain tops and table tops. I had 7 lots of discussion with lots of institutions about 8 that time. 9 Q. Now, have you ever spoke in any 10 conferences at which the USAT people attended? 11 A. I believe I have, yes. The Home Loan 12 Bank put on loan put on several conferences. 13 Q. What time period? 14 A. '87, maybe as late as early '88. 15 Q. And what essentially was your 16 presentation? 17 A. My presentation was primarily what you 18 see in these articles dressed up. 19 Q. Now, did you ever have any specific 20 discussions with anyone to your knowledge about 21 the specifics of USAT's risk-controlled arbitrage? 22 A. I do not recall having that, no. 2812 1 Q. Have you ever had discussions with the 2 regulators in which they brought you something and 3 not identifying the institution and asked you for 4 your views? 5 A. I've had many of those. 6 Q. Okay. Do you know whether or not any 7 of those entailed advice with regard to USAT? 8 A. With respect to their mortgages, no, I 9 don't. 10 Q. Only the regulators would know that? 11 A. (Witness nods head affirmatively. 12 Q. Now, were you involved in the unwinding 13 of the swaps at USAT in late 1988 time period? 14 A. I was involved in unwinding swaps, yes. 15 Q. Okay. Can you tell us what that 16 involvement was? 17 A. We had been the intermediary for 18 several interest rate swaps that USAT had entered 19 into I guess in the '85, '86 time period. Those 20 swaps were at a loss to United Savings. We had 21 the swaps matched off in the sense that whatever 22 we were doing with United Savings, we had mirrored 2813 1 to a Wall Street firm so that we were a 2 pass-through. We were providing essentially a 3 credit enhancement function for United Savings 4 toward Wall Street. 5 So, we may have, for example, a 6 hundred -- may have had a hundred million-dollar 7 swap with USAT in an exactly offsetting hundred 8 million-dollar swap with Morgan Stanley or Salomon 9 Brothers to offset the risk at the home loan bank 10 from that swap position. 11 Q. Now, who else was involved in the sale 12 of the swaps in that 1988 time period? 13 A. The unwinding of the swaps? 14 Q. Uh-huh. 15 A. John Scott was involved somewhat at 16 that point. 17 Q. Anyone else? 18 A. On the financial side of the bank, no. 19 Q. Now, the discussion that you had with 20 the people at USAT with regard to the termination 21 of Smith Breeden, do you recall who was in that 22 meeting? 2814 1 A. It was Sandy Lawrenson and there was 2 another gentleman there, and I don't recall 3 offhand who that was. 4 Q. Was one of the people Michael Crow? 5 A. It may well have been. 6 Q. Was one of the people Bruce Williams? 7 A. I don't recall ever meeting Bruce 8 Williams. 9 Q. Was one of the people Walter Mueller? 10 A. I've never met Walter Mueller. 11 Q. Who is Walter Mueller? 12 A. Walter Mueller was at that time a 13 consultant that Bank United indicated they were 14 bringing in to replace Smith Breeden. 15 Q. Now, did the people from USAT at that 16 meeting ask you why you were meeting with them? 17 A. Yes. 18 Q. And what did they ask? 19 A. They asked what our concern about their 20 changing of an investment -- what our concern 21 could be with their changing of investment 22 advisors. 2815 1 Q. Can you describe for us what the tone 2 of that meeting was? 3 A. I think it was a fairly tense meeting 4 from both sides. 5 Q. Why was that? 6 A. They were unsure what I was doing 7 poking around in their business, coming from the 8 bank side, and I was unsure totally what I was 9 being sent down to find out. 10 Q. Was anyone else with you from the bank 11 side? 12 A. Just me. 13 Q. Now, you also were asked to go meet 14 with the people at USAT with regard to their junk 15 bond portfolio, were you not? 16 A. That's correct. 17 Q. And did you meet with any specific 18 individuals at USAT regarding their junk bond 19 portfolio? 20 A. The only person I recall specifically 21 was Joe Phillips, who I believe was the portfolio 22 manager. 2816 1 MR. GUIDO: I'd like to at this point 2 in time offer into evidence, Your Honor, 3 Exhibit B1042 which is a memorandum from Jonathan 4 Scott to Neil Twomey dated June 12th, 1986. 5 MR. NICKENS: Your Honor, we have no 6 objection to B1042. 7 THE COURT: All right. B1042 is 8 received. 9 Q. (BY MR. GUIDO) Now, have you had an 10 opportunity to review B1042 before your testimony 11 today? 12 A. Yes, I have. 13 Q. And when did you first see that 14 document recently? 15 A. I saw it when our in-house attorney 16 showed it to me last week before I came down here. 17 Q. Okay. Did you also see it yesterday? 18 A. Yes, I did see it yesterday. 19 Q. And did I provide it to you? 20 A. Yes, you did. 21 Q. So, you've had an opportunity to review 22 the document? 2817 1 A. Correct. 2 Q. Now, look at the second -- the first 3 paragraph. Excuse me. It says "On Tuesday, 4 June 10th, 1986, Terry Smith and I met with 5 Michael Crow, Bruce Williams, Joe Phillips, and 6 Art Berner of United Savings to discuss the 7 corporate bond portfolio." 8 Do you see that? 9 A. Yes. 10 Q. Do you recall that meeting? 11 A. Vaguely. 12 Q. Did this memorandum refresh your 13 recollection? 14 A. Yes, it did, when I read it. 15 Q. Now, why did you meet with the people 16 of United Savings to discuss the corporate bond 17 portfolio? 18 A. John and Neil asked me to attend a 19 meeting that they had set up to discuss how their 20 junk bond portfolio was being managed. 21 Q. Okay. And did you look at the 22 portfolio yourself? 2818 1 A. No. 2 Q. Did you look at any general ledgers 3 with regard to the portfolio yourself? 4 A. Nothing that I recall. 5 Q. Okay. Do you recall being provided 6 with the attachment, the United Financial Group 7 investment policy? 8 A. I don't recall being given that. 9 Q. Look at the second paragraph which says 10 "In September of 1984, United conceived the 11 strategy of purchasing less than investment grade 12 corporate debentures. B and BB rated bonds 13 financed with term deposits issued through 14 brokers." 15 Do you see that? 16 A. Correct. 17 Q. Where did you get -- where did you-all 18 get that information? 19 A. It was provided by Bank United or by 20 United Savings. 21 Q. So, they described essentially the 22 portfolio and their strategy to you? 2819 1 A. Correct. 2 Q. And then in the second sentence, it 3 says "The idea behind the strategy was to have a 4 dedicated portfolio of bonds duration matched the 5 term deposits to achieve a 200 basis point 6 spread." 7 Do you see that? 8 A. Yes, I do. 9 Q. Did you have any discussions with 10 regard to that sentence with them? 11 A. We asked them, as I recall, how they 12 planned to fund and manage the portfolio and that 13 was the answer they gave us. 14 Q. Did you ask them whether it was a 15 trading portfolio? 16 A. I did not. 17 Q. Did they tell you it was a trading 18 portfolio? 19 A. Not that I recall. 20 Q. Then it says "The term CDs financing 21 this portfolio range from seven to ten years 22 straight coupons to six to twelve year zeros with 2820 1 a total all-in cost of 11.6 percent." 2 Was that information also provided 3 orally at that meeting? 4 A. Since it's in the memo, I'm assuming it 5 was. 6 Q. But you don't have any independent 7 recollection? 8 A. (Witness shakes head negatively.) 9 Q. But you don't have any reason to 10 dispute the accuracy? 11 A. No, I do not have any reason to. 12 Q. Then it says "While the portfolio does 13 require certain adjustments to keep the durations 14 matched, the management of the bond portfolio does 15 not involve bond swapping to pick up yields or 16 premature selling to capture capital gains 17 currently unrealized." 18 Do you recall discussions along those 19 lines? 20 A. Somewhat. That is, we were told that 21 it was not a trading portfolio and by not being a 22 trading portfolio, it should be consistent with 2821 1 that sentence. 2 Q. Now, did you at any time thereafter 3 have discussions with the people of Bank United 4 regarding their junk bond portfolio? 5 A. I don't recall speaking with USAT after 6 that. 7 Q. Do you recall having any meetings in 8 which you discussed with them the -- bringing in 9 someone to look at the high-yield bond portfolio 10 for them? 11 A. I don't recall talking to USAT 12 specifically about that. I was asked by the 13 regulators to provide them with the names of some 14 institutions, Wall Street type firms, that could 15 do such an analysis for them. And I gave them 16 some names. 17 Q. I have a handwritten memorandum which 18 is a little difficult to read. 19 MR. GUIDO: Your Honor, these are 20 handwritten notes that came from the supervisory 21 file. They are Deposition Exhibit 69 from the 22 Ginger Baugh deposition. They have been marked 2822 1 by -- as Exhibit A11008. 2 MR. NICKENS: Your Honor, if Mr. Guido 3 is offering A11008, I have no objection. 4 THE COURT: Received. 5 Q. (BY MR. GUIDO) This appears to be a 6 summary of a meeting which a number of people had 7 in March of 1988 and it mentions you and John 8 Scott being there and it looks like it mentions a 9 Mr. Lemanski, Ray Lemanski, as being there. 10 Do you recall attending such a meeting? 11 A. I don't recall attending the meeting. 12 Q. The -- so, you couldn't shed any light 13 on what this memorandum is discussing? 14 A. Not only that, I can't even read it. 15 Q. Okay. Do you recall having any 16 meetings with any of the regulatory people at the 17 Federal Home Loan Bank of Dallas with regard to 18 recommendations of people that they might consult 19 if they wanted to have the high-yield bond 20 portfolio looked at again? 21 A. Yes. 22 Q. Do you recall who those discussions 2823 1 were with? 2 A. Probably Neil Twomey or one of his 3 analysts. 4 Q. And did you make any recommendations? 5 A. I did give them two or three names of 6 firms. 7 Q. And who were they -- which firms' names 8 did you give them? 9 A. I don't even recall which names I gave 10 them at that point. It was whoever was big in 11 high-yield, I would assume. 12 Q. Did you give them the names of any 13 particular individuals? 14 A. What I typically did when the 15 regulators asked me for something like this was to 16 determine what firms we thought were the best in 17 that area and then call our sales coverage at that 18 institution and ask who was in the junk bond area 19 that would be willing to do this. Since we did 20 not trade junk bonds or hold junk bonds at the 21 Home Loan Bank, we didn't have any direct 22 knowledge of who the good junk bond analysts were. 2824 1 So, what I would assume we did was call 2 the sales coverage at three or four firms, ask 3 them who at their company would do that, and then 4 provide those names to the regulators. 5 Q. Did you ever look at the general 6 ledgers of United Savings Association to ascertain 7 whether the portfolio was being managed as a 8 trading portfolio or an investment portfolio as 9 described in Exhibit B1042? 10 A. Not to my knowledge. 11 Q. Did you ever look at the portfolio to 12 ascertain whether there were any affiliated party 13 transactions in that portfolio? 14 A. Not to my knowledge. 15 Q. Did you ever look to ascertain whether 16 or not there was proper diversification in that 17 portfolio in terms of industry groups? 18 A. Not to my knowledge. 19 Q. Did you ever look at that portfolio to 20 ascertain whether or not there was appropriate 21 diversification in the brokers through whom 22 high-yield bonds were purchased? 2825 1 A. I don't recall that either. 2 THE COURT: Mr. Guido, we'll take a 3 short recess. 4 MR. GUIDO: Yes, sir. 5 6 (A break was taken.) 7 8 THE COURT: We'll be back on the 9 record. 10 Mr. Guido, you may continue your 11 examination of the witness. 12 MR. GUIDO: Thank you, Your Honor. 13 Q. (BY MR. GUIDO) I'd like to at this 14 point in time turn back to the mortgage-backed 15 securities and close off the questioning on 16 direct, if I may. 17 We talked about the meeting that you 18 had when you met with the people at USAT regarding 19 the Smith Breeden termination, and we also 20 discussed the meeting which you described as -- 21 the tone as being tense. 22 Do you recall that? 2826 1 A. Yes. 2 Q. Were those one and the same meeting? 3 A. Yes. 4 Q. And at that meeting, what was it that 5 you addressed with Sandy Lawrenson and the people 6 from USAT? 7 A. My specific questions in that meeting 8 were why had they chosen to break the relationship 9 with Smith Breeden and go another way. That was 10 the focus of my discussions. 11 Q. Did you ask them anything about their 12 portfolio? 13 A. Not that I recall. 14 Q. Did they say anything about their 15 portfolio to you at that meeting? 16 A. The only thing they said was that they 17 believe they can do an adequate job of managing 18 the portfolio using Walter and that they didn't 19 need Smith Breeden. I don't recall getting into 20 any details of the composition of the portfolio. 21 Q. Did they tell you that the portfolio 22 was a risk-controlled arbitrage? 2827 1 A. We may have used the term. I don't 2 recall. 3 Q. Was it your understanding that the 4 mortgage-backed security portfolio that they had 5 was a risk-controlled arbitrage? 6 A. If Smith Breeden was running it, it was 7 a risk-controlled arbitrage. That's what they 8 did. 9 MR. GUIDO: No further questions, Your 10 Honor. 11 THE COURT: Mr. Nickens. 12 MR. NICKENS: Your Honor, we've got 13 some documents. It'll just take a second. 14 15 CROSS-EXAMINATION 16 17 (11:39 a.m.) 18 Q. (BY MR. NICKENS) Mr. Smith, first let 19 me introduce myself. My name is J.C. Nickens. 20 We've never met before, have we? 21 A. No. 22 Q. I represent MAXXAM as one of the 2828 1 respondents in this matter. 2 Have you ever read the notice of 3 charges? 4 A. I have not read the entire notice of 5 charges, no. 6 Q. Has anyone ever explained to you about 7 what the nature of the claims were in the notice 8 of charges relating to mortgage-backed securities? 9 A. Technically, no. 10 Q. Has anyone ever indicated to you that 11 one of the charges was that the roll-down was 12 imprudent? 13 A. That has been discussed. 14 Q. But that the reaction to a lowering of 15 interest rates, a roll-down, was imprudent. Has 16 anyone ever said that to you? 17 A. No. I was told that there was a 18 roll-down after rates had moved significantly. 19 Q. And a roll-down would be the 20 appropriate response to declining rates with a 21 risk-controlled arbitrage, isn't it? 22 A. In certain situations. 2829 1 Q. And that was the recommended response 2 that the Federal Home Loan Bank published in those 3 circumstances, isn't it? 4 A. I believe that what we published was 5 when rates declined, you should be rolling down 6 the mortgage assets along with some other things. 7 Q. Yes, sir. Now, with regard to your 8 testimony about rebalancing, one of the articles, 9 the third one that you wrote, was on the subject 10 of rebalancing; isn't that correct? 11 A. I wrote one on rebalancing. If you say 12 it was the third one, I guess I'll go along with 13 you for now. 14 Q. You'll see -- 15 A. It says "Rebalancing: The Final Leap 16 in Risk-controlled Arbitrage?" 17 Q. Yes, sir. And the very first sentence 18 is "This is the third article in a series of 19 risk-controlled arbitrage programs." 20 A. That's it. 21 Q. Okay. And now, rebalancing was an 22 essential part of a risk-controlled arbitrage, 2830 1 wasn't it? 2 A. Correct. 3 Q. So, if you had a volatile interest rate 4 market, you would expect to see a good manager 5 responding to that? 6 A. That is correct. 7 Q. And that rebalancing involves buying 8 and selling assets or at least one possible 9 response. Right? 10 A. That is part of the rebalancing trade. 11 Q. And so, one would expect to see 12 turnover in the portfolio in a volatile interest 13 rate market? 14 A. Depending on the degree of volatility 15 and the time span, yes. 16 Q. Because that is the nature of 17 rebalancing? 18 A. All hedges have to be rebalanced. 19 Q. Yes, sir. But my only point -- let's 20 see if we can get agreement -- is that the way you 21 do that or one way to do that on the asset side is 22 to buy and substitute other assets? 2831 1 A. I can agree with that, yes. 2 Q. And if one is looking at that 3 portfolio, one would see turnover? 4 A. You are going to see some turnover. 5 Q. And it would be entirely expected? 6 A. In a volatile rate environment, yes. 7 Q. Now, you indicated to Mr. Guido that a 8 manager in a declining -- or in an interest 9 rate -- in a volatile interest rate market would 10 make changes in the portfolio with changes of 25 11 basis points? 12 A. What I'm saying is at 25 to 50, you 13 need to be very sure that the transaction is 14 remaining intact -- and by "intact," I mean that 15 the risk return profile that you started with 16 still looks like it's there -- that if you're 17 using assets that are behaving peculiarly or if 18 you have initially put volatile assets in at 25 to 19 50, you probably need to start rebalancing. 20 Q. And was that your thought back in 1987 21 when you wrote this article? 22 A. My thought was that 50 was about the 2832 1 right number, yeah. 2 Q. And have you done any -- have you made 3 any effort to remind yourself of your thinking 4 back in 1987? 5 A. My -- I have read my articles over, 6 yes. 7 Q. Well, you don't see anything in there 8 about 50 basis points, do you? 9 A. What you see in the articles -- 10 Q. Sir, let -- 11 A. Okay. 12 Q. Do you see anything in there -- 13 A. Specifically, do I see anything in here 14 relating to 50? I do not see the number 50 in 15 here, no. 16 Q. Do you see any number for recommended 17 rebalancing in your article on rebalancing in the 18 fall of 1987? 19 A. I do not see any recommendations 20 related to basis point changes requiring 21 rebalancing in this article. 22 Q. Where would we find such information? 2833 1 A. There are, I believe, in some of the 2 other articles and in some of the talks I gave 3 references to 100 and 200 basis point moves. 4 Q. Now, have you published anything more 5 than these four articles plus another -- anything 6 more than these four articles on the subject of 7 risk-controlled arbitrage? 8 A. I don't believe so; but since you were 9 looking for one, I may have. 10 Q. Well, I'll tell you what I was looking 11 for. There is another article that is a version 12 of one of these. 13 A. Yes. There are two versions of one of 14 these articles running around, that's correct. 15 Q. So, are you telling us we ought to be 16 able to find a reference in one of those articles 17 to when you rebalance? 18 A. What I think you should be able to find 19 in here is that you need to be monitoring the 20 program and you need to be making sure that the 21 program is behaving as expected. In terms of 22 looking at an article and finding a reference to 2834 1 25 or 50 or even 100 as a specific recommendation 2 for rebalancing, I don't believe they are in here. 3 Q. In fact, for those people who did 4 publish on the subject in 1987, they recommended a 5 hundred basis points, didn't they? 6 A. I can't speak for everything that was 7 published. 8 Q. Well, let's -- let's see if we can look 9 at some of that. 10 Are you familiar with the Institutional 11 Investor? 12 A. Yes. 13 Q. Is that a well-accepted authority 14 publication? 15 A. It's a fairly reputable -- yeah. 16 Q. Okay. Let me see if I can find this 17 for you: B823. Do you know Barbara Donnelly? 18 A. I do not. 19 Q. Have you seen this article before? 20 A. I vaguely remember reading this article 21 years ago. 22 MR. NICKENS: Your Honor, we offer 2835 1 B823. 2 MR. GUIDO: If this is being offered, 3 Your Honor, as a statement of what was being 4 published at the time about risk-controlled 5 arbitrage, we have no objection. 6 THE COURT: Received. 7 Q. (BY MR. NICKENS) Now, with regard -- 8 I'm going to come back to this, Mr. Smith. But 9 with regard to this specific point, if you'd look 10 over on the third page of B823. And you'll see in 11 the middle column, there is an identifier "leaner 12 pickings." 13 Do you see that? 14 A. Uh-huh. 15 Q. And if you come down to the second full 16 paragraph, it starts out "most practitioners." 17 Do you see that? 18 A. Uh-huh. 19 MR. GUIDO: What page are we on? 20 MR. NICKENS: On the third page. 21 Q. (BY MR. NICKENS) If you read along 22 with me and make sure I read this right. It says 2836 1 "Most practitioners feel it's sufficient to 2 reevaluate their arbitrages either monthly or with 3 every hundred basis point change in rates, 4 whichever comes first. One particularly rigorous 5 Drexel customer that rebalanced daily found its 6 initial spread of 125 basis points gradually 7 whittled down to about 35 basis points by early 8 December." 9 Now, does that refresh your 10 recollection that most practitioners were looking 11 at changes of a hundred basis points as being -- 12 this is in February of 1986. 13 MR. GUIDO: Objection. That 14 mischaracterizes what the sentence says, Your 15 Honor. The sentence says "It's sufficient to 16 reevaluate their arbitrages either monthly or with 17 every hundred basis point change in rates, 18 whichever comes first," Your Honor. It does not 19 say wait until a hundred basis point move. 20 Q. (BY MR. NICKENS) Sir, does that 21 refresh your recollection as to what -- 22 THE COURT: Can you answer the 2837 1 question? 2 A. Okay. I don't know that it refreshes 3 my recollection. Certainly, whoever this author 4 talked to told her that that's where the market 5 was. What was happening at that time period is 6 most analysis was done in hundred basis point 7 increments. 8 Q. (BY MR. NICKENS) Now, if you -- let's 9 talk about -- you've got this person that's 10 managing the portfolio and it comes to the end of 11 the day and he sees that interest rates have 12 fallen 25 basis points. That is a quarter of 13 one point. Right? 14 A. That's correct. 15 Q. So, it wouldn't be so unusual to see a 16 25 basis point movement? 17 A. In a seven- or ten-year security, 18 that's not that often. 19 Q. Well, it still happens. Right? 20 A. It does happen, yes. 21 Q. And so, at the end -- he immediately 22 sells the entire portfolio because he's had a 2838 1 25 basis point movement? 2 A. What I was indicating or what I was 3 trying to indicate with the 25 to 50 is that is 4 the point at which you need to be assuring 5 yourself you don't have to sell the portfolio or 6 if you do need to do something, to go do it. 7 Q. That means -- you mean you need to 8 monitor it? 9 A. That's correct. 10 Q. But you didn't mean to imply to the 11 Court that at 25 basis points, you ought to sell? 12 A. There are some transactions that you 13 may want to sell at 25 basis point changes. 14 Q. But were you indicating to the Court 15 that it is your opinion that with a 25 basis point 16 movement, that you ought to sell the portfolio? 17 A. Not in all cases. 18 Q. What you ought to do is evaluate the 19 situation. Right? 20 A. That's correct. 21 Q. Because that manager doesn't know 22 whether that movement is going to reverse itself 2839 1 the next day, does he? 2 A. It may reverse itself. The -- what the 3 manager needs to assure himself of at 25 basis 4 points is that the integrity of the transaction is 5 still in place, that his expectations for how that 6 trade is going to behave will still occur. 7 Q. And so, what he looks at is he looks at 8 his -- in this particular case, he looks at his 9 prepayment assumptions; and he calculates his 10 durations. Right? 11 A. That's correct. 12 Q. Or maybe he runs another scenario 13 analysis? 14 A. More than likely, that would -- 15 Q. And then he has to make a decision, you 16 know. Have things changed enough that I need to 17 make a drastic move or a less drastic move in 18 adjusting the portfolio? 19 A. I would agree with that. 20 Q. And that's true whether you're talking 21 about 25 basis points, 50 basis points, 100 basis 22 points. He has to continually monitor -- 2840 1 A. Monitor. 2 Q. -- the portfolio. Right? 3 A. Yes. 4 Q. And you don't have any information, do 5 you, that USAT didn't do that? 6 A. I do not have any information. 7 Q. Now, let me ask you to also look at 8 B1475. 9 Are you familiar with the firm of First 10 Boston? 11 A. Yes, I am. 12 Q. Are they a recognized investment 13 banking firm -- 14 A. Yes, they are. 15 Q. -- with expertise in this area? 16 A. Yes, they are. 17 Q. And do you know the name Lisa Wolfson? 18 A. I know the name. 19 Q. Or Carol Garfein? 20 A. And, again, the names only. 21 Q. Okay. 22 MR. NICKENS: Your Honor, this 2841 1 indicates that it came from the research and 2 information library, the Federal Home Loan Bank of 3 Dallas. 4 Q. (BY MR. NICKENS) Right? February, 5 1987. Is that -- you're shaking your head. She 6 can't get that down, Mr. Smith. 7 A. I'm sorry. Yes, that is what it says. 8 MR. NICKENS: Your Honor, we offer 9 B1475. 10 MR. GUIDO: No objection, Your Honor. 11 THE COURT: Received. 12 Q. (BY MR. NICKENS) So, this is 13 information that was available in your own library 14 about what was the prevailing practice back in 15 1987, correct? 16 A. First Boston's view of it, yes. 17 Q. Let me ask you, with regard to this 18 particular point, to look over at Page 12. Are 19 you with me? 20 A. I'm with you. 21 Q. Okay. On the left side, I want to draw 22 your attention to the third paragraph that starts 2842 1 "risk-controlled arbitrages." It says here, 2 "Risk-controlled arbitrages should be monitored 3 frequently, perhaps monthly or quarterly, or with 4 significant changes in market rates to ensure that 5 a properly hedged position remains." 6 Do you see that, sir? 7 A. Yes, I do. 8 Q. You certainly agree with that. Right? 9 A. I do agree with that. 10 Q. And you have no information to suggest 11 that USAT didn't do that? 12 A. I have no information. 13 Q. Then if you look over on the right 14 side, you can see there that they are talking 15 about a specific event. Right? 16 A. Uh-huh. (Witness nods head 17 affirmatively.) 18 Q. And if you go down to the bottom of the 19 left side, it says "In highly volatile interest 20 rate environments" -- 21 A. Hang on. I lost you. Where are you 22 again? 2843 1 Q. At the bottom on the left column in the 2 paragraph -- 3 A. Highly -- okay. 4 Q. Okay. It's the carry-over paragraph. 5 A. Right. 6 Q. It says "In highly volatile interest 7 rate environments such as during the past two 8 years" -- and this is February of 1987, correct? 9 A. That's correct. 10 Q. -- "an arbitrage transaction should be 11 rebalanced fairly often. As an example, the 12 current coupon Fannie Mae arbitrage purchased in 13 January 1985 would have involved Fannie May 13s 14 with initial yield of 13.37 percent based upon a 15 projected prepayment rate of .60 SMM." 16 Now, let's pause there. And can you 17 tell the Court what SMM is? 18 A. SMM is a measure of the speed that the 19 mortgages are prepaying. I think that means -- 20 and First Boston used a little different measure 21 than everyone else, but I think that means that 22 .6 percent of the balance was repaying monthly. 2844 1 Q. Yes, sir. This was a monthly measure, 2 correct, as opposed to the CPR which was an annual 3 measure? 4 A. I believe that's correct. 5 Q. But you can't simply multiply by 12 6 because there is a compounding feature to it. 7 Right? 8 A. That's correct. 9 Q. And so, there are tables to allow 10 people to translate SMMs to CPRs. Right? 11 A. That's right. 12 Q. But a rough measure would be 12 times? 13 A. At a number that small, it's got to be 14 pretty close. 15 Q. Right. If you get into larger numbers, 16 then you have a problem? 17 A. Then compounding does kick in. 18 Q. It says "From January 1985 to November 19 1986, the yield based upon projected SMM on the 20 Fannie Mae 13 fell to 8.75 percent." 21 Now, that's a significant move, isn't 22 it, Mr. Smith? 2845 1 A. That would be significant. 2 Q. "The actual latest three months' SMM 3 rose over that period from .38 percent SMM in 4 January 1985 peaking at 8.71 percent SMM in 5 August." 6 Do you see that? 7 A. I see that. 8 Q. "And then down to 6.9 percent in 9 November 1986. In January 1985 when the arbitrage 10 was initiated, a hedged design was structured 11 based upon the projected SMM at the time, 12 .60 percent SMM." 13 Do you see that? 14 A. I see that. 15 Q. So, that -- the manager that planned 16 this arbitrage had projected an SMM of 17 .60 percent, right? 18 A. That's correct. 19 Q. And in August, it reached 8.71 percent? 20 A. That's correct. 21 Q. Now, that is a huge difference, isn't 22 it? 2846 1 A. That was the historic change until 2 recent years. 3 Q. That is off by 16 or 17 times what the 4 person had planned for? 5 A. Close to that. 6 Q. "Left unbalanced over this period, the 7 arbitrage would have been drastically overhedged 8 by the end of 1985. If the arbitrage, however, 9 was rebalanced at every change in yield of the 10 asset of approximately 100 basis points, the hedge 11 would be restructured based on the new SMM 12 projection." 13 Do you see that? 14 A. I see that. 15 Q. Now, does that refresh your 16 recollection that practitioners of the time -- 17 this is in 1987. Right? 18 A. Uh-huh. 19 Q. Yes, sir? 20 A. Yes. I'm sorry. 21 Q. And this was after the market had 22 experienced the huge increase in prepayments that 2847 1 had occurred and is described here. Right? 2 A. That's correct. 3 Q. And this person is telling people -- 4 First Boston is telling people if you had changed 5 at 100 basis points, you still could have 6 preserved the arbitrage? 7 A. What they are saying is if you 8 rebalanced at every hundred basis points, you 9 certainly would have been better off if you didn't 10 hedge it. I don't know about preserving the 11 arbitrage. But I also don't know that this 12 sentence is stating that they recommend doing it 13 every hundred basis points. I think what you may 14 be seeing here is an artifact of the modeling 15 technology; that instead of looking at every rate 16 move, they looked at it in hundred basis point 17 increments in order to run this example. 18 Q. Well, you certainly don't see them 19 saying anywhere in there "and we recommend that 20 you move every 25 basis points"? 21 A. I see no recommendations as to when it 22 should be rehedged in here. 2848 1 Q. And you don't draw any inferences from 2 the fact that they chose 100 as the standard other 3 than the fact that that may be what was used in 4 the modeling? 5 A. That would be my inference from -- 6 Q. In fact, the duration changes have to 7 do with changes in interest rate of a hundred 8 basis points, don't they, or changes in 9 price/yield? 10 A. The standard McCauley duration is for a 11 hundred -- for a 1 percent change in rates which 12 is not a hundred basis point move in rates unless 13 you're at 10 percent. 14 Q. It involves a 1 percent movement in 15 interest rates, correct? 16 A. It is -- a McCauley duration is a 17 1 percent -- it is the change in the value of the 18 transaction for a 1 percent change in rates. That 19 is, if you're at 10 percent, a 1 percent change is 20 one. If you're at 5 percent, a 1 percent change 21 is 50. 22 Q. All right. 2849 1 MR. NICKENS: Your Honor, I'm preparing 2 to move into another subject and I -- whatever 3 your lunch plans are. And I will inform the Court 4 that I am planning to speak -- to ask the witness 5 about McCauley duration, effective duration, 6 modified duration, implied duration, empirical 7 duration, things of this subject matter. 8 THE COURT: So, you will have some 9 extensive more -- 10 MR. NICKENS: Your Honor, I expect and 11 I have informed Mr. Guido that I will have 12 Mr. Smith, if not for the entire afternoon -- that 13 with redirect, it is likely to take the entire day 14 with this witness. 15 THE COURT: All right. We'll adjourn 16 till 1:30. 17 18 (Luncheon recess.) 19 20 MR. GRIFFITH: Your Honor, may I take 21 up just a short matter? 22 THE COURT: We'll be back on the 2850 1 record. 2 MR. GRIFFITH: My name is David 3 Griffth. I represent MAXXAM. 4 As you may recall, the Judge entered a 5 protective order dealing with MAXXAM and that was 6 handed to the Court the day after. An order was 7 handed up concerning Federated's shareholder list. 8 Mr. Veis has called and asked -- he wants to see 9 those shareholders lists at 2:00 o'clock today. 10 There was no order in place yet to allow for that, 11 but we don't want to hold up his review. I've got 12 a copy of those orders that were handed up for the 13 Court's execution if that's agreeable so that he 14 can stick with whatever schedule he's got. 15 THE COURT: All right. This is the 16 same order I signed. 17 MR. GRIFFITH: Exactly. Except for 18 this one, Your Honor, says FedRe instead of 19 MAXXAM. 20 THE COURT: But it doesn't make a time 21 certain for the production. 22 MR. GRIFFITH: He's coming over at 2851 1 2:00. And so that we can show him the documents 2 and have an order in place to allow for the 3 procedure, I was trying to get the order signed. 4 Otherwise, I'd have to tell him to wait and I 5 didn't want to do that if we didn't have to. 6 Thank you, Your Honor. 7 MR. NICKENS: May I proceed, Your 8 Honor? 9 THE COURT: Yes, you may, Mr. Nickens. 10 (1:34 p.m.) 11 Q. (BY MR. NICKENS) Mr. Smith, when we 12 took our lunch break, I sort of fast forwarded to 13 some issues. So, let's come back a little bit and 14 see if we can address some other issues. 15 Your professional or educational 16 training is as an economist, not in the area of 17 finance; is that correct? 18 A. That's correct. 19 Q. And your teaching assignments at SMU, 20 was that in the business school? 21 A. Yes, in the business school. I was a 22 member of the finance department. 2852 1 Q. Now, have you ever managed an MBS 2 portfolio? 3 A. Yes. 4 Q. In what context? 5 A. The Federal Home Loan Bank of Dallas 6 MBS portfolio. 7 Q. And how long have you been responsible 8 for that duty? 9 A. Directly and indirectly since 1988. 10 Q. And you also are in charge of some of 11 the Dallas bank's hedging activities? 12 A. Correct. 13 Q. Have you ever managed a risk-controlled 14 arbitrage? 15 A. In the sense of using reverse 16 repurchase agreements to manage levered positions, 17 no. In the sense of managing hedged mortgage 18 positions, yes. 19 Q. Would you consider yourself adequately 20 trained and competent to manage a risk-controlled 21 arbitrage? 22 A. With the proper technology, yes. 2853 1 Q. That is if you had the proper tools? 2 A. The proper tools. 3 Q. And was there ever any question raised 4 about Joe Phillips' ability to manage a 5 risk-controlled arbitrage? 6 A. Not to my knowledge. 7 Q. You personally met with Mr. Phillips, 8 did you not? 9 A. I met with -- the meeting I recall with 10 Mr. Phillips was related to the junk bond 11 portfolio, yes. 12 Q. Yes, sir. But you -- 13 A. Yes, I have met with Mr. Phillips. 14 Q. And you were impressed with his 15 qualifications? 16 A. As a manager of a junk bond portfolio, 17 yes. I thought he seemed to understand all the 18 important pieces of that. 19 Q. And did you ever reach any -- did you 20 ever have any question about his ability to manage 21 a risk-controlled arbitrage? 22 A. I never thought about it in that 2854 1 context. 2 Q. And so, I gather, therefore, that 3 question never arose? 4 A. Not to my knowledge or memory. 5 Q. In fact, you were never asked to look 6 at USAT's risk-controlled arbitrage, were you? 7 A. I do not recall looking at any detailed 8 data about the arbitrage, no. 9 Q. And to your knowledge, was any issue 10 ever raised about the competence of the people 11 there to manage a risk-controlled arbitrage? 12 A. Aside from the meeting we've already 13 discussed when Smith Breeden was released, not to 14 my knowledge. 15 Q. And with regard to that meeting, I 16 gather that you reached the conclusion that their 17 explanation for not using Smith Breeden was 18 adequate? 19 A. It was adequate for my purposes, yes. 20 Q. And, in fact, that was sort of the end 21 of it, wasn't it? 22 A. I did not follow up, no. 2855 1 Q. And there is no known documentation of 2 that meeting? 3 A. There is none from my side, no. 4 Q. And if there had been a concern that 5 had come out of the meeting, we would have seen 6 some documentation? 7 A. I do not know what the regulatory 8 concerns were. I reported back. I did not 9 document anything. 10 Q. Yes, sir. Okay. Your report was 11 positive? 12 A. My report was essentially that the 13 reasons that they gave for changing from Smith 14 Breeden to Walter Mueller seemed reasonable. I 15 did not know Mr. Mueller. I did not know what his 16 competencies were. 17 Q. And did you reach any -- have you ever 18 reached any conclusion on that point? 19 A. Not really. 20 Q. Now, you did meet with Sandy Lawrenson? 21 A. Yes. 22 Q. You know of her background? 2856 1 A. Yes. 2 Q. And you'd have to say that that was a 3 stellar background for managing a risk-controlled 4 arbitrage? 5 A. Sandy had a very strong background and 6 very good work experience for that, yes. 7 Q. She had been hedge manager for Freddie 8 Mac? 9 A. I knew she was in hedging at Freddie. 10 I didn't realize she was manager. 11 Q. A MIT Ph.D. in mathematics? 12 A. Actually, I didn't know she had a Ph.D. 13 Q. Okay. Well, was there ever any 14 question raised about her qualifications? 15 A. Not to my knowledge. 16 Q. Do you know of anybody in the southwest 17 that you could point to at that period of time 18 that you would have said was more qualified to 19 manage a risk-controlled arbitrage? 20 A. Other than myself? 21 Q. Yes, sir. We'll exempt yourself from 22 that. 2857 1 A. In terms of having the background and 2 the training, I think Sandy was as good as there 3 was in this area. 4 Q. Thank you. Now, who is your -- or 5 who -- when you came to work at the Dallas bank, 6 who was your immediate supervisor? 7 A. John Scott. 8 Q. And tell the Court something about Mr. 9 Scott's background. 10 A. John was a Ph.D. in finance, I believe, 11 or economics from Perdue. John and I had talked 12 taught together at SMU for I guess a year and a 13 half before he left to join the Home Loan Bank, 14 and I believe he was vice president of financial 15 management when he hired me. 16 Q. And you worked closely with Mr. Scott? 17 A. John and I worked closely, very closely 18 for about two and a half years. 19 Q. And where is Mr. Scott today? 20 A. He's a professor at Temple University 21 in Philadelphia. 22 Q. When did he leave the bank? 2858 1 A. I'm guessing '91, somewhere in the 2 early Nineties. 3 Q. So, he was at the bank along with you 4 during this entire time frame involving USAT. And 5 that is, you said you started in January of '86? 6 A. '86, yes. 7 Q. And both of you were there from that 8 time period through December of '88? 9 A. Yes, we were. 10 Q. Now, USAT wasn't the only institution 11 in the southwest using risk-controlled arbitrage, 12 was it? 13 A. I don't believe they were. 14 Q. Well, are you aware of others? 15 A. I cannot -- I can't tell you names of 16 institutions but I'm -- I know that I was aware 17 that there were others using this risk-controlled 18 arbitrage technology, yes. 19 Q. And, in fact, basically all of the 20 large Texas institutions were using 21 risk-controlled arbitrage, were they not? 22 A. I don't know that I can confirm that 2859 1 blanket of a statement. 2 Q. Well, let me ask you -- I asked you 3 about this earlier and it should be there. 4 Exhibit B823 which is the Donnelly article. You 5 see on the right side column at the very bottom, 6 it's a carry-over paragraph? 7 A. Uh-huh. (Witness nods head 8 affirmatively.) 9 Q. "Among the practitioners of 10 risk-controlled arbitrage which include a big 11 percentage of the top 100 thrifts and almost all 12 of the large Texas S&Ls" -- and it goes on to talk 13 about Franklin Savings. 14 A. Okay. 15 Q. Would you be able to dispute the 16 conclusion reflected in B823? 17 A. I think "almost all" is an adequate 18 hedge. I can't dispute this. 19 Q. But it certainly wasn't unusual or rare 20 among the larger thrifts at that point in time? 21 A. That's correct. 22 Q. And tell the Court why that was. Why 2860 1 was risk-controlled arbitrage so attractive? 2 A. It was originally presented and I 3 believe sold by the Wall Street firms as a way to 4 effectively lever your capital, increase your 5 earning streams, and mitigate the risk of the 6 levered mortgage positions. 7 Q. You could actually lower your -- or 8 lessen your gap position through risk-controlled 9 arbitrage. Right? 10 A. Potentially, yes. 11 Q. And to the extent that you could 12 substitute MBS for whole loan, that was thought to 13 be a good development, wasn't it? 14 A. That was perceived to substantially 15 lessen the credit risk which was a positive, yes. 16 Q. As well as improve your liquidity 17 position? 18 A. And would improve the liquidity, that's 19 correct. 20 Q. So the bank didn't ever oppose even 21 just the substitution of MBS for whole loans? 22 A. I don't know that the bank ever did 2861 1 oppose that. 2 Q. And you mentioned Wall Street. If you 3 could turn over to the second page of B823 there 4 on the left side, it's the second full 5 paragraph -- the third paragraph actually. It 6 starts out "oddly enough." 7 A. Got it. 8 Q. Okay. "Oddly enough, this strategy 9 which can be implemented only with the aid of 10 highly sophisticated duration analysis and 11 optimization programs didn't originate among Wall 12 Street fixed-income brain trusts but at the 13 thrifts themselves." 14 Do you see that? 15 A. Uh-huh. 16 Q. Now, you have some contrary 17 information? 18 A. I don't know where it started. I only 19 know where I learned about it. 20 Q. And it caught on at least among the 21 investment banks as a product? 22 A. Yes. 2862 1 Q. Partly because it was so popular among 2 the thrifts. Right? 3 A. It was very popular among the thrifts, 4 and it was a source of trading profits for the 5 institutions. 6 Q. And that was part of the reason that 7 you were writing four articles on it in 1987 and 8 1988? 9 A. Correct. 10 Q. Now, let me ask you to take a look at 11 Exhibit 2684. Are you familiar with the 12 publication The American Bankers? 13 A. Yes. 14 Q. And is that a well thought of 15 publication? 16 A. Yes. 17 Q. One that you rely on in your business? 18 A. I don't read it, but several people in 19 our business do. 20 MR. NICKENS: Your Honor, we offer 21 B2684. 22 MR. GUIDO: No objection, Your Honor. 2863 1 THE COURT: Received. 2 Q. (BY MR. NICKENS) Now, B2684 purports 3 to show the top 100 thrifts and mortgage-backed 4 securities holdings as of June 30th, 1988. And 5 can you look down there and see at No. 39, United 6 Savings Association of Houston? 7 A. Yes. 8 Q. And do you see above at No. 30, the 9 Bright Banc, S.A. Dallas? 10 A. Yes. 11 Q. And at No. 25, Gibraltar Savings 12 Association of Houston? 13 A. Yes. 14 Q. And at No. 11, Guardian Savings and 15 Loan-Houston? 16 A. Yes. 17 Q. Now, looking at these numbers, can you 18 say that the mortgage-backed securities holdings 19 of United Savings were out of line in any way? 20 A. As mortgage-backed as a percent of 21 assets on here? 22 Q. Yes, it is. It's in the column about 2864 1 five over, five or six. It says 26.09 for United 2 Savings. 3 A. Okay. Since you're doing so well and I 4 can't read this, would you like to read me the 5 others and I'll see if we're in the -- 6 Q. Okay. Here. Try my copy. 7 A. Highlights would be helpful, yes. 26, 8 32,25, 87. It looks like it's in the zone with 9 certainly Gibraltar and Bright Banc and below 10 Guardian. 11 Q. Nothing unusual from examining the 12 percentage of assets? 13 A. No. You would expect a thrift to have 14 a lot of mortgage assets. 15 Q. And hopefully have those as opposed in 16 this situation to whole loans? 17 A. Depending on what strategies they were 18 following, yes. Securities would -- could be much 19 better than whole loans. 20 Q. Now -- I took this away. The very top 21 one is American Savings and Loan Association of 22 Stockton, California? 2865 1 A. Right. 2 Q. Are you familiar with that institution? 3 A. By reputation only. They were a large 4 California thrift. 5 Q. And was the Federal Home Loan Bank 6 involved or the regulators involved in running 7 that bank? 8 A. At one point, it was run out of the San 9 Francisco bank. There was a large merger out 10 there, but I really don't know the details. 11 Q. And that was Mr. Popejoy? 12 A. Name's familiar. 13 Q. But in any event, there wasn't any 14 regulatory -- well, let me ask you this question. 15 What were the regulatory concerns with 16 risk-controlled arbitrage? 17 A. I think you'd probably have to ask a 18 regulator to get a very precise answer for that. 19 I believe the concerns would have been along the 20 lines of adequate management, adequate capital to 21 absorb any risk that's in the position. 22 Q. And adequate capital was defined in two 2866 1 ways: By regulation and by the marketplace, 2 correct? 3 A. That's generally, yes. 4 Q. Would you know of anything else? 5 A. One is going to be binding at some 6 times and one will be binding at others. 7 Q. And, in fact, Mr. Hjerpe -- is that the 8 correct pronunciation? 9 A. Yes. 10 Q. -- has looked at that very issue, 11 hasn't he? 12 A. In the -- would you like to point me to 13 where before I -- 14 Q. Okay. I'm going to hand you 15 Exhibit B1619 -- 16 A. Okay. 17 Q. -- which purports to be a research 18 working paper No. 128 of the Federal Home Loan 19 Bank Board dated May 11th, 1987. 20 MR. NICKENS: Your Honor, we offer 21 B1619. 22 MR. GUIDO: No objection, Your Honor. 2867 1 THE COURT: Received. 2 Q. (BY MR. NICKENS) Mr. Smith, let me 3 ask you to look over at Page 6 which has Figure 2. 4 MR. GUIDO: Page 6? 5 A. I don't have a Page 6. 6 MR. GUIDO: Me either. 7 A. I have a Page 5 and a Page 7. 8 MR. NICKENS: Your Honor, I recall that 9 this document is one where I think, at one time, 10 every other page was copied and apparently I got a 11 complete copy but everybody else didn't. 12 A. And we have the odds. 13 Q. (BY MR. NICKENS) Let me ask you if -- 14 MR. NICKENS: And we'll supply the 15 pages, Your Honor, so it's complete. 16 Q. (BY MR. NICKENS) But let me ask you 17 to look at Page 6, Figure 2? 18 MR. GUIDO: Your Honor, I object to the 19 use of a incomplete document. There is no way 20 that I can prepare for redirect based on the 21 incomplete document. 22 MR. NICKENS: Okay, Your Honor. We can 2868 1 make the copies right now. 2 THE COURT: All right. We'll take a 3 minute. We'll be off the record. 4 5 (A break was taken.) 6 7 THE COURT: All right. We'll be back 8 on the record. 9 Q. (BY MR. NICKENS) Mr. Smith, while 10 we're waiting for the document to be copied, would 11 it be your conclusion that it was the Federal Home 12 Loan Bank's policy that risk-controlled arbitrage 13 was an appropriate investment strategy for thrifts 14 in the nineteen -- in the mid-1980s? 15 A. I think it's documented that there are 16 certain circumstances, if properly managed, where 17 risk-controlled arbitrages could enhance the 18 earnings and profitability fairly safely of a 19 thrift institution, yes. 20 Q. And part of the reason for the 21 popularity of it had to do with new net worth 22 credits that the bank put in place in this time 2869 1 frame; isn't that correct? 2 A. I vaguely remember that there was 3 regulatory changes on net worth credits, but I 4 don't recall the details. 5 Q. The maturity matching credits? 6 A. There were -- yes, I do recall the 7 maturity matching credits. 8 Q. And that would be very well suited for 9 a risk-controlled arbitrage. Right? 10 A. Incremental growth through a 11 risk-controlled arbitrage would generally make the 12 gap positions of a savings institution better and 13 help them qualify for that, yes. 14 Q. Now, was it also recommended for 15 problem institutions? 16 A. There were occasions where institutions 17 were looking at or were headed toward potentially 18 not meeting their risk capital requirements or 19 regulatory capital requirements and those 20 institutions who had requested risk-controlled 21 arbitrage authority, I believe, in many cases were 22 granted that. 2870 1 Q. So, is the conclusion that it was 2 recommended for problem institutions? 3 A. And I guess -- we may be trying to 4 split a hair here. I don't think that it -- I 5 don't know that it was recommended for all problem 6 institutions. I believe there were a subset of 7 problem institutions in certain circumstances 8 where it was considered acceptable by the 9 regulatory group. 10 Q. And that was in part because it was 11 considered to be safer than other investment 12 alternatives? 13 A. Depending on the structure of the 14 transaction, that's correct. 15 Q. Now, let's talk about the purpose of a 16 risk-controlled arbitrage for a few minutes. The 17 purpose, is it not, is to create spread income 18 based upon the difference between the coupon of 19 the mortgage-backed securities and the cost -- the 20 hedged cost of funds? 21 A. Yes. 22 Q. And it was thought at the time that the 2871 1 profit was very durable for the risk-controlled 2 arbitrage over a fairly large range of interest 3 rate scenarios? 4 A. It was thought that structures could be 5 created that would give you that kind of 6 protection, yes. 7 Q. And that was what you've referred to as 8 scenario analysis? 9 A. Correct. 10 Q. And the common scenario for the time 11 period was down 200 basis points and up 200 basis 12 points, was it not? 13 A. Those were generally the limits that 14 you set. There were usually a couple in between 15 that you would run, also. 16 Q. In other words -- 17 A. Plus and minus 100, plus and minus 200 18 were the traditional modeling scenarios. 19 Q. And -- now, what is the modeling today 20 that is prevalent? 21 A. The modeling is -- the transactions 22 today are being packaged somewhat differently so 2872 1 the modeling -- it's hard to kind of give you an 2 apples to apples comparison. But generally if 3 you're looking at hedging mortgage securities, 4 you're using much more -- many more scenarios, 5 rotating the yield curve as opposed to just 6 parallel shifts of the yield curve. And in terms 7 of evaluating the risk in the mortgage securities, 8 you may or may not be using something called 9 option adjusted spread technology. 10 Q. Well, that just wasn't available in 11 1985 or '86, was it? 12 A. No. '87 is when OAS came on. Option 13 adjusted spread was, I believe, '87. 14 Q. And back in 1985, were there any, to 15 your knowledge, prepayment models available? 16 A. Well, there certainly were in January 17 of '86 when I got to the Home Loan Bank, and I 18 assume that that means they were built and 19 available in '85. 20 Q. Well, wasn't it the case generally in 21 1985 and 1986 that people used yield tables for 22 looking at prepayments? 2873 1 A. Some people were probably still using 2 yield tables. What we were seeing in '86 3 certainly coming out of Wall Street were 4 econometric models trying to explain and then 5 predict prepayment behavior based on lots of 6 factors. 7 Q. Okay. Is risk-controlled arbitrage a 8 cash flow model or a total return model? 9 A. It's presented a lot of different ways. 10 The total return model -- let's make sure we're 11 using the same vocabulary before I go forward. 12 For a total return model, what you're looking at 13 generally is put the transaction on, measure your 14 cash flows across the horizon of the transaction, 15 and at the end, terminate the transaction. 16 Calculate all your gains and losses and look at 17 that, look at the total return of all those gains 18 and losses divided by the original capital that 19 you put in the transaction. That's total return. 20 And that is what we were certainly recommending, 21 was a market value total return basis. 22 The cash flow analysis implicitly 2874 1 assumes that you're going to sell the securities 2 at the same price you bought them less accretions 3 over the time period. People did look at the cash 4 flow. We were -- in fact, in my articles, I think 5 we talk about using market value which would be 6 closer to a total return concept. 7 Q. And do you know one way or another as 8 to what tools USAT used for managing its 9 risk-controlled -- 10 A. No, I don't. 11 Q. -- arbitrage? So, you can't provide 12 any information as to whether or not they were 13 using the tools of the time or whether they were 14 or not using the tools of the time? 15 A. I do not recall ever having information 16 about what they were using. 17 Q. Let's talk a few minutes about the 18 decisions that a person whose putting a 19 risk-controlled arbitrage together has to make and 20 the considerations that go into it. And let's, if 21 we might, focus on the asset side first. 22 A. Okay. 2875 1 Q. The manager has to make a decision 2 about what assets to choose. Right? 3 A. Correct. 4 Q. And there are a variety of choices out 5 there, but let's focus on mortgage-backed 6 securities. And we're before the CMO market comes 7 along, correct? 8 A. It was in its very, very early stages 9 in '87. So certainly in '85, you were. 10 Q. So, what we're basically looking at in 11 that area were the pass-through certificates. 12 Right? 13 A. Yes. 14 Q. And so, the decision maker has to 15 decide whether to pick premium securities, current 16 coupon securities, or discount securities 17 essentially; is that correct? 18 A. Yes, but I think you're deconstructing 19 the way you put this together somewhat 20 artificially. 21 Q. Well, what do you mean? 22 A. You can't pick the asset without 2876 1 looking at the liability -- the ability to 2 structure the liabilities because if you're really 3 looking at a total return scenario, as you change 4 coupons on the mortgages, you're also going to 5 have to restructure your liabilities and those are 6 all going to generate different spreads. So, it 7 is a simultaneous solution, in a sense, that you 8 have to look at both sides. 9 So, we can go forward talking about 10 asset choice but I think it's very important to 11 understand that each asset may have specific 12 liability structures that are optimal that will 13 give different spread relationships. 14 Q. Well -- okay. Let's see if I can 15 follow that. You may pick an asset and think it's 16 good and then when you look at the liability side, 17 you decide, "Well, I better try something else"? 18 A. When you jam it in the model, what you 19 may find is that to meet your risk preferences, 20 you would have an unacceptable spread. 21 Q. Well, let's, if we can, focus on this 22 choice between premium, current, or discount. If 2877 1 you choose premium coupon mortgages, your spread 2 is going to be greater typically. Right? 3 A. Typically, right, unless they are very, 4 very premium. 5 Q. Meaning that they are prepared to be 6 paid off tomorrow? 7 A. Right. You expect them to last about 8 an hour and a half. 9 Q. Okay. So, the -- if you were just 10 going for spread outside of any other 11 consideration, you normally would pick a premium 12 coupon mortgage? 13 A. If you're looking at a spread to some 14 sort of reference security, those are going to 15 have the highest spread, yes. 16 Q. And the reference security is normally 17 treasuries? 18 A. Treasuries. 19 Q. Okay. So -- and let's be clear what we 20 mean by "premium." Premium security would be one 21 where the coupon is above the current market 22 coupon. Right? 2878 1 A. Yes. 2 Q. So that if current market is 12, you 3 can go out there and buy a 14 but you have to pay 4 a premium for it? 5 A. Right, because you have to get back to 6 an expected 12 percent yield. 7 Q. And so -- and then opposed to that is 8 the current coupon and the discount coupon. 9 Right? 10 A. Yes. 11 Q. And the exact example, if you've got 12 12 percent, you can go out there and buy a 13 10 percent paper and you would normally buy that 14 at a discount? 15 A. Yes. 16 Q. So, that's what we call discount MBS. 17 And explain to the Court the advantages vis-a-vis 18 interest rate risk and prepayment risk about this 19 choice between discount, premium, or current 20 coupon. 21 A. Okay. Let's start with discount and 22 build up from there. If you have a security that 2879 1 has 5 percent -- or let's stay with your 2 12 percent example for a minute. Right now if 3 mortgage rates are 12 percent and you have a 4 security that's a pass-through of a pool of 5 mortgages with 10 percent payments on them, 6 10 percent coupons on them, then you would expect 7 that to be very insensitive to rate changes from a 8 purely prepayment driven perspective. Rates would 9 have to fall dramatically before anyone with a 10 10 percent mortgage could be convinced to 11 refinance into a 12 percent mortgage unless they 12 were moving or something like that. So, you 13 expect the prepayments to be generally low and you 14 expect them to be fairly stable for small changes 15 in the current coupon rate around 12. 16 At 12 percent, you have roughly the 17 same story as you do with discounts but you're 18 getting closer to the point rates have to move a 19 smaller amount downward before those people would 20 refinance out. So, at 12, for example, rates may 21 have to fall to 10 and a half before you would 22 start to see huge increases in prepayments on the 2880 1 12s. But even down 150 for the 12s to start 2 prepaying, it still wouldn't be economic for the 3 10s to prepay, if I did my math right. So, you 4 would expect the current coupon to have a little 5 more prepayment risk than the deep discounts. And 6 as you move upward from current coupon, you're 7 going to start to trigger prepayments if rates 8 drop because people are going to have incentives 9 to refinance out of 11 percent or 12 percent down 10 depending on how much longer their mortgage has to 11 run and what the cost of going through the 12 refinancing process is. 13 So, you would expect somewhere at this 14 point in time 125, 150ish out of the money, those 15 mortgages -- those securities would have a 16 tremendous amount or they would have more 17 prepayment volatility, meaning if rates move from 18 10 to 9.75 or 9.50, the 11 percent securities 19 would start to prepay and the 12 percents would be 20 prepaying even faster. 21 Q. So, the discounts offer you some 22 protection against prepayments? 2881 1 A. Against prepayments. 2 Q. But the cost is that you reduce your 3 spread? 4 A. That's right. Risk return. Low risk, 5 low return. 6 Q. I mean, there is nothing -- these -- 7 this investment strategy is like most in that you 8 are paid to take a risk? 9 A. That's correct. 10 Q. And the manager has to decide whether 11 or not that risk is worth taking? 12 A. He has to decide if that risk is worth 13 taking and he has to also measure that against the 14 amount of capital he has to put at risk. 15 Q. And what decision did USAT make in 16 setting up Joe's portfolio? 17 A. I really don't know. 18 Q. Do you have any information that they 19 didn't take the most conservative approach? 20 A. If I did, I sure don't remember it 21 anymore. 22 Q. And among those choices, the more 2882 1 conservative approach would be the current coupon 2 or discount coupon -- 3 A. Those are generally thought to be more 4 conservative, yes. 5 Q. Now, let's -- in addition to how much 6 or which security to pick, he must decide how much 7 to pick. Right? 8 A. Yes. 9 Q. And how does he go about making that 10 decision? 11 A. Again, the -- he has to decide what 12 degree of leverage he's comfortable with given his 13 capital position. And how much -- he has a couple 14 of binding constraints. One binding constraint is 15 the regulatory capital constraint that he cannot 16 be levered more than the regulators allow him to. 17 The other is that if credit markets, the firms who 18 are providing the credit, are concerned about the 19 degree of leverage or concerned about what they 20 perceive the risk on the balance sheet to be, they 21 may cut him short before he hits the regulatory 22 limit. So, those are the two binding constraints 2883 1 he has. 2 The other constraint is again the risk 3 return tradeoff, how levered does he have to get 4 the transaction in order to earn the returns that 5 he thinks are appropriate. 6 Q. Okay. Now, let me bring you back, 7 since we've got it now, to Mr. Hjerpe's paper. 8 MR. NICKENS: Your Honor, we now have 9 provided hopefully everybody with a complete copy 10 of B1619, and I offer it at this time. 11 MR. GUIDO: No objection, Your Honor. 12 THE COURT: Received. 13 Q. (BY MR. NICKENS) Now, Mr. Smith, we 14 can go over to Page 6, Figure 2. Do you see Mr. 15 Hjerpe's doing an analysis about various amounts 16 of the capital requirement or the leverage? 17 A. I see that's what that is. 18 Q. And he defines "prudent liquidity" as 19 being somewhere between approximately five to 20 about seven and a half multiple. That's Item D. 21 A. Oh, okay. Down in the bottom right 22 corner? 2884 1 Q. Yes, sir. 2 A. Okay. 3 Q. Is that -- 4 A. I see where he is. 5 Q. He's got that word "prudent" and he 6 indicates that it's somewhere, you know, five to 7 seven and a half approximately. Right? 8 A. I don't see the five to seven and a 9 half. I'm sorry. 10 Q. Well, 4.8 to 7.4. 11 A. Then I am looking in the wrong -- oh, 12 okay. Are you talking about the leverage 13 multiples? 14 Q. Yes, sir. 15 A. Okay. I thought you were looking at 16 something different. Okay. 4.8 to 7.4. 17 Q. And he gets there by taking a 5 percent 18 hair cut that would be the market discipline, 19 correct? 20 A. Correct. 21 Q. And then 4 percent capital requirement 22 at that point in time with the Federal Home Loan 2885 1 Bank Board, and he's got a little note down here 2 indicating that he's assuming that this 3 institution is able to take the full maturity 4 matching credit. Right? 5 A. Yes. 6 Q. And then he's got an estimate of 7 hedging costs and then he puts in what he calls 3 8 to 10 percent elective liquidity? 9 A. Uh-huh. 10 Q. And then from that, he derives a 11 leverage multiple for what he calls prudent 12 liquidity at between 4.8 to 7.4. Do you see that? 13 A. I see the analysis, yes. 14 Q. Now, do you agree or disagree with that 15 conclusion if that is the conclusion? 16 A. Let me take a minute and read the 17 elective liquidity paragraph at the bottom and 18 then -- 19 Q. Okay. 20 THE COURT: Mr. Nickens, the copy that 21 I've got in my hand does not have the exhibit 22 number on it. I'll just call that to your 2886 1 attention. The reporter's copy should be marked. 2 MR. NICKENS: We'll make that 3 arrangement, Your Honor. Mine's got this yellow 4 sticky, and I don't see any other mark on it. So 5 we will make sure that we get a page with the 6 appropriate number on it. 7 A. Okay. I read the paragraph. Would you 8 repeat your question, please? 9 Q. (BY MR. NICKENS) Would you agree or 10 disagree with his conclusion here with regard to 11 what he has said is prudent liquidity? 12 A. Certainly the 5 percent, 4 percent, one 13 and a half percent look very reasonable. The 3 to 14 10 percent, I really don't have much insight into 15 why he picked those numbers. I think the ultimate 16 leverages that you come out with are probably 17 about right, the 4.8 to 7.4. 18 Q. And, in fact, if you could look at the 19 next page, you'll see that he indicates that there 20 are -- and this is on Page 7 at the bottom of the 21 first full paragraph. "The reciprocal of 10 and a 22 half percent or 9.5 percent is the number of 2887 1 multiples of the original investment that can be 2 attained for an institution meeting these 3 criteria," and he's talking about a Scenario B 4 institution. 5 A. And Scenario B is referring to the 6 previous page or -- 7 Q. Well, let's see. It's a 4 percent 8 regulatory capital requirement. 9 A. I was going to say I can agree that 10 that's what the paragraph says. I just don't know 11 what Scenario B -- 12 Q. Okay. Fair enough. Now, this article 13 was dated May of '87. Let me ask you to look over 14 at Page 9. 15 Would you agree with him or disagree 16 with him that there was -- and I'm reading at the 17 bottom of the page -- a lack of consensus as to 18 the appropriate theoretical and practical models 19 for pricing the securities at that point in time? 20 A. Are you on that last paragraph? 21 Q. Yes, sir. 22 A. Okay. Hang on. Let me read the 2888 1 paragraph. 2 Q. Okay. 3 A. I pretty much have to agree with that 4 because that's what actually happened. 5 Q. Well, I want to focus on the time 6 period. In May of 1987, what was being widely 7 used were static models? 8 A. Discounted cash flow models based on 9 particular prepayment speeds. 10 Q. And the option adjusted trading models 11 which later came to dominate the market were not 12 in widespread use? 13 A. In what year were you -- 14 Q. We're talking about May of '87. 15 A. '87. They were just starting to come 16 on-line in May of '87. 17 Q. And he was noting the fact that there 18 is a lack of consensus? 19 A. Depending on how -- in a static 20 discounted cash flow model, the art of getting the 21 right answer is picking the right prepayment 22 speed. And if you have a difference of opinion 2889 1 about the prepayment speed, you're going to get a 2 different price for the security out of those 3 models. 4 Q. And the -- that's true for either 5 model -- static model or the option adjusted 6 spread models. Right? 7 A. That's right. The option adjusted 8 spread models, because they run so many more 9 scenarios, will average that out a little bit. 10 Q. It wouldn't be a fair criticism of USAT 11 to say to them that in 1985 they weren't using 12 option adjusted spread models? 13 A. I don't think you can criticize someone 14 for using something that hasn't been invented. 15 Q. I can agree with that. And, indeed, 16 there has continued even today to be debate over 17 whether option adjusted spread models have 18 significant problems? 19 A. OAS models, option adjusted spread 20 models, still error. And, in fact, if you look at 21 mortgage securities today, you will generally get 22 an option adjusted spread and you will also 2890 1 generally get scenario analysis just like we were 2 running back in '86 and '87. 3 Q. Let me ask you to look over to Page 13 4 of Exhibit 1619. At the very top of the page, 5 he's talking about the risk-controlled arbitrage 6 and he says, "That goal" -- and go back a page -- 7 "Hedging the value of the assets or the cost of 8 the liabilities can both accomplish the same goal. 9 That goal is to lock in a rate of return on the 10 assets over the cost of funding for the life of 11 the asset regardless of changes in the interest 12 rate environment." 13 Do you see that? 14 A. Yes. 15 Q. Do you agree with that? 16 A. Can I have a running start into it? 17 Q. Of course. Of course. 18 MR. GUIDO: Does this start on 12? Is 19 that it? 20 MR. NICKENS: Starting where you want. 21 A. I was starting at "financial futures" 22 on the middle of Page 12 just to get a sense for 2891 1 what that paragraph is. 2 What Mr. Hjerpe, I believe, is 3 referring to here is that whether you think of the 4 interest rate swaps as swapping the mortgage 5 assets back to a short-term asset or if you think 6 of them as swapping a short-term liability to a 7 long-term liability, you get the same overall 8 answer. And I agree with that. You're trying to 9 hedge a spread and you're attaching the hedge to 10 one or the other. 11 Q. (BY MR. NICKENS) And the goal is to 12 lock in a rate of return on the assets over the 13 cost of funding for the life of the asset 14 regardless of changes in the interest rate 15 environment? 16 A. That is the way he states the goal. 17 I'm not sure that is the goal in the sense that 18 it's not clear to me that you can lock in a rate 19 of return on assets regardless of changes in the 20 interest rate environment, that there are 21 potential interest rate environments that will 22 blow the hedges apart. 2892 1 Q. And he would expect -- well, one 2 schooled in this area would know that. Right? 3 A. They should. 4 Q. It violates the basic principle you've 5 told us about about having to -- in order to get a 6 return, you have to take a risk. Right? 7 A. That's correct. And I think the 8 question here -- and you probably need to ask Mr. 9 Hjerpe -- but it seems to me that if the goal is 10 to lock in a rate of return -- and it's not clear 11 what kind of rate of return he's talking about -- 12 regardless of changes in the interest rate 13 environment, that's an admirable goal but one that 14 probably can't be achieved and leave any spread in 15 the transaction. 16 Q. And everybody in this business would 17 understand that? 18 A. I believe that's correct. 19 Q. If someone came to you and said, "I 20 have found a risk-free way of making money," you'd 21 tell them to go back home and try again? 22 A. I would ask them why they are telling 2893 1 me about it instead of using it themselves. 2 Q. Okay. If there was a risk-free way of 3 making money, everybody would do it? 4 A. That would be the assumption. 5 Q. And the people at the bank board in 6 1985 knew that, didn't they? 7 A. I believe they did. 8 Q. Now, are you aware of the fact that one 9 of the claims in this case is that we misled the 10 bank board by telling them that we had hoped to 11 lock in a spread in the risk-controlled arbitrage? 12 A. No. 13 Q. Would you have been misled by such a 14 statement into thinking that we were telling the 15 bank board that we had discovered a risk-free way 16 of making money? 17 A. It would depend on how you phrased it. 18 If you said you had hoped to lock in a spread, we 19 would all have hoped to lock in a spread. I mean, 20 I don't know -- 21 Q. I'm going to show you the statement 22 before we're finished, Mr. Smith. I'm not going 2894 1 to -- but let's say that someone had come in and 2 said, "I have found a way to lock in a spread 3 regardless of interest rate environments." 4 Would you have been fooled by that 5 statement? 6 A. I would have assumed they didn't know 7 what they were doing. 8 Q. And you would have asked questions 9 about what they were doing? 10 A. Either that or I would have -- 11 certainly if they were in a position to execute 12 trades, I probably would have asked. 13 Q. And in general, you would understand 14 such a statement to mean that over certain 15 interest rate scenarios, you have locked in a 16 spread? 17 A. That is what I would have assumed they 18 were talking about. 19 Q. And the scenarios that were prevalent 20 for risk-controlled arbitrages in 1985 and '86 21 were 200 basis points up and down? 22 A. That is what the market standard was 2895 1 then. 2 Q. Why do you use mortgage-backed 3 securities instead of treasuries? 4 A. Because you'd like to have a positive 5 spread out of your transaction. You cannot -- 6 Q. Couldn't get a spread with treasuries. 7 Right? 8 A. You can't because the swaps, the 9 liability cost is above the treasury rate. So if 10 you're starting off the liability cost above the 11 treasury rate and you're adding treasuries, then 12 you can't get a positive spread. 13 Q. And the reason you get -- 14 A. Now -- 15 Q. I'm sorry. Go ahead? 16 A. In terms of a true risk-controlled 17 arbitrage with hedges. You can lever treasuries 18 without hedges and get a positive spread. 19 Q. Now -- okay. We've talked about the 20 choice of the assets. We've talked about the 21 choice about how much and the 5 to 7 and a half 22 times being at least one area that was defined as 2896 1 prudent at the time, correct? 2 A. Correct. 3 Q. And now let's talk about the liability 4 side. 5 What was the typical hedging instrument 6 used for these types of scenarios in 1986? 7 A. Typically, they were doing reverse 8 repurchase transactions and interest rate swaps. 9 There were a few institutions that were using 10 futures and options. There were a few 11 institutions that were using interest rate caps. 12 But predominantly, it was either reverse repo or 13 deposits with interest rate swaps. 14 Q. Now, why was that? Why swaps over caps 15 or futures? 16 A. Well, certainly over future swaps are 17 much easier to manage. Futures require a level of 18 day-to-day margin management and hedge management 19 that is subsumed in the interest rate swap. 20 Essentially with the interest rate swap, you have 21 done a futures transaction and you don't have all 22 the contracts to manage. 2897 1 Q. Do you manage your -- or hedge your 2 swap positions through the use of futures? 3 A. At the Home Loan Bank? 4 Q. Yes, sir. 5 A. We actually did for a little while. We 6 do not anymore. 7 Q. All right. But frequently, the 8 counter-party on these swaps were investment 9 bankers. Right? 10 A. Correct. 11 Q. Or financial institutions like the 12 bank? 13 A. Uh-huh. 14 Q. Right? 15 A. Right. 16 Q. And those institutions would hedge 17 their swap position by going into the futures 18 markets? 19 A. They would either do that or attempt to 20 do an offsetting swap with another counter-party. 21 Q. And some people with sufficient level 22 of sophistication sort of skipped that middle step 2898 1 and went into the futures market rather than 2 paying the cost of the swap? 3 A. Well, let's also -- let's get the time 4 frames clean. 5 Q. Yes, sir. 6 A. In certainly the '85 time frame and 7 before the swap market was very, very young, it 8 was just beginning. It was not well understood, 9 and we -- I think the Home Loan Bank of Dallas did 10 its first swap in 1985 and we were one of the 11 first large swap counter-parties anywhere, the 12 Home Loan Bank system was. 13 By '86, very late '85 and '86, you were 14 seeing more and more use of swaps. But prior to 15 early '85, you really didn't have a swap market 16 and you had to use futures and options. So, 17 let -- so, give me the time frame and I'll try to 18 give you -- 19 Q. And the so-called swaption didn't exist 20 in '85 and '86, did it? 21 A. Swaptions? 22 Q. Yes. 2899 1 A. Not to my knowledge. Not in '85 and 2 '86. 3 Q. Now, let's explain to the Court what a 4 swaption is. If you have -- this swap agreement 5 is for a specified time period, isn't it? 6 A. Yes. 7 THE COURT: How do you spell it? 8 MR. NICKENS: S-w-a-p-t-i-o-n, Your 9 Honor. 10 THE COURT: Thank you. 11 Q. (BY MR. NICKENS) And that was -- you 12 were stuck with that agreement unless you could 13 buy your way out of it? 14 A. Or you did a mark to market of that 15 transaction just like it was a treasury. 16 Q. You'd have to go to the counter-party 17 and say, "I want out of this agreement" and you'd 18 have to pay that person some money to get out of 19 it? 20 A. That is how swaps get terminated. 21 Q. All right. And that turned out to be 22 sort of an unsatisfactory situation, did it not? 2900 1 A. You could build up very large 2 positions, but it's the same position you would 3 have had in futures -- 4 Q. In order to -- 5 A. -- in order to get -- 6 Q. -- hedge it? 7 A. Yes. If you had a swap that had a very 8 large payment attached to terminating it because 9 it was out -- it was in the money, you would have 10 had the same loss in the futures. 11 Q. And a new product was developed 12 sometime in the late Eighties called the swaption 13 to allow people to have an option beforehand to 14 get out of the swap? 15 A. What they did was to put a call option 16 into a swap agreement where one counter-party or 17 the other would have the opportunity quarterly, 18 semi annually, or on some basis to terminate the 19 swap at no payment and there was a higher coupon 20 up front, obviously, to compensate for that. 21 Q. But that didn't exist in the time frame 22 we're talking about? 2901 1 A. '85, '86, early '87, I don't remember 2 swaptions being issued. 3 Q. So, when you were telling the Court 4 that what they should have done in the roll-down 5 was to take the funds and buy out of the swap, 6 that was a very expensive proposition, wasn't it? 7 A. Not necessarily. If they have a gain 8 in the mortgage assets that they have just sold 9 and they have to then turn around and use those to 10 pay the loss in the swap, it may be zero. 11 Q. But you have to pay your counter-party 12 what he thinks it's worth because he can -- 13 A. But there is a market out there that 14 you can transfer the swaps so if you have a 15 particular counter-party who is trying to hold you 16 up, then find another counter-party and transfer 17 the swap. That was being done in '86 and '87. 18 Q. Well, wasn't the most frequent way to 19 get out of these swaps before the invention, if 20 you will, of the swaption the so-called mirror 21 swap? 22 A. Mirror swaps were done. 2902 1 Q. And explain to the Court what a mirror 2 swap was. 3 A. There are -- there is a couple of ways 4 to do mirror swaps, and I'll try to explain both. 5 If you're paying the counter-party 10 percent and 6 currently market rates would be 8 percent, then 7 you could do another swap to receive 8 percent, 8 the current market, and lock in the 200 basis 9 point loss. 10 So, you would essentially be paying 11 10 percent to the counter-party, receiving 8, 12 which is a net 2, and the short-term indices LIBOR 13 or bills would offset. So, you've locked in a 14 200 basis point loss. That's one way to do mirror 15 swaps. 16 The other way to do mirror swaps is to 17 essentially write a check for the expected -- for 18 the present value of that 200 basis point 19 difference to the counter-party and then he would 20 give you a new swap at 10 percent and everything 21 would completely offset. 22 So, if you were unable to terminate the 2903 1 swap for some reason, mirror swaps were used. And 2 you can do them either way. You can either lock 3 in the loss and take it over time or you can write 4 the check up front, have the swaps cancel out, and 5 essentially be done with it. 6 Q. Let me -- with regard to this point, 7 Mr. Smith, let me ask you to look over to your 8 article at B2055. 9 A. Is that the one with my picture on the 10 front? 11 Q. Yes, sir. And I think we may, in fact, 12 be in Mr. Hjerpe's article. I'd like for you to 13 look at Page 73 which is at F37. 14 A. F37 or 36? 15 Q. 37. 16 A. Got it. 17 Q. Is this a continuation of your article? 18 A. I believe it is. My left-hand margin's 19 cut off, but I think it is a continuation of my 20 article. 21 Q. And this was what you and Mr. Scott 22 wrote in March of 1988, correct? 2904 1 A. Uh-huh. (Witness nods head 2 affirmatively.) 3 Q. And let me refer you to the paragraph 4 that's entitled "Rebalancing the arbitrage." And 5 if we can look down at the third paragraph. 6 MR. GUIDO: What page are you on? 7 MR. NICKENS: It's Page 73 or F -- the 8 Bates number is F37. It's the fourth page of the 9 document. 10 Q. (BY MR. NICKENS) Do you have your 11 place, Mr. Smith? 12 A. I have the "rebalancing the arbitrage" 13 section. 14 Q. And what you wrote in the third 15 paragraph is as follows: "If prepayment speeds 16 change because of significant rate movements, 17 returns will deviate from the expected values. 18 Should rates move down and the mortgage balance 19 begin to prepay, the mortgage securities should be 20 rolled into lower coupons." 21 That's the roll-down. Right? 22 A. Correct. 2905 1 Q. No equivocation. That's what you 2 should do? 3 A. Uh-huh. 4 Q. Right? 5 A. (Witness nods head affirmatively.) 6 Q. You have to answer audibly. 7 A. I'm sorry. Yes. 8 Q. Okay. "The alternative is to partially 9 unwind swaps to reach the desired short funding." 10 You mentioned that possibility, but then you go on 11 to say "The latter rebalancing strategy is much 12 more costly than rolling down coupons because a 13 payment will be required to unwind the swap. By 14 rolling down coupons, the capital gain will help 15 offset the cost of reduced interest income on the 16 assets." 17 Now, does that refresh your 18 recollection that that was your position back in 19 1988? 20 A. That is my position. And I think what 21 I am saying, when you look at that sentence that 22 says "The alternative is to partially unwind 2906 1 swaps," what that's saying is you can either roll 2 down your coupons, which is what you should do if 3 you're expecting rapid prepayment and that is a 4 rebalancing you can do. If you just choose in 5 that environment to not roll down the coupons and 6 to unwind the swaps, you will take a loss and that 7 is probably not the most efficient way to do it. 8 So, rolling down the coupon is an 9 acceptable transaction. The thing you have to be 10 careful about which is, I will agree, embedded in 11 this but subtly, when you roll down the mortgage, 12 your liability structure may not work anymore. 13 And I think that also was understood at that time 14 period. 15 Q. You have to look at the structure, but 16 we can agree that what you told everybody who was 17 prepared to read your writing at that time was 18 that the latter rebalancing strategy is much more 19 costly than rolling down? 20 A. In the absence of rolling down the 21 coupons, that's right. 22 Q. And it would have been a perfectly 2907 1 acceptable means of doing that to mirror the 2 swaps? 3 A. You still took the -- okay. I'm 4 confused. Would you please -- 5 Q. Okay. You've indicated -- and as I 6 understood your direct testimony, you said that 7 they should have bought out of the swaps, in 8 effect. 9 A. Right. 10 Q. And I'm exploring whether those 11 possibilities were real possibilities at the time. 12 A. In mirroring the swaps, if you've set 13 up the same maturity dates, same reset dates, you 14 have economically bought out of the swaps as long 15 as all the dates are matched. You've chosen when 16 you want to take your losses and how you want to 17 do it, but you've essentially taken your losses. 18 Q. And you take them out over time. You 19 pay out over time? 20 A. Depending on which accounting treatment 21 you go for, yes. 22 Q. Now, let's come back to this choice of 2908 1 the hedge. The swap was the most popular choice 2 in this time frame, correct? 3 A. I believe that's correct. 4 Q. Now, why -- why the swap as opposed to 5 a cap or collar? 6 A. Caps and collars were not well 7 understood at that time period. The other issue 8 is that the pricing of caps -- there was also a 9 lot of discussion about what proper cap prices 10 were. And I think swaps were just viewed as a 11 much cleaner, easier, more easily understood way 12 to do this. And I think it was -- it was as much 13 the ability to understand and to explain and to 14 capture how they work as opposed to getting into 15 some esoteric pricing option about what my caps 16 were. 17 Q. And it was also the case that a cap 18 required an up-front payment, didn't it? 19 A. But that payment was generally 20 amortized over the life of the of the cap. So I'm 21 not sure the accounting was. 22 Q. But you would have had to have had 2909 1 cash -- 2 A. You did have to have -- 3 Q. -- to pay the up-front payment? 4 A. That's true. 5 Q. And with the swap, you didn't have such 6 an up-front payment? 7 A. That's right. It was embedded into the 8 spread. 9 Q. But in any event, most people, for 10 whatever reason, chose swaps? 11 A. That's true. 12 Q. And you don't have any criticism of 13 swaps itself as a hedging instrument, do you? 14 A. I do not. 15 Q. Now, another decision that the manager 16 has to make once he's chosen his assets and chosen 17 his liabilities and how much of the assets to 18 have, he has to choose how much liabilities -- how 19 much in the way of swaps, correct? 20 A. He has to choose the term and the 21 amount. 22 Q. Now, if one chooses -- let's give an 2910 1 example. Let's say we've got a hundred million 2 dollars in face amount of the assets. Okay? And 3 let's say that we choose a hundred million dollars 4 notional amount swaps. Are you with me? 5 A. Okay. They are the same. 6 Q. Okay. The same number. And you are 7 there fully hedged, correct? 8 A. Not necessarily. 9 Q. Isn't that the way that term was used 10 at that time? 11 A. Not with mortgage securities. 12 Q. Well, what would "fully hedged" mean? 13 A. Fully hedged -- a hundred million 14 dollars of swaps on day one? 15 Q. Yes, sir. 16 A. I'm sorry. Yes. That would be fully 17 hedged. 18 Q. Okay. Okay. 19 A. I was thinking about the serialization 20 issues. 21 Q. Yes, sir. Okay. You're fully hedged 22 if you -- that's the way that term was used during 2911 1 that time period. Right? 2 A. Yes. 3 Q. And if you had 300 -- if you had 4 75 million of hedges, you were underhedged 5 according to the nomenclature at the time? 6 A. There were -- yes, that's true. 7 Q. And if you had 125 million of swaps, 8 then you were thought to be overhedged. Right? 9 A. Yes. 10 Q. Nobody understood "fully hedged" to 11 mean that you could not possibly, no matter what 12 happened in the world, have a loss, did they? 13 A. Fully hedged was not used in that 14 context. 15 Q. Fully hedged was used to mean that the 16 notional amount of the swap in this example was 17 approximately equal to the amount -- the starting 18 amount of the MBS? 19 A. And I believe it also implied that the 20 term of the swaps would be staggered to match what 21 you thought the amortization stream did. 22 So, there is also -- there is a dollar 2912 1 amount up front for fully hedged and there is also 2 a term -- a maturity structure to the swaps that 3 was expected. 4 Q. You would try to match the life of your 5 assets to that of your liabilities? 6 A. Yes. And since the asset is paying 7 down over time, you would hope -- you would expect 8 the liabilities to also pay off. 9 Q. Now, if you are fully hedged in this 10 sense, you are hedged against rising interest 11 rates, correct? 12 A. On a market value basis, you may or may 13 not be. 14 Q. Depending upon whether your durations 15 are matched? 16 A. Depending on whether your durations are 17 matched. And there is actually a little more to 18 it than that because convexity also plays in if 19 you get big interest rate changes. 20 Q. Well, let's talk about if interest 21 rates go up in the circumstances, then your swap 22 payment, your swap partner is going to have to 2913 1 start making payments to you, correct? 2 A. If they go far enough up. Short-term 3 rates go far enough up, yes. 4 Q. And so -- and if you've got a hundred 5 percent match, then those payments ought to offset 6 any loss in income from the higher rates? 7 A. In the liability side of the portfolio, 8 yes. 9 Q. Now, what you are not hedged against in 10 that circumstance is prepayments? 11 A. Changes in prepayments. 12 Q. So, what you do in order to get 13 protection against higher rates, higher interest 14 rates, you have to give up -- you have to have 15 some exposure on prepayments. Would you say 16 that's fair? 17 A. You either have to have -- you have to 18 give up exposure or you have to give up spread to 19 buy more insurance up front. 20 Q. Okay. There is an inherent tradeoff. 21 Right? 22 A. The nature of the tradeoff changes 2914 1 through time, but there is a tradeoff. 2 Q. And there is nothing evil, bad, or 3 inherently so with the fact that that tradeoff 4 exists and you've got to make a choice? 5 A. I don't think it's inherently evil. 6 It's the nature of the market. 7 Q. And events may prove that being that 8 being fully hedged in this circumstance that we're 9 talking about is worse than not being hedged at 10 all? 11 A. There are scenarios where that can come 12 up. 13 Q. Indeed, if interest rates go down and 14 prepayments accelerate, the person who is fully 15 hedged actually turns out to be worse off than the 16 person who hadn't hedged at all? 17 A. If they -- they will be worse off 18 unless they are able to monitor and roll the 19 coupons and come out of the swaps on the way down, 20 but they have to be very nimble in order to make 21 that a zero. But you could do it. 22 Q. And indeed, even so, they are going to 2915 1 be behind the guy that hasn't hedged? 2 A. They will be behind the guy that hasn't 3 hedged, but they could come out close to the guy 4 that hasn't hedged. 5 Q. Somebody has made a choice to try -- 6 well, would you say that is -- maybe it doesn't -- 7 isn't subject to this kind of nomenclature. But 8 is that a conservative -- I'm talking about the 9 fully hedged now -- is that a conservative choice 10 or not conservative? 11 A. That is -- in the context of -- are you 12 asking that in terms of what possible rate 13 movements can occur or are you asking that in the 14 context purely of a one-sided rate movement? 15 Q. I'm really asking you whether the 16 nomenclature applies. 17 A. Fully hedged is generally thought to be 18 more conservative than unhedged. 19 Q. But it may not be more conservative 20 than underhedged. Right? 21 A. The way it's spoken about, it is more 22 conservative than underhedged. 2916 1 Q. Okay. But in your example, if we come 2 back to the paper that you wrote, "ABCs of RCAs," 3 you didn't recommend -- to the extent this is a 4 recommendation, you didn't show a fully hedged 5 position, did you? 6 A. In the analysis, we did not show a 7 fully hedged position. 8 Q. You showed a position 70 percent 9 hedged? 10 A. Three and a half out of five. 11 Q. Right? 12 A. Right. 13 Q. And that meant that this structure 14 would do a little better off in declining rates. 15 Right? 16 A. It meant that -- the main thing it 17 meant is that because we were doing swaps in this 18 model, we were doing swaps of two and three and 19 four years and we knew there were going to be 20 amortizations in the mortgage securities between 21 those years. Had we done a hundred -- or 22 5 million of swaps on day one against 5 million of 2917 1 mortgage assets on day one, at the end of the 2 first month, we would have been overhedged. 3 So, what we were attempting to do with 4 the underhedging as opposed to necessarily making 5 a rate call was to allow us to rebalance the 6 liabilities more smoothly over the life of the 7 transaction. 8 Q. Now, let's talk a little bit if we 9 can -- and I know this is difficult, going back to 10 what the prevailing knowledge was in 1985. Isn't 11 it the case that most people in that time frame 12 were concerned about rising rates as opposed to 13 declining rates in prepayments? 14 A. I can't help you there. I do not know. 15 Q. So, we'd have to go back and look at 16 the literature to try to -- 17 A. Yes. 18 Q. -- look at that issue? 19 A. Yes. 20 Q. It certainly was a time frame where it 21 was thought that interest rates had to fall 200 or 22 more basis points before you'd begin to see 2918 1 significant prepayments? 2 A. In the current coupons, I think that's 3 correct. 4 Q. And, in fact, there was sort of a rule 5 of thumb, was there not, that persisted for a very 6 long time that it was not economically -- it was 7 not economically viable for someone to refinance 8 unless you had a 200 basis point move? 9 A. It was really more around 125 or 150, I 10 think. 11 Q. You think that was true in 1984? 12 A. I don't know about 1984. I know about 13 late '85 and early '86. 14 Q. And in more recent times, that number 15 has gotten very much smaller. Right? 16 A. Very much. 17 Q. In part because of competition? 18 A. In the mortgage market. 19 Q. Right? 20 A. Yes. 21 Q. And also because of regulations? 22 A. What regulations? 2919 1 Q. Well, the regulations that allowed 2 people to put the refinancing cost into their new 3 mortgage. 4 A. Yes. That would -- that would have 5 something to do with it. 6 Q. Now, with regard to -- let's look at 7 the question, if we can, of what was thought about 8 prepayments back in 1984. Let me see if I can 9 find for you B505. I'm going to hand you an 10 article entitled "Constructive Use of Fixed Rate 11 Mortgages" that is Exhibit B505 published by 12 Salomon Brothers. 13 Salomon Brothers was a leading 14 authority in mortgage-backed securities. Would 15 you agree with that? 16 A. Yes. They were one of two or three 17 firms that were considered to be state-of-the-art 18 in the mid-Eighties. 19 Q. Indeed, if we look at your article, you 20 say right there at the very beginning -- 21 A. "Salomon Brothers" -- 22 Q. "Salomon Brothers seems to have been 2920 1 the innovator in this field"? 2 A. Yes. 3 Q. And just after -- "Just three years 4 after their first program went into effect, the 5 major firms on the street have over $15 billion in 6 assets under management"? 7 A. Yes. 8 Q. That was in risk-controlled arbitrages? 9 A. That was in my article. 10 Q. Now -- 11 MR. NICKENS: Your Honor, we would 12 offer B505. 13 MR. GUIDO: No objection, Your Honor. 14 THE COURT: Received. 15 Q. (BY MR. NICKENS) Do you know 16 Mr. Waldman? 17 A. I met him once. 18 Q. He was well-known in the field -- 19 A. Yes, he was. 20 Q. -- of mortgage-backed securities in the 21 early Eighties, correct? 22 A. Yes, he was. 2921 1 Q. And was thought to be authoritative on 2 the subject? 3 A. As investment bankers go, he was 4 probably one of two authorities at that time. 5 Q. Okay. Who was the other? 6 A. Ken Sullivan at Drexel Burnham. 7 Q. Okay. And he published this article, 8 Constructive Use of Fixed Rate Mortgages. 9 Arbitrage Opportunities for Financial 10 Institutions" in May of 1985. Right? 11 A. Okay. 12 Q. And let me ask you to look over at 13 page -- we'll just look at several. Just look at 14 the third page. In the middle of the page, you've 15 got a chart and then below that you've got a 16 paragraph that starts the page. And at the end of 17 that paragraph, you've got a parenthetical that I 18 would like to focus on, if you can. 19 It says "In fact, Salomon Brothers, 20 Inc. and Financial Publishing Company recently 21 published yield books that give cash flow yields 22 for mortgage securities for a series of constant 2922 1 prepayment rates." Do you see that? 2 A. Yes, I do. 3 MR. GUIDO: Where are you reading? 4 MR. NICKENS: In the parenthetical at 5 the end here. 6 MR. GUIDO: What page is it? 7 MR. NICKENS: The third page of the 8 document. It's numbered Page 2. 9 Q. (BY MR. NICKENS) Does that refresh 10 your recollection that people in 1985 were using 11 yield books for determining prepayment rates? 12 A. That's -- it's the other way around. 13 They were looking at prepayment books to calculate 14 what the appropriate yields were. 15 Q. Well, that -- okay. But that's where 16 you -- 17 A. But that's an important difference. 18 Q. Yes, sir. But that's where you would 19 go to look for prepayment predictions, is that not 20 the case? 21 A. I don't believe that's where you'd go 22 to look for prepayment predictions. 2923 1 Q. Well, in any event -- 2 A. What it says is we can look at cash 3 flow yields based on a particular prepayment rate. 4 And what that means is give me a prepayment rate 5 and I can tell you the cash flow yield on the 6 security. I don't think that this references 7 going backward, going from yields to prepayment 8 rates. 9 Q. Well, let's -- okay. Let's -- let's go 10 forward then and see what the point is. If we 11 turn to the next page and it's the carry-over 12 paragraph and -- let's see. In the middle, to get 13 everybody there, it says "at this rate" -- or no. 14 If you can start where it says "The average 15 prepayment rate." 16 Are you with me? 17 MR. GUIDO: What paragraph are you on? 18 MR. NICKENS: The carry-over paragraph 19 at the top, right about here. Page 3, right here. 20 MR. GUIDO: Halfway down the first full 21 paragraph at the top of the carry-over paragraph? 22 MR. NICKENS: Thereabouts. 2924 1 Q. (BY MR. NICKENS) It starts out "The 2 average prepayment rate." Are you with us, 3 Mr. Smith? 4 A. Yeah. "The average prepayment rate for 5 7584 period has been" -- 6 Q. -- "has been about seven and a half 7 percent." 8 A. Uh-huh. 9 Q. So, a period of nearly ten years, and a 10 half percent was the average prepayment rate? 11 A. Okay. 12 Q. "In the current housing environment 13 which has been reasonably stable for close to two 14 years, season discount Freddie Mac PCs have 15 experienced a 6 percent prepayment rate." 16 Do you see that? 17 A. I see the sentence. 18 Q. And you go -- it goes forward and the 19 assumption just below under "risk-controlled 20 arbitrage" is he's assuming a 6 percent annual 21 rate? 22 A. Okay. Well, let's -- can we back up 2925 1 just a moment to the paragraph above? 2 Q. Wherever you want to go. 3 A. It says "The average prepayment rate 4 for the 7584 period has been about seven and a 5 half percent." 6 Q. Yes, sir. 7 A. If you go to Figure 3 at the bottom of 8 the page, you'll see that that rate has been 9 anywhere on Freddie Mac PCs from 1 percent to 10 18 percent in that time period. So, I don't think 11 the average -- I don't think the statement of this 12 being an average was meant necessarily to 13 characterize that entire ten-year period. 14 Now, what he has shown is that over the 15 two years prior, which I guess would have been, 16 what, '83 and '84, is that what he's referencing, 17 you still see -- even though the average has been 18 much lower, you still see some cyclical 19 volatility. So, I don't think even at this time 20 period assuming that prepayment rates never moved 21 was an appropriate assumption. 22 Q. Well, bear with me. What he says is 2926 1 that the average prepayment rate was seven and a 2 half percent for that time period -- 3 A. Right. 4 Q. -- right? So, he's gone to the trouble 5 to find out what it was. 6 A. Okay. 7 Q. Okay? And as we've seen before, in a 8 single month in 1986, it got above 8. 9 A. Well, that -- that was -- are you 10 referring to the First Boston paper this morning? 11 Q. Yes, sir. Is that right? 12 A. I think you got annuals and monthlies 13 confused, but -- 14 Q. Well, this -- this is annual? 15 A. This is annual, and I think those were 16 monthlies that we were talking about. 17 Q. Yes, sir. That's my point. That in a 18 single month in 1986, after this was published -- 19 A. The monthly was 8 percent, yes. 20 Q. Right. So, something drastically 21 changed in 1986 from the environment that 22 Mr. Waldman was writing about in 1985? 2927 1 A. Okay. 2 Q. Okay? Now, if we turn over to Page 5, 3 Figure 6, now, what range of CPRs does he show for 4 this risk-controlled arbitrage? 5 A. Looks like 4, 6, and 8 if I can make 6 out the copy. 7 Q. And he says "Three interest rate 8 scenarios, an unchanged market, and rates moving 9 higher and lower by 200 basis points." 10 Do you see that? 11 A. Uh-huh. 12 Q. So, Mr. Waldman in 1985 was assuming a 13 movement of between 4 and 8 percent CPR to 14 correspond with an interest rate movement of 200 15 basis points up and down, correct? 16 A. That is what his paper shows. 17 Q. And if we go over two more pages and 18 looking at Page 9, Mr. Waldman concludes -- and 19 I'm looking at the paragraph right at the bottom 20 of that graph or chart. Figures 12 and 13 21 indicate that the profitability of the arbitrage 22 stands up well to massive changes in interest 2928 1 rates and a fair amount of variation in prepayment 2 rates. Thus, the duration match has accomplished 3 its intended purpose. 4 Do you see that? 5 A. I see that. 6 Q. Now, look at the chart up above. He's 7 got a chart that says "reinvestment rate." Right? 8 A. Yes. 9 Q. And he's running it from zero to 10 20 percent? 11 A. What is the reinvestment rate? 12 Q. Wouldn't that be the current interest 13 rate? 14 A. I guess that's what it is. 15 Q. That would be what you would put your 16 money into as you got it and that would be your 17 treasury rate or something. Right? 18 A. Okay. 19 Q. And then he shows -- 20 A. So, that's what he's using as the 21 current interest rate environment for each 22 scenario? 2929 1 Q. Right. And he's doing this scenario 2 and he's showing 4, 6, and 8 percent prepayments. 3 A. Uh-huh. 4 Q. And he's showing those prepayments to 5 vary independently of the reinvestment rate. 6 Right? 7 A. Okay. 8 Q. And what conclusions would you draw 9 from that? 10 A. That if mortgage prepayments aren't 11 sensitive to interest rates, you have a wonderful 12 duration hedge. 13 Q. Isn't it -- doesn't it indicate, 14 Mr. Smith, that the thinking at the time was that 15 prepayments were far less volatile than they 16 proved to be later? 17 A. It proves -- potentially it proves that 18 certainly in the context of what Salomon's trying 19 to sell. But, again, if we go back to the picture 20 on the bottom of Page 2 with historic prepayment 21 rates discount -- season discount mortgages, if 22 you just look at what happened to Freddie Mac 2930 1 within a one-year period, prepayment rates 2 doubled. 3 Q. Yes, sir. 4 A. And certainly if you looked at what the 5 interest rate changes were back in '76, '77, '78 6 associated with that doubling of prepayments, I 7 think you would find it to be in the realm of 8 changes he's shown here. So, he is in this 9 analysis certainly dampening even the historical 10 effect of interest rates on changes in 11 prepayments. 12 Q. Well, if you looked at a five-year 13 period or six-year period, he wouldn't have been 14 far off, would he? 15 A. Depends on where you started. 16 Q. Well, just about any five-year period, 17 if your average is seven and a half percent -- 18 A. Over that. 19 Q. Over a nine-year period? 20 A. Over this particular nine-year 21 period -- 22 Q. Yes. 2931 1 A. -- his average would not have moved 2 very much. But what I think you also -- anyone 3 who reading this article should have remembered in 4 '84 is what the interest rate scenarios were in 5 late '78 and '79. Those are unprecedented. 6 Q. Right. And let's talk about that for a 7 second. In 1984, interest rates had fallen 8 something -- or excuse me -- in the early part of 9 1985, interest rates had fallen some 200 basis 10 points, hadn't they? 11 A. Somewhere in that range. 12 Q. And people were concerned about a 13 rebound? 14 A. Possibly. 15 Q. And that would be a concern that would 16 lead somebody to fully hedge the way we talked 17 about; isn't that true? 18 A. That is possibly what a portfolio 19 manager would have done. 20 Q. If you had just experienced interest 21 rates coming down 200 basis points and you're 22 thinking that it's not likely to go much lower and 2932 1 it could very well go higher, it wouldn't be 2 unreasonable to look at a fully hedged position? 3 A. No. But it also wouldn't be 4 unreasonable to look at that trend continuing. I 5 mean, there -- 6 Q. You have to make a choice, don't you? 7 A. You have to make a choice. 8 Q. And it's not that there are bad choices 9 or evil choices. They are just choices that 10 reasonable people have to make? 11 A. They are choices that reasonable people 12 have to make. There are also choices that are 13 dominated by other choices in the sense that 14 people still make dumb mistakes. Even high-paid 15 professional people make mistakes. 16 Q. All the time. Right? 17 A. Uh-huh. (Witness nods head 18 affirmatively.) 19 MR. NICKENS: On that note, Your Honor, 20 I might ask for a break. 21 THE COURT: All right. We'll take a 22 short recess. 2933 1 (A break was taken at 2:55 p.m.) 2 3 THE COURT: We'll be back on the 4 record. Mr. Nickens, you may continue with your 5 cross-examination. 6 (3:26 p.m.) 7 Q. (BY MR. NICKENS) Mr. Smith, when we 8 took our break, I was questioning you about 9 movement of interest rates in 1985, and I want to 10 hand you a document that we've identified as B586. 11 MR. NICKENS: Your Honor, we would 12 offer B586 which on its face appears to be a 13 research -- a mortgage research bulletin 14 September 4th, 1985, of Salomon Brothers by 15 Michael Waldman, Mark Gordon, and Steve Guterman. 16 MR. GUIDO: Your Honor, I have no 17 objection with the same caveat that we have no 18 objection that this is information that was 19 published. We're not conceding the accuracy or 20 inaccuracy of the allegations that are made or 21 statements that are made in the paper. 22 THE COURT: Fine. Received. 2934 1 Q. (BY MR. NICKENS) Mr. Smith, the very 2 first paragraph under summary -- and by the way, 3 this is the Salomon Brothers prepayment model 4 impact of the market rally on mart prepayments and 5 yields. Do you see that? 6 A. Yes, I do. 7 Q. And it indicates that between March 7th 8 and June 18th, current coupon Ginnie Mae yields 9 fell by almost 200 basis points from 13.16 percent 10 to 11.17 percent. Now, of course, we were talking 11 about interest rate movements. Right? 12 A. Yes. 13 Q. But in these -- and then it goes on, 14 "While the market has backed off from the highs 15 reached in mid-June, mortgage interest rates are 16 still nearly as low as any time during the past 17 five years." 18 Do you see that? 19 A. I see the statement. 20 Q. And do you recall that time frame? Do 21 you have any different information? 22 A. I don't have any different information. 2935 1 Q. And in these circumstances, it would be 2 reasonable to conclude that one might begin to see 3 interest rates rising? 4 A. I don't follow that. 5 Q. Well, if interest rates are still 6 nearly as low as at any time during the past five 7 years and you had just experienced a 200 basis 8 point drop in the yield for Ginnie Maes, do you 9 reach any conclusion from that? 10 A. I would not reach any conclusion from 11 that. 12 Q. Now, with regard to the timing about 13 prepayment models, you remember I asked you about 14 that? 15 A. Uh-huh. (Witness nods head 16 affirmatively.) 17 Q. And you see that this is announcing the 18 Salomon Brothers' prepayment model in September of 19 1985? 20 A. Yes. 21 Q. And if you go down to the third 22 paragraph, it says "In order to create a logical 2936 1 framework for valuing mortgage issues, Salomon 2 Brothers has developed a statistical model of 3 prepayment rates based upon historical prepayment 4 experience." 5 Now, was this something that Salomon 6 Brothers was again out front on? 7 A. I don't know how out front they were. 8 There were three or four companies that were all 9 projecting or that were all showing prepayment 10 models in early '86 when I was involved in the 11 market. 12 Q. Now, let me ask you to look at 13 Exhibit B1785 which is entitled "The Mortgage 14 Securities Market Analytical Methods and New 15 Product Opportunities by Michael Waldman," 16 October 1987. Do you see that? 17 A. I see that. 18 MR. NICKENS: Your Honor, we offer 19 B1785. 20 MR. GUIDO: No objection with the same 21 caveat, Your Honor. 22 THE COURT: Received. 2937 1 Q. (BY MR. NICKENS) Let me ask you to 2 look over at Page 6 under "modeling prepayments." 3 And again, October of 1987, Mr. Waldman writes "To 4 project applicable prepayment rates in a given 5 market environment, Salomon Brothers developed a 6 prepayment model about two years ago." And it 7 cites the article we just looked at. 8 A. Uh-huh. (Witness nods head 9 affirmatively.) 10 Q. "And this is an econometric model that 11 describes prepayment rates as a function of 12 several variables but mainly mortgage interest 13 rates based upon the experience of past years." 14 Then let me ask you to look over at 15 Page 10. This goes back to something we were 16 discussing before, but you see he's got a 17 paragraph there, "option adjusted yield spreads"? 18 And the third sentence or second sentence, I 19 guess, "To address this issue, we developed an 20 option pricing model for mortgage securities," and 21 then he's got Footnote four. And that is dated in 22 September of 1986. 2938 1 A. Yes. 2 Q. And do you know of any earlier mention 3 of a working option adjusted spread model? 4 A. As I recall, the first one that came to 5 my attention was one from Drexel Burnham that I 6 believe came out a little bit before Salomon 7 Brothers. 8 Q. So, sometime toward the end of 1986? 9 A. That was the first one I was aware of. 10 Q. Let me ask you, do you know Mr. Kenney? 11 A. I spoke with him on the phone a couple 12 of times. 13 Q. Okay. I've handed you Exhibit B2054 14 which purports to be an article on risk-controlled 15 arbitrage by Mr. Kenney in March of 1988. 16 MR. NICKENS: Your Honor, we offer 17 B2054. 18 MR. GUIDO: No objection, Your Honor, 19 with the same qualification. 20 THE COURT: Received. 21 Q. (BY MR. NICKENS) Mr. Kenney is with 22 the -- or was with the Federal Home Loan Bank 2939 1 system? 2 A. Yes. The Chicago bank. 3 Q. And it says here that -- checkmarks 4 "active management is critical." 5 Do you agree with that? 6 A. Yes. 7 Q. "Prepayment changes can threaten 8 spreads." 9 Do you agree with that? 10 A. Yes. 11 Q. Now -- and this is on the right side 12 there, the beginning of the article, the second 13 paragraph. "Recently, many proponents of the 14 strategy, including investment bankers and 15 independent have advocated more widespread use. 16 Typically, the arbitrage features the purchase of 17 mortgage-backed securities funded by reverse 18 repurchase agreements. The term of the" -- and I 19 can't -- "repo is synthetically extended using 20 interest rate swaps, futures, or options." 21 In other words, the typical model was 22 the same in March of '88 as it had been for 2940 1 several years, correct? 2 A. According to Mr. Kenney. 3 Q. Do you have some different information? 4 A. That's the basic model. By 1988, there 5 were many alternatives that were being used by 6 late '87, early '88 using different securities and 7 different hedge techniques. 8 Q. And again, looking at the second page, 9 Mr. Kenney -- and let's see -- fourth full 10 paragraph down. "Some advocates have proposed the 11 use of risk-controlled arbitrage to mitigate the 12 problems of ailing savings institutions. They 13 contend that these institutions could use the 14 additional returns to offset the low or negative 15 returns from distressed assets." 16 Now, that's something we discussed 17 earlier in the day. Right? 18 A. Yes. 19 Q. And, again, this was something that the 20 bank board did in certain circumstances recommend 21 for ailing institutions -- that is, 22 risk-controlled arbitrage? 2941 1 A. You again would have to talk to one of 2 the regulatory people to see what they 3 recommended. We were asked on several occasions 4 if it was possible to construct very low 5 risk-controlled arbitrages for ailing 6 institutions, and our answer was yes. 7 Q. Now, I want to talk for a few minutes 8 about the question of active management. This 9 goes to that rebalancing issue that we've 10 discussed. Right? 11 A. Yes. 12 Q. Now, is it also the case that active 13 management involved the identification of values 14 in the marketplace? 15 A. I need more definition. 16 Q. Okay. It is the case that there is a 17 historical relationship between mortgage-backed 18 securities of various coupons and treasury 19 securities, is there not? 20 A. There are historical averages. 21 Q. And one who watches these things can 22 see those averages and occasionally, if we can 2942 1 think of it graphically, you can get anomalies in 2 the curve where the spread widens or narrows from 3 the relationship to treasuries? 4 A. Yes. Those spreads are not constant 5 over time and they will wander. 6 Q. And in watching those, someone can 7 identify what could be regarded as being cheap or 8 expensive issues of mortgage-backed securities? 9 A. There are trading strategies that are 10 based on that kind of analysis. 11 Q. Okay. And so, you watch those 12 historical trends and if they narrow, for example, 13 one might think that the -- and this is based upon 14 price or yield -- one might think that the 15 security was cheap. Right? 16 A. Certainly, you have had a positive 17 basis change in your favor if you're owning the 18 securities because that spread has collapsed. 19 Rich/cheap analysis is something I never really 20 bought into so -- 21 Q. Well, it's certainly out there. Right? 22 A. It's out there and there are people 2943 1 that trade based on those relationships. 2 Q. And very successfully? 3 A. Some have. 4 Q. And it's a low risk kind of strategy, 5 is it not? 6 A. It depends on how it's implemented. 7 Q. Okay. Now, you can -- if you have a -- 8 if you have a particular coupon, for example, that 9 would be regarded as being expensive on this 10 analysis, you could sell it and swap it for one 11 that you had identified as being cheap? 12 A. That's the way the strategy works, is 13 you buy the security. When the spread will 14 narrow, you sell it. When it widens, you buy back 15 in or you swap to something that's wider. That's 16 the way the strategies typically work. 17 Q. And you're simply trading based upon 18 this spread from the treasury curve? 19 A. That's correct. 20 Q. You're still having mortgage-backed 21 securities in both instances but with perhaps 22 slightly different coupons? 2944 1 A. Right. And the issue -- generally, 2 what I have seen is the coupons are the same and 3 you're trading between agencies or the coupons are 4 very, very close, within half a point. 5 Q. Now, I want to refer back to 6 Exhibit B823. Let me see if I can find it for us. 7 A. Where are you? 8 Q. 823 is the cashing in on 9 risk-controlled arbitrage article. One of the 10 first ones I asked you about. 11 A. The one from Institutional Investor? 12 Q. Yes. 13 A. Okay. 14 Q. And I'm going to ask you about the 15 third page of the article on the far right-hand 16 column down toward the bottom right here that says 17 "most thrifts." It says "Most thrifts that pursue 18 risk-controlled arbitrage also actively trade 19 pass-throughs, buying those they think are 20 undervalued and selling the ones they think are 21 too dear to add incremental returns. They view 22 this as another means of obtaining insurance 2945 1 against prepayment risk. They constantly sift 2 through the peculiarities of the different 3 pass-through issues and track the pattern of 4 supply and demand in the market much the same way 5 active managers of fixed income portfolios do to 6 boost returns." 7 Now, do you agree or disagree with that 8 observation? 9 A. I would agree that many of the 10 institutions that did risk-controlled arbitrages 11 or had them in place were trading the securities 12 based on rich/cheap. 13 Q. They had thought they had identified 14 bargains and would swap back and forth? 15 A. That was their expectations. That's 16 why they were doing the transactions. 17 Q. Do you know that -- is there anything 18 wrong with that? 19 A. The issue really comes down to are the 20 swaps that you're doing bringing back in 21 securities that are sufficiently similar to the 22 ones that went out to keep the liability structure 2946 1 in place and appropriate so that you don't have to 2 rebalance. That's one. 3 The other thing that this talks about, 4 it compares it to the way active managers of fixed 5 income portfolios do to boost returns. 6 Rebalancing and active management are really two 7 different things in risk-controlled arbitrages. 8 Rebalancing is generally in response to some 9 market shock as opposed to a richening or 10 cheapening of coupons. If you're trading on 11 richening and cheapening coupons, what you're 12 attempting to do is boost returns and hopefully 13 keep the transaction in place. 14 So, you are -- at least in my belief, 15 you are trading the assets in order to increase 16 the returns from the base structure. 17 Q. Okay. But if you could improve your 18 yield without changing your durations, that's 19 certainly possible to do, isn't it? 20 A. That is possible to do. 21 Q. And if you can improve your yield 22 without changing your durations, isn't that 2947 1 something that a manager ought to do? 2 A. That is, but the manager can't create 3 these opportunities in the sense that there may be 4 times when the market is for some reason 5 dislocated in a particular coupon that you own for 6 some sort of supply or demand reason. You happen 7 to be in a lucky enough in position to have that 8 coupon. 9 Q. Are you talking about dollar rolls or 10 something? 11 A. Dollar rolls or something like that. 12 Or essentially there may a collateral shortage 13 that somebody needs Fannie 8s to cover a position. 14 As a result, they are willing to pay a little bit 15 more to get their hands on them so you would sell 16 Fannie 8s and buy Freddie 8s. And that's 17 generally an opportunistic transaction and 18 generally those things don't occur on a regular 19 basis. 20 Q. But nevertheless, it's something that a 21 good manager ought to be on the lookout for? 22 A. At least looking for. 2948 1 Q. And if you can improve your yield while 2 not changing your duration or other important 3 characteristic of your hedge, then that's 4 something that you ought to do? 5 A. Or the asset. And one of the things 6 that we've seen in mortgage land is things that go 7 in these rich/cheap relationships many times are 8 related to changes in prepayment speeds. For 9 example, Fannie 8s and Freddie 8s are not going to 10 prepay exactly the same and if for some reason 11 there is an expectation in the market that the 12 Fannie 8s are going to accelerate in prepayments, 13 you may move to the Freddie 8s. That may help 14 your hedge. That may hurt your hedge depending on 15 what's in place. 16 Q. But that ought to show up in your 17 durations? 18 A. Ultimately in your durations. 19 Q. Let me ask you to go back to an earlier 20 document, one of your articles, which is 21 Exhibit B2055. 22 A. Got it. 2949 1 Q. Now, earlier, I had asked you -- I want 2 to go over to Page 4. 3 A. Is that Page 73 or F -- 4 Q. 73, yes, sir. 5 MR. GUIDO: Which exhibit is this 6 again? 7 MR. NICKENS: This is Exhibit 2055. 8 Q. (BY MR. NICKENS) Earlier, I had asked 9 you at Page 73 about the roll-down. 10 Do you recall that? 11 A. Yes. 12 Q. That's in the bottom right-hand column? 13 A. Uh-huh. 14 Q. And I want to ask you, if you turn over 15 the page, about the opposite side. You write "If 16 interest rates rise, the desired rebalancing 17 strategy for an underhedged position is to add 18 swaps instead of rolling up coupons. Rolling up 19 coupons would require the immediate recognition of 20 a capital loss. Adding swaps will reduce the 21 expected spread because the average fixed rate 22 paid is now higher. Whether rates rise or fall, 2950 1 the rebalancing cost of the risk-controlled 2 arbitrage in terms of a lower than expected spread 3 over the remaining time horizon" -- did I lose a 4 verb there? 5 A. No. I probably can't write English. 6 Go ahead. 7 Q. I'm sure that I misread that in some 8 way. "This cost should not be viewed as a 9 deterrent from doing a risk-controlled arbitrage." 10 Do you see that? That was your view 11 back in 1988? 12 A. You see that when you do the up-front 13 modeling, that if you go through the up-front 14 modeling looking to change over time, you see that 15 there is going to be a cost and, in fact, I 16 believe in one of those Hjerpe papers you had me 17 looking at earlier, he had an expected cost of 18 rebalancing. So, I agree that that is where I was 19 in 1988. 20 Q. Okay. And what you were talking about 21 here is if you were underhedged -- and by that, we 22 mean that you're in an example swap, the notional 2951 1 amount of your swaps are less than the amount of 2 your MBS. Right? 3 A. (Witness nods head affirmatively.) 4 Q. That the way to adjust to acquire the 5 desired match -- the suggested way here is to add 6 swaps? 7 A. In an upright world. 8 Q. What do you mean by that? 9 A. If rates increase -- 10 Q. Yes. That's what we're talking about 11 here. 12 A. Right. 13 Q. So, your suggestion was that if rates 14 decrease, you roll down on the assets and if rates 15 increase, you purchase additional swaps or 16 liability hedges, correct? 17 A. That would be your first -- the first 18 thing you would look to do. 19 Q. And in both instances, you would 20 recognize current gains and defer the payments, 21 wouldn't you? 22 A. If I'm adding swaps, I don't -- by 2952 1 adding swaps in this scenario -- 2 Q. As opposed to rolling up the coupon. 3 A. I understand. You would see a capital 4 loss if you sold the mortgages. And what we are 5 suggesting in this paper is that by adding swaps, 6 you can avoid looking that capital loss up front 7 and hopefully retain your expected spread. 8 However, the market value of the total 9 transactions moved against you, most likely. 10 Q. And on the other side of it, on the 11 roll-down, you would be recognizing gains while 12 deferring those losses on your swaps or paying 13 them out over time? 14 A. If you were -- for small changes, yes. 15 Q. And that's a perfectly acceptable 16 management tool to take into account, isn't? 17 A. As long as you are properly modeling it 18 and look at what the spreads are going to be and 19 the market values are going to be and making the 20 decision that those are acceptable resultant 21 spreads. 22 Q. Proper management includes recognition 2953 1 of the accounting rules that you live under. 2 Right? 3 A. Yes. 4 Q. And if you can take advantage of them, 5 there is nothing wrong with that as long as it 6 doesn't destroy the structure of the arbitrage? 7 A. As long as you're not fooling yourself. 8 Q. Okay. But you do have to take 9 recognition of those rules? 10 A. You have to take the accounting -- you 11 have to manage the economics of the transaction. 12 If the economics of the transaction can be altered 13 in ways where you can have better accounting 14 treatment, then you should look to do that. If 15 you can't fix the problem, the economic problem, 16 without seeing -- without having -- let me try to 17 get the double negatives out. 18 If you cannot fix the economic problem 19 without creating a bad accounting outcome, then 20 you've fooled yourself if you don't do anything. 21 You need to address the economics. And to the 22 extent some economics have better accounting than 2954 1 others, those are the ones you should use if they 2 get you to the same place. 3 Q. Okay. If we can hold the economics 4 equal, the same, you would say as you did here 5 that a proper management is to take into account 6 the accounting issues? 7 A. Economics the same, you would take the 8 accounting issues into account. 9 Q. Now, I want to spend just a few minutes 10 talking about some of these terms that I promised 11 before lunch that we would do. The term 12 "duration" we've used frequently here. And there 13 are a variety of uses of that term, correct? 14 A. Yes. 15 Q. Currently, the term is most frequently 16 used to indicate price elasticity. 17 Would you agree with that? 18 A. That's what duration is supposed to be. 19 Q. So, even though it measured in years, 20 it, in fact, is a percentage change in price, is 21 it not? 22 A. I don't know how much math you want to 2955 1 get into. The McCauley duration is measured -- 2 actually, I believe, in years squared. 3 And McCauley duration just drops out of 4 the math of saying if you write down what the 5 relationship between a bond price and its yield is 6 and take a derivative to ask what's the change in 7 price for a change in yield and just drop through 8 the math, what comes out of that is something 9 called McCauley duration. And because of the 10 goofiness of mathematics, it's measured in years 11 squared. You can transform that number into 12 something called modified duration -- 13 Q. Yes, sir. 14 A. -- which is a percentage. 15 Q. And that's the one that's most 16 frequently used? 17 A. That's the one that's generally used 18 for things other than mortgages today. 19 Q. And then what's used with mortgages is 20 something called effective duration? 21 A. Or depending on what -- what time frame 22 am I in? 2956 1 Q. Let's talk about today. 2 A. Today you would either use effective 3 and implied, which I think are about the same 4 thing, or option adjusted duration. 5 Q. Now, let's see if we can use an 6 example. Let's say you have a mortgage-backed 7 security that's priced at a hundred par and there 8 is a movement in interest rates up so that the 9 price goes to 95. 10 A. Okay. 11 Q. And let's say the interest rate 12 movement is a hundred basis points. 13 A. Okay. 14 Q. What is your effective duration? 15 A. 5. 16 Q. 5. So, what you're measuring with 17 duration is the change in price for each hundred 18 basis point movement in interest rates, correct? 19 A. That's correct. 20 Q. And -- in other words, price 21 elasticity? 22 A. That's right. 2957 1 Q. Now, what is the relationship between 2 that concept and average life? 3 A. Average life is given the coupon of a 4 mortgage, the final maturity of the mortgage, and 5 an expected prepayment speed, you can calculate 6 what's essentially a half-life of that mortgage, 7 how long you expect the mortgage to be outstanding 8 on average. And I can't get much closer than that 9 without math. 10 Q. Okay. 11 A. What it's trying to -- what -- the 12 issue that average life is trying to reach is that 13 a mortgage with a very high coupon that's 14 prepaying very rapidly is going to go away 15 quickly. A pool of mortgages are going to go away 16 quickly because of those refinancings. So, the 17 pool isn't expected to stay out -- the entire pool 18 is not expected to live for 30 years. It's only 19 expected to live perhaps for five years. And 20 average life is trying to capture the expectation 21 of how long the pool's outstanding given 22 prepayment speeds of coupons. 2958 1 Q. And going back to the risk-controlled 2 arbitrage, the basic tools that the manager had in 3 the '85, '86 time frame was duration matching or 4 scenario analysis, correct? 5 A. They had a little bit more than 6 duration matching in the sense that they did 7 understand -- we did understand convexity, which 8 is the fact that that duration number is not 9 constant over changes in interest rates. So, they 10 did understand convexity, which is what drove 11 people to scenario analysis. But 12 duration/convexity and scenario analysis were the 13 predominant techniques that were used. 14 Q. And if one was using both, you were 15 using the basic tools available in the 16 marketplace? 17 A. The stuff that was available to 18 98 percent of the market participants. 19 Q. And you couldn't be critical of someone 20 using both in trying to manage its arbitrage? 21 A. You would want to look at everything 22 you have to provide you information about the 2959 1 transaction. 2 Q. Sure. And -- but on the assumption 3 that one was using both, you were using basically 4 what was available to try to manage the situation? 5 A. Yes. 6 Q. And you don't know, I take it, of what 7 USAT was doing one way or another? 8 A. I do not. 9 Q. Knowing Sandy Lawrenson, would you have 10 expected her to use both of those tools? 11 A. Knowing Sandy today, I would have 12 assumed that she was using whatever tools she had 13 at her -- whatever she had access to. If she 14 didn't have computers, she wasn't going to be 15 doing scenario analysis even though she may have 16 wanted to. So, not knowing the tools she had at 17 hand, I don't know how to answer that question. 18 Q. Well, actually Bloomberg may have been 19 available not in the sense that it is available 20 today, correct? 21 A. Bloomberg was available. What time 22 frame are we in again? 2960 1 Q. We're talking about '85, '86. 2 A. '85, '86, there were several investment 3 firms starting early analytics out to the desktops 4 at that point, too. 5 Q. Bloomberg, for example, today, you can 6 with a key stroke pull up and perform a stress 7 analysis on basically any mortgage-backed security 8 in the market. Right? 9 A. And anything that's in Bloomberg's 10 database. 11 Q. So, with a single stroke, you can look 12 at a scenario analysis up and down 300 basis 13 points or anything you want to look at for any 14 mortgage-backed security in the market? 15 A. It's four or five key strokes and it's 16 anything in Bloomberg's database. But, 17 effectively, that's correct. 18 Q. That was not available in 1985 or '86? 19 A. Bloomberg's database was not that big. 20 Their analytics were not that developed in '85, 21 '86. 22 Q. Mr. Smith, I'd like to turn our 2961 1 attention to a document that is already in the 2 record at Tab 172. 3 MR. NICKENS: Your Honor, this is also 4 B337 and it's the October 24th, 1984, Salomon 5 Brothers presentation. I'm told that it should 6 be -- Your Honor, your number may be T4069. Let 7 me see if I can find it over here. 8 MR. GUIDO: Your Honor, could we have a 9 short break? I wasn't given notice that these 10 documents were going to be used, and I have to 11 have them brought down unless Mr. Nickens has 12 another copy. 13 MR. NICKENS: We can get you one. I 14 apologize, Your Honor. There are some attachments 15 and they don't appear to have made it in the copy 16 and I want to make sure that we have them. I'll 17 get that fixed, Your Honor, so we can ask some 18 questions. 19 Q. (BY MR. NICKENS) But let me just ask 20 a few questions, Mr. Smith. If you look over -- 21 this document has already come into evidence and, 22 as indicated, it was a presentation made by 2962 1 Salomon Brothers concerning risk-controlled 2 arbitrage to USAT in October of 1984. And does 3 your document have Bates numbers on them? 4 A. Yes, they do, at the bottom. 5 Q. Okay. 6 MR. NICKENS: Your Honor, I'm going to 7 give up. We'll see if we can get the appropriate 8 copies. Your Honor, if you could indulge us, I've 9 got one copy and we can get the other one. 10 Q. (BY MR. NICKENS) Okay. Now, let's 11 see if we can go through this. 12 Does your copy have down at the bottom 13 a CN number? 14 A. CN253017. 15 Q. Right. And I would ask you to turn 16 over to CN253030. 17 A. 253030. 18 Q. Actually, let's start with 3028. 19 A. CN253028? 20 Q. Right. 21 A. Okay. It says "by assets" on the top? 22 Q. Right. And it's indicating a proposal 2963 1 of buying Ginnie Mae 223F, which it indicates is 2 current coupon. Right? 3 A. Yes. 4 Q. And Fannie Mae 8s and Freddie Mac 8s, 5 correct? 6 A. Correct. 7 Q. And then if we turn over to Page 8 8 which is CN253030, you see at the top those same 9 suggestions except here you've got amounts, 10 yields, and durations? 11 A. Yes. 12 Q. And it's indicating 500 million of the 13 Ginnie Mae 223F, and the Fannie Mae 8s, 14 250 million, Freddie Mac 8s, 250 million. 15 Do you see that? 16 A. Yes, I see that. 17 Q. Okay. Now, at -- now, let me ask you 18 to turn over to CN253053 and compare it to the 19 next page, 3054. 20 Does that indicate to you that Salomon 21 Brothers was suggesting to USAT that it could 22 improve its gap position from a negative 121.1 to 2964 1 a positive 19.0 through the use of this arbitrage? 2 A. The gap is improving, although it's 3 hard for me to tell if that is what's driving 4 that. We have a branch sale in here, but I can't 5 find the mortgage securities or the swap in this 6 gap statement. 7 Q. And so, you don't have a way of 8 analyzing? 9 A. I really don't because if you look at 10 the -- if you look at the interest-earning assets 11 for June 30 and then June 30 after branch sale, 12 they went down. And if you're buying mortgage 13 securities, it seems to me your investment 14 securities in stuff ought to go up. So, I'm -- I 15 can say that the gap -- the gap does improve, but 16 I can't tell you what's driving it from this level 17 of detail in this sheet. 18 Q. Okay. Let me ask you then to turn over 19 to CN253058. 20 A. 3058. Okay. 21 Q. And that, again, it was attached as the 22 risk-controlled arbitrage for thrift institutions 2965 1 by Mr. Waldman and Mr. Lupo. Do you see that? 2 A. I see that. 3 Q. And if you turn over to the page 4 before, do you have a date? 5 A. The page before -- oh, the main cover 6 page, the one that says October 1983? 7 Q. Yes, sir. 8 A. Yes. 9 Q. So, would you conclude from that this 10 is what Salomon Brothers was telling its customers 11 about risk-controlled arbitrage for thrift 12 institutions in October of 1983? 13 A. That would be the implication. 14 Q. Then let me ask you to look over at 15 CN253060. It says at the beginning, "Thrift 16 institutions have undergone receiver earnings 17 pressures during the past few years." 18 Now, you were in -- teaching at SMU or 19 were you in Chicago at this time? 20 A. In '83, I was at SMU. 21 Q. "Spreads have eroded drastically due to 22 asset liability mismatch and rising interest 2966 1 rates. While the rally and the credits markets 2 during the second half of 1982 has helped restore 3 some of the lost grounds, serious difficulties 4 remain. A technique that allows an institution to 5 generate additional earnings without creating a 6 further mismatch is the use of risk-controlled 7 arbitrage." 8 Now, would you agree or disagree with 9 that statement? 10 A. That last sentence, I don't know that I 11 would have said so stridently in the sense that it 12 is a technique that could allow you to do that. 13 If you do it improperly, then it'll give you the 14 opposite results. 15 Q. And you had a lot more experience with 16 it over -- in the last 15 years or so. Right? 17 A. Yes. 18 Q. And if you look down at the next 19 paragraph, the last sentence, "To a great extent, 20 these arbitrages can be structured so as to 21 minimize the interest rate exposure of the 22 institution." 2967 1 Do you agree or disagree with that 2 statement? 3 A. I agree with that statement. 4 Q. Now, would you ever understand such a 5 statement to mean to minimize it to zero? 6 A. No. 7 Q. And I know this is very small, but the 8 footnote if you can read -- if you can read along 9 with me, if I can read it, Footnote 2. "The 10 duration of an asset or liability is defined as 11 the average life of its cash flows weighted by 12 their present values. The duration measures the 13 rate at which interest rate movements affect the 14 value of the instrument." 15 That's what we were talking about. 16 Right? 17 A. Uh-huh. (Witness nods head 18 affirmatively.) 19 Q. "Thus, at least in theory, duration 20 matching projects or immunizes the institution 21 against changes in interest rates. For a 22 discussion of some of the issues involved in this 2968 1 complex subject, see bond immunization, et 2 cetera." 3 Now, what was the understanding or use 4 of the word immunization in this context? 5 A. Immunization originally, in the 6 academic literature and as it moved into portfolio 7 management, talked about using duration matching 8 to protect an asset from -- an asset and liability 9 from changes in interest rate movements in the 10 sense that you would immunize a treasury bond 11 portfolio by using either futures or short 12 positions and other treasuries so that the market 13 values would be offset. Immunization in the pure 14 academic -- in the pure academic content meant 15 pure total hedging with no changes when it was 16 originally developed. And if you look in this 17 book, there is -- I'm sorry. There is a book that 18 this article is actually put in on bond 19 immunization, and the academic view is that you 20 were trying in perfect markets to take all the 21 risk out of a portfolio. 22 Q. Okay. And let me ask you to turn over 2969 1 a couple pages and tell the Court what Mr. Waldman 2 in 1983 was using as an assumed prepayment rate. 3 MR. GUIDO: What page are you on? 4 MR. NICKENS: At the bottom, it says 3 5 or CN253062. 6 A. 5 percent prepayment rate, it looks 7 like, on conventional mortgage loans with a 9.875 8 coupon. 9 Q. (BY MR. NICKENS) And then if we turn 10 the page it says something called horizon 11 analysis? 12 A. Yes. 13 Q. Is that what you understand to be 14 scenario analysis? 15 A. It looks to be. 16 Q. And it's using a 200 basis points up or 17 down? 18 A. Yes. 19 Q. And it shows at 200 basis points up and 20 down an investment -- reinvestment rate going from 21 8 to 12 percent. It's indicating a 3 to 7 percent 22 CPR? 2970 1 A. I see the 3 to 7 percent CPR. What was 2 the middle number you wanted me to look at? Oh, 3 the reinvestment rates? 8 to 12. And prepayment 4 speeds from 3 to 7. But he's also hedged himself 5 on that table on the first page of the article. 6 Q. What do you mean by that? 7 A. He has a paragraph on the very first 8 page that you showed me that goes into the fact 9 that one risk of an arbitrage using mortgage loans 10 is the reinvestment of the actual prepayment 11 experience. "Yield cash flow average life and 12 duration of a mortgage investment depend greatly 13 on the prepayment rate that the pools and the loan 14 undergo." A slower -- and then he goes on to talk 15 about higher and slower or slower and faster 16 prepayment speeds. 17 So, I think what he's trying to do up 18 front, given the disclosure in describing the 19 arbitrage, is to give some sense that these 20 prepayment speeds are unpredictable or hard to 21 predict. 22 Q. Do you feel like this would be 2971 1 indicative of what was the thinking at the time 2 concerning risk-controlled arbitrage? 3 A. In '83, I wasn't in the market. I 4 can't tell you what the market was thinking in 5 '83. 6 Q. But Mr. Waldman, we have established I 7 believe, was a recognized expert in this 8 particular area? 9 A. Mr. Waldman, along with a couple of 10 other firms, were the leaders in trying to apply 11 analytics to mortgages. 12 Q. And you wouldn't expect him to publish 13 something that was false? 14 A. I would expect him to publish something 15 that showed how transactions can work. I don't 16 know that when he publishes anything that he's 17 using real numbers. 18 Q. Well, he is talking to a very 19 sophisticated audience, is he not? 20 A. Yes. 21 Q. Institutional investors? 22 A. Institutional investors. 2972 1 Q. People that know how to evaluate 2 numbers and claims with regard to performance? 3 A. And know how to discount claims made by 4 investment bankers in their marketing literature. 5 Q. Okay. But nevertheless, it would have 6 to be something out there that he would think at 7 least would be persuasive to them? 8 A. That's fair. He would think it was 9 persuasive. 10 Q. Now, let me ask you to go over to Page 11 3 -- 253069, which is Page 10 of this pamphlet. 12 At the top of the page, the first full paragraph, 13 "As this analysis indicates, the profits on the 14 arbitrage stand up to a surprising degree under an 15 assortment of tests relating to the market price 16 of the mortgages, the general level of interest 17 rates, and the timing of interest rate 18 fluctuations. In other words, the duration match 19 provides a significant degree of protection 20 against interest rate movements." 21 Now, would you say that was typical of 22 the thinking of risk-controlled arbitrage in 2973 1 October of 1983? 2 A. In October of 1983, I don't know. 3 Q. Okay. Let me ask you to look over to 4 Page 12, again with the idea of what people were 5 thinking back in this time frame. 6 MR. GUIDO: What page? 7 MR. NICKENS: Page 12. It's 253071. 8 Q. (BY MR. NICKENS) Mr. Smith, it's 9 difficult to forget what you've learned in the 10 last 14 years. Right? 11 A. Well, it's difficult. It's also 12 difficult to put yourself into a market that you 13 weren't in. 14 Q. Yes, sir. 15 A. I mean, I was writing academic articles 16 on international finance in 1983. 17 Q. But if the judge is going to evaluate 18 the behavior of these respondents, then he would 19 need to look back at what was prevailing at the 20 time. Wouldn't you agree with that? 21 MR. GUIDO: Your Honor, at this point 22 in time I would like to object to this whole line 2974 1 of testimony because 1983 I don't think there was 2 a risk-controlled arbitrage at USAT. I think that 3 the allegations in the complaint go with the 4 portfolio that didn't start until late 1984 and 5 we're talking about events -- we're complaining 6 about events in late 1985 or middle of 1985 7 through, I think, 1987 and '88. We are not 8 talking about 1983 so that all of this testimony 9 that's being sought here -- and I've listened to 10 it for a great deal -- I believe is not materially 11 relevant to any of the claims in the notice of 12 charges. 13 THE COURT: Well, it's a little hard to 14 separate these dates. The exhibit we're looking 15 at here is October 24, '84. So, that's getting 16 pretty close to '85. 17 MR. NICKENS: And, Your Honor, I 18 believe the evidence will show that the 19 risk-controlled arbitrage was begun shortly after 20 this date, the early part of 1985, and then built 21 up through the early part of 1985. 22 Now, I've been trying to cover a fairly 2975 1 wide range of time frame but this was a document 2 that was in the files of United Savings with 3 regard to risk-controlled arbitrage. In any 4 event, if I might continue, I will get through 5 this. 6 THE COURT: All right. Continue. 7 Q. (BY MR. NICKENS) If you would turn 8 over to Page 12. Have you got it there? 9 A. I'm on Page 253071. 10 Q. Yes, sir. 11 A. Okay. 12 Q. And at the top, "Consider an arbitrage 13 where Federal Home Loan Bank finances the purchase 14 of 13 and a half percent current coupon 15 single-family mortgages. Suppose interest rates 16 declined, say, 400 basis points setting off rapid 17 refinancing of the mortgages. A 250 basis point 18 decline probably would trigger some refinancing by 19 homeowners." 20 Do you see that? 21 A. I see that. 22 Q. Now, does that indicate to you that at 2976 1 least Mr. Waldman was acting on the assumption 2 that a 250 basis point decline would not set off 3 significant repayments? 4 A. That appears to be what Mr. Waldman is 5 saying. 6 Q. Now, let me ask you to turn over to 7 Page -- it's Page 6 because we are now into 8 another article, which is Salomon Brothers "High 9 Coupon Mortgage Securities and Short-term 10 Investments," April 12th, 1983. 11 MR. GUIDO: What's the Bates number? 12 MR. NICKENS: It's Bates No. 253082. 13 Q. (BY MR. NICKENS) Again, I'm asking 14 you about the thinking at the time and I want to 15 look at the sixth paragraph that starts out "the 16 question." 17 A. Okay. 18 Q. "The question in making prepayment rate 19 assumptions involves trying to estimate the 20 ultimate stabilized levels for these issues as 21 well as the band of the coupons that eventually 22 will show some activity. Rules of thumb for the 2977 1 rate savings necessary to make refinancing 2 worthwhile vary between two and a half to three 3 and a half percent. This implies that the edge of 4 the refinancing bands for Ginnie Maes should be 5 somewhere between Ginnie Mae 14 and 15s, which is 6 consistent with the experience so far." 7 Now, earlier, I had questioned you 8 about what the thinking was concerning rules of 9 thumb back in the Eighties. This indicates that 10 Mr. Waldman believed, at least in 1983, that those 11 rules of thumb were two and a half to three and a 12 half percent, correct? 13 A. That appears to be his thinking for 14 '83. 15 Q. And if one based -- models based upon 16 that experience, then you would have had some 17 severe problems in 1986, wouldn't you? 18 A. You would have -- if that was your 19 model, you would have underestimated the 20 prepayments in 1986. 21 Q. Severely? 22 A. Yes. 2978 1 Q. Right? 2 A. Yes. 3 Q. I want to ask you some questions 4 about -- 5 MR. NICKENS: Your Honor, this document 6 is in evidence in another form but since I have 7 read from it, I would offer -- what is the number? 8 It's B377. It's confusing to me because it's 9 written on here 377, but I'll take -- 10 THE COURT: All right. 11 MR. NICKENS: This is the one I've been 12 reading from in that form, and I'm going to offer 13 that because it includes the exhibits. 14 MR. RINALDI: Is it 377 or 337? You 15 said 377 when you started using this article. 16 MR. NICKENS: I am told that it is 377 17 and I was -- there was perhaps some confusion 18 because it happened that my little tab had 337 on 19 it. But I am corrected and I'm told that it 20 should be 377. 21 MR. GUIDO: The document that I have 22 has B377 on it. I mean, I don't know what anybody 2979 1 else has. 2 MR. NICKENS: Okay. Good. I just -- 3 MR. GUIDO: I think the witness has 4 B377. And this with the attachments is what came 5 from the Kestrel warehouse? Is that it? 6 MR. NICKENS: I mean, Your Honor, I 7 can't testify -- 8 MR. GUIDO: The code numbers are 9 supposed to identify where the document came from. 10 MR. NICKENS: And we have an agreement 11 on that and I just don't remember what they are. 12 MR. GUIDO: I have no objection to the 13 use of this rendition of the document if CN refers 14 to the documents taken out of the Kestrel 15 warehouse. 16 THE COURT: All right. Exhibit B377 is 17 received. 18 MR. GRIFFITH: I just want to correct 19 something on the record. CN does not mean that 20 it's from the Kestrel warehouse. You have to look 21 at a specific CN number. CN is just a control 22 number. 2980 1 MR. GUIDO: Okay. I'm sorry. I 2 thought most of the CNs came from the Kestrel 3 warehouse. 4 MR. GRIFFITH: Just because of the 5 volume of production, that's probably true that 6 it's out of the Kestrel warehouse. 7 THE COURT: All right. Thank you. 8 Q. (BY MR. NICKENS) Mr. Smith, I'm 9 handing you the documents that have been -- 10 A. Are these two separate documents? 11 Q. They are. 12 A. Okay. 13 Q. And I'm going to first ask you about -- 14 it's Tab No. 176 and it's Exhibit No. A10566. 15 A. 66. Okay. 16 Q. And it is the letter of July 12th, 17 1985, from Mr. Gerald Williams to Mr. Robert S. 18 Bonczak. 19 A. Okay. 20 MR. GUIDO: I always thought that our 21 arrangement was if you're using any exhibits, that 22 you were going to either notify us or you were 2981 1 going to have them in the courtroom. This is the 2 second time today that I haven't had the exhibit. 3 MR. NICKENS: My understanding of the 4 arrangement, Your Honor, is that if we are using 5 exhibits that have already been admitted, then we 6 don't have to provide additional copies. If we're 7 using new exhibits, then we provide additional 8 copies. Now, having said that -- 9 MR. RINALDI: The problem is if you 10 tell us in advance, that's fine. 11 MR. NICKENS: Well, let's see. It's 12 Tab 17 of 178 and it's Exhibits A10566 and A10575. 13 Q. (BY MR. NICKENS) Mr. Smith, I'd like 14 to refer you to the first page of Exhibit A10566 15 which is the letter of July 12th and particularly 16 this paragraph numbered No. 1. "The association's 17 use of reverse repurchase agreements and purchase 18 of mortgage-backed securities has increased from 19 $60 million at the base period of November 30th, 20 1984, to 539 million at June 30th, 1985. All of 21 these investments are fully hedged and the related 22 interest rate gap exposure has been reduced to the 2982 1 maximum amount possible." 2 What would you understand that term 3 "fully hedged" to mean consistent with our earlier 4 discussion about that term? 5 A. I would assume that "fully hedged" 6 meant that you had something along the lines of 7 the same dollar amount of notional and the swaps 8 that you do on the mortgages. It's the second 9 extension to that sentence that would make me 10 think that part about "reduced to the maximum 11 amount possible" would indicate to me that you 12 would have a very, very flat expected return for a 13 wide range of interest rates. 14 Q. And, in other words, you would expect 15 the amount of the -- in a risk-controlled 16 arbitrage that we have been discussing, you would 17 expect the amount of the swaps to be -- the 18 notional amount to be approximately the amount of 19 these assets? 20 A. Yes. Fairly close. 21 Q. And you would never understand it to 22 mean that they were telling you that they could 2983 1 not lose money on these securities under any 2 circumstances? 3 A. From the "fully hedged," they could 4 still lose money in a fully hedged position would 5 be my understanding if they mismanaged the hedge 6 going forward or if -- there are several things 7 that may possibly occur. It's the "reduced to a 8 maximum amount possible" I think that would give 9 me more comfort than the "fully hedged" as to how 10 safe this transaction is. 11 Q. Okay. And they could lose money in 12 particular if interest rates declined -- 13 A. Uh-huh. 14 Q. -- in comparison with those people who 15 didn't have the hedges at all. Right? 16 A. That would be -- 17 Q. So, you're still subject to prepayment 18 risk? 19 A. You're still subject to prepayment risk 20 but since you've hedged the maximum amount 21 possible, you should be fairly immunized against 22 prepayment risk. 2984 1 Q. Well, by making -- what you've 2 protected against is rising rates, haven't you? 3 A. I don't know what they protected 4 against. 5 Q. Well, if you have a fully hedged 6 position, that's what I'm saying. 7 A. You have hedged against rising rates 8 and you have given up some potential earnings in a 9 down rate environment to buy that protection. 10 Q. You have -- the tradeoff that's done 11 here is that you have protected against rising 12 rates at the expense of being able to benefit from 13 lower rates? 14 A. That's correct. 15 Q. Now, it says further that "United's 16 management believes that its growth is 17 consistent" -- this is at the bottom of the page 18 after Paragraph 2. "United's management believes 19 that its growth is consistent with the principles 20 of safety and soundness and provides a stable 21 income with a very limited interest rate risk." I 22 know this is very difficult to read. "The 2985 1 decisions made in implementing these investment 2 policies were based on the premise of reducing the 3 association's interest rate exposure." 4 Now, do you see anything inconsistent 5 with those two statements? 6 MR. GUIDO: In what? 7 MR. NICKENS: Between the two I just 8 read. 9 A. The two sentences in that last 10 paragraph? 11 Q. (BY MR. NICKENS) Yes, sir. The one 12 that we read numbered Paragraph 1 and the 13 statement of management's belief that its growth 14 was consistent with the principles of safety and 15 soundness, et cetera. 16 A. Can you read that last sentence in the 17 second paragraph because I've just got Morse code 18 here. 19 Q. It says "The decisions made in 20 implementing these investment policies were based 21 on the premise of reducing the association's 22 interest rate exposure." 2986 1 A. I think if you read those two in tandem 2 and cut out the second paragraph that I haven't 3 read, that they would flow and they would be 4 consistent in terms of what they were attempting 5 to say. 6 Q. They are saying that we have tried to 7 hedge against interest rate exposure by being 8 fully hedged. Right? 9 A. What they have said is that they have 10 reduced their gap exposure to the maximum amount 11 possible. 12 Q. Okay. Let me ask you to turn over to 13 Page CN52883. And I'm going to read the paragraph 14 with regard to investments and mortgage-backed 15 securities. "During 1984, United established an 16 investment department to coordinate a brokerage CD 17 program and the investment of these funds in 18 corporate and mortgage-backed securities. The 19 deposits generated by this program generally have 20 maturities of from 2 to 12 years and are matched 21 against high-yield securities with similar 22 maturities and duration. This approach virtually 2987 1 locks in a spread between USAT's asset yield and 2 funding costs, giving significant protection 3 against interest rate fluctuations. Additionally, 4 this department is upgrading United's investment 5 portfolio to increase its overall yield." 6 Do you see anything there, Mr. Smith, 7 deceptive or misleading? 8 A. Well, I don't -- 9 MR. GUIDO: With regard to what? 10 MR. NICKENS: This is about -- 11 MR. GUIDO: I have an objection. I 12 mean, the question doesn't indicate what it's 13 misleading with regard to. 14 MR. NICKENS: Well, I think it's 15 misleading as to nothing, Your Honor. But this 16 man has -- this is talking about the junk bonds as 17 well as the mortgage-backed securities which he 18 said that he had some familiarity with. 19 MR. GUIDO: Well, the witness I think 20 also testified earlier that he didn't look to see 21 whether or not the junk bond portfolio had been 22 traded as a trading portfolio as opposed to held 2988 1 as an investment portfolio. 2 MR. NICKENS: Let me rephrase the 3 question, Your Honor. 4 MR. GUIDO: Thank you. 5 Q. (BY MR. NICKENS) Assuming that you 6 have a risk-controlled arbitrage for the 7 mortgage-backed securities and with regard -- what 8 they say here, deposits matched against as to 9 maturities on the junk bonds, okay? 10 A. Uh-huh. 11 Q. Would you see anything wrong with the 12 statement "this approach virtually locks in a 13 spread between United's asset yield and funding 14 cost giving significant protection against 15 interest rate fluctuations"? 16 A. Assuming they do what they said they 17 did -- 18 Q. Yes, sir. 19 A. -- then I think that's consistent. I 20 probably wouldn't have been that strident if I 21 were writing it and left myself an out. 22 Q. You would have been a little -- you 2989 1 would have had a few more conditions in there? 2 A. Yes. 3 Q. But you're not misled by reading it? 4 A. When I read that, what that is -- and 5 maybe it's the extra 14 years of experience. But 6 what that means is they are attempting to maturity 7 match and they are attempting to hedge and they 8 believe if they have done everything that they 9 have indicated they are going to do in this 10 paragraph, that they have protected their yield 11 spread I think is what they are talking about. 12 Q. Okay. Let me ask you to go over to 13 Page 52885. It's at the bottom of the page that I 14 want to focus on. The program for matching 15 maturities and controlling interest rate risk. 16 Same subject. Right? 17 A. Uh-huh. I guess. 18 Q. "The impetus for United's liability 19 growth in the first two quarters of 1985 was 20 attributed to a program in which investments in 21 mortgage-backed and corporate securities were 22 matched with liabilities of similar maturity and 2990 1 duration. Specifically, this asset liability 2 match program supported incremental asset growth 3 with minimal interest rate risk and consisted of 4 mortgage-backed securities, Fannie Mae, Freddie 5 Macs, and Ginnie Maes purchased and funded with 6 reverse repurchase agreements. Concurrent with 7 the transaction, interest rate swaps were 8 initiated which effectively lengthened the 9 maturity and duration of the liabilities and 10 locked in a net interest spread." 11 And then, two, "Long-term corporate 12 securities were purchased and match funded with 13 retail broker deposits of similar maturity and 14 duration which also locked in a net interest 15 spread." 16 Now, was that term consistent with the 17 use at the time of the term "locked in"? 18 A. I think in the financial market -- when 19 was this done? 20 Q. This is 1985. 21 A. I think certainly I can talk to the 22 '86, '87 time frame where people were talking 2991 1 about locking in a spread. There was generically 2 some sort of risk that could make that lock go 3 away. So people would -- 4 Q. But still people used the term? 5 A. What people would say were things like 6 "With respect to the maturity matching of these -- 7 this asset to that liability, as long as the asset 8 doesn't default, I have locked in a spread." 9 So, there were usually some unstated 10 caveats that went along with "locked in." 11 Q. And if you look back over at Page 12 52891, you see there "The investments, 13 mortgage-backed securities" -- again, the similar 14 paragraph here. Mr. Williams or whoever writes 15 this expresses it as "this asset liability match 16 program virtually locks in a spread between 17 United's asset yield and funding costs giving 18 significant protection against interest rate 19 fluctuations." 20 So, there are a few more conditions in 21 there. Right? 22 A. Uh-huh. (Witness nods head 2992 1 affirmatively.) 2 THE COURT: Would you speak up, please? 3 A. I'm sorry. Yes. I guess my question, 4 if I were reading this, is were these truly their 5 expectations and were they taking the actions to 6 make these things occur and, if so, then that 7 paragraph is consistent. 8 Q. (BY MR. NICKENS) Should, in your 9 view, people be liable for $250 million 15 years 10 later based upon use of those terms? 11 MR. GUIDO: Objection. The question is 12 the use of those terms in relationship to what, 13 and this witness has testified he's never looked 14 at the USAT portfolio -- 15 THE COURT: All right. I'll sustain. 16 Q. (BY MR. NICKENS) Let me ask you to 17 look over at Page 52892. This is the next page. 18 And do you see there -- and you'll need to look at 19 the next two pages -- that this person is trying 20 to illustrate the effects of certain hedging 21 instruments -- in the first instance, swaps. In 22 the second instance, caps, and a third instance, 2993 1 collars? 2 A. Uh-huh. (Witness nods head 3 affirmatively.) 4 Q. And it indicates a decrease in rates of 5 two and a quarter points and an increase of rates 6 if you assume the first column is current up to 13 7 and a half? 8 A. Okay. 9 Q. Is this a fairly typical presentation 10 for illustrating something such as the effect of 11 swaps, different types of hedging instruments? 12 A. Generally the way it was done -- there 13 were really two ways to do it. This was one way. 14 The other was to put cash values in this and then 15 calculate the returns at the end. They should 16 have led to the same results, though. This is 17 certainly one of the two ways I've seen it -- I 18 was seeing it done. 19 Q. Okay. Let me ask you to look briefly 20 over at Exhibit 10575 which is the October 28th, 21 1985, letter. 22 MR. GUIDO: 10575. 2994 1 MR. NICKENS: Yes. Well, that's okay, 2 Your Honor. I think I will forego that. Pretty 3 much the same. 4 Your Honor, at this time, I would be 5 moving to the junk bond subject which I estimate 6 will take approximately an hour. Mr. Smith has 7 indicated that he cannot be here tomorrow. We 8 have a witness that is coming from out of town, 9 one of the experts, but I believe he's indicated 10 that he might be able to come back on Thursday 11 morning to complete his testimony. 12 Is that where we are? 13 MR. GUIDO: That's where we are, Your 14 Honor. The witness tomorrow has a very heavy 15 teaching schedule. He's a professor at Stanford 16 and the next time he'll be available is sometime 17 in November. So, we would like to make sure that 18 we have his time available tomorrow to put him on. 19 THE COURT: All right. I'll suspend 20 the examination of this witness and we'll adjourn 21 till tomorrow at 9:00 o'clock. 22 MR. NICKENS: Thank you, Your Honor. 2995 1 (Whereupon at 4:39 p.m. 2 the proceedings were recessed.) 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 2996 1 STATE OF TEXAS COUNTY OF HARRIS 2 REPORTER'S CERTIFICATION 3 TO THE TRIAL PROCEEDINGS 4 I, Marcy Clark, the undersigned Certified 5 Shorthand Reporter in and for the State of Texas, 6 certify that the facts stated in the foregoing 7 pages are true and correct to the best of my ability. 8 I further certify that I am neither 9 attorney nor counsel for, related to nor employed 10 by, any of the parties to the action in which this 11 testimony was taken and, further, I am not a 12 relative or employee of any counsel employed by 13 the parties hereto, or financially interested in 14 the action. 15 SUBSCRIBED AND SWORN TO under my hand 16 and seal of office on this the 14th day of 17 October, 1997. 18 ____________________________ MARCY CLARK, CSR 19 Certified Shorthand Reporter In and for the State of Texas 20 Certification No. 4935 Expiration Date: 12-31-97 21 22 2997 1 STATE OF TEXAS COUNTY OF HARRIS 2 REPORTER'S CERTIFICATION 3 TO THE TRIAL PROCEEDINGS 4 I, Shauna Foreman, the undersigned Certified 5 Shorthand Reporter in and for the State of Texas, 6 certify that the facts stated in the foregoing 7 pages are true and correct to the best of my ability. 8 I further certify that I am neither 9 attorney nor counsel for, related to nor employed 10 by, any of the parties to the action in which this 11 testimony was taken and, further, I am not a 12 relative or employee of any counsel employed by 13 the parties hereto, or financially interested in 14 the action. 15 SUBSCRIBED AND SWORN TO under my hand 16 and seal of office on this the 14th day of 17 October, 1997. 18 _____________________________ SHAUNA FOREMAN, CSR 19 Certified Shorthand Reporter In and for the State of Texas 20 Certification No. 3786 Expiration Date: 12-31-98 21 22