Free Motion for Summary Judgment - District Court of Federal Claims - federal

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 1 of 42

IN THE UNITED STATES COURT OF FEDERAL CLAIMS

) STERLING SAVINGS ASSOCIATION, a state-chartered savings association, STERLING FINANCIAL CORPORATION, a Washington corporation ) ) ) ) ) Plaintiffs, v. ) ) ) UNITED STATES OF AMERICA, ) ) Defendant. ) ) ) No. 95-829-C (Judge Wheeler)

UPDATED AMENDED EXPERT REPORT OF MUKESH BAJAJ, Ph.D.

February 15, 2007

App. 776

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 2 of 42

Sterling would have offset the lower spread on its incremental assets and liabilities through RTC acquisitions is unverifiable and speculative. 6.4.7.2 Dr. Horvitz's Assumptions Regarding Sterling's Wholesale Growth Are Speculative

Apart from retail growth, a thrift, like any entity, can acquire wholesale assets such as mortgage-backed securities (MBS). Wholesale markets for financial assets are

international markets and the prices for assets in those markets are set by worldwide competition. A thrift can also fund assets using market-priced wholesale liabilities like Federal Home Loan Bank (FHLB) advances or reverse repurchase agreements although the ability to use these sources is limited by the availability of sufficient acceptable collateral.158 However, a thrift has no comparative advantage in tapping the wholesale

markets for assets or liabilities, and a strategy of adding market priced assets funded with wholesale liabilities to its balance sheet is not a source of economic profit to a thrift. In 1993, Sterling implemented a growth strategy predicated on buying wholesale assets funded with wholesale liabilities with the intention of replacing these assets over time with retail assets and liabilities.159 As a result, Sterling's assets grew by $399 million, or 61%,160 over the prior year. The bulk of Sterling's growth was attributable to an increase of 245% in mortgage-backed securities (MBS) and investments.161 As a percent of total assets, MBS and investments grew from 17% in 1992 to 36% in 1993. The funds to purchase these additional assets came primarily from FHLB advances, which increased by 243% between 1992 and 1993.162 Dr. Horvitz readily admits that the foregone assets and liabilities would have a lower spread than average:

I also note that as a member, Sterling was required to purchase stock in the FHLB of Seattle according to a formula based on the amount of its outstanding loans and advances. (Sterling Financial 1993 Form 10K, p. 19 [VSG001 0458]).
159 160 161 162

158

ALCO Minutes, December 14, 1992 [WON935 1493-1494]. Revised Updated Horvitz Report, Exhibit 6, Table 3e. Sterling Financial 1994 Annual Report, p. 1 [WON8871543]. Ibid.

App. 777

50

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 3 of 42

Q. All right. Why would the addition of assets decrease the net interest margin? A. Because they're assets -- wholesale assets being funded with wholesale liabilities at a narrower spread. Q. Does that mean the incremental assets have a lower than average net interest margin? A. These incremental assets, yes.163 Thus, in 1993, by Dr. Horvitz's own admission, Sterling's incremental assets earned a lower rate of return than the rest of Sterling's asset portfolio. As noted in its ALCO minutes, Sterling understood this as well: The wholesale nature of the growth strategy will pull down the interest margin in the future.164 To the extent that the growth in Dr. Horvitz's model is wholesale growth, his assumption that Sterling would be able to earn the equivalent of its actual ROAA on its incremental assets is also not credible. Also, Sterling's growth strategy was implemented with the intention of replacing wholesale assets and liabilities with loans and deposits. However, at the time its growth strategy was implemented, Sterling expressed concerns that it would be unable to generate assets to replace the wholesale assets: Norm Judd presented a draft proposal to the Board of Directors for the growth strategy...Howard Gilkey emphasized that the important component of this plan is the ability to replace wholesale activity with retail activity over the next few years. Norm Judd noted that the Equity/Assets ratio grows over the next five years as we are unable to generate earning assets as fast as the increase in equity. As our existing portfolio of balloon mortgages and securities begins to mature over the next few years the need to generate earning assets will increase. While we have successfully emphasized consumer and commercial loans, we are adding about $2 million in total balances monthly...If we just double

163 164

Horvitz Deposition, December 6, 2001, p. 451. ALCO Minutes February 8, 1993 [WON935 1487]; ALCO Minutes March 15, 1993 [WON935 1483].

App. 778

51

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 4 of 42

Similarly, in fiscal 1993, Sterling sold its mortgage-backed securities portfolio, which resulted in a gain of $0.638 million.180 Although Dr. Horvitz has adjusted the but-for Sterling's ROAA for the penalties from the one-time prepayment of FHLB advances, he has not adjusted it for this one-time sale of Sterling's fixed rate securities portfolio. Dr. Horvitz has selectively adjusted Sterling's but-for ROAA for non-recurring items. This approach is not credible. Dr. Horvitz has also made adjustments to the but-for Sterling's ROAA for its claimed "wounded bank" costs. I discuss the flaws in some of those claimed costs in the next section. 6.5 Dr. Horvitz's Presentation Of "Wounded Bank" Damages Is An Uncritical Repetition Of Sterling Management's Views And Obviously Flawed Since filing his initial report, Dr. Horvitz has reduced the claimed amount of the "Wounded Bank Damages" he includes in his damages estimate to $4.556 million from $5.278 million.181 These damages consist of: · · · · · Cost of Great American branch denied because of OTS Capital Standards ($1.835 million) Loss on CJ-4 loan ($1.241 million) Excess Supervision Costs ($172,000) Cost of 1989 Units Offering ($1.004 million) Increased Legal and Accounting [Costs] to Defend against Regulatory Interference ($304,000).182 For all of these categories, Dr. Horvitz relies upon conversations with, and calculations prepared by, Daniel Byrne, CFO of Sterling, without any independent review of underlying documents.183

180 181 182 183

Sterling 1993 Annual Report, p. XIX [PSG027 0540]. Horvitz 2006 Report, Exhibit 8. Ibid. Horvitz Deposition, December 6, 2001, pp. 252-253, December 7, 2001, pp. 553-554.

App. 779

56

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 5 of 42

Dr. Horvitz does not even appear to have provided a conceptual framework for the wounded bank damages to Mr. Byrne. He simply accepted a damages calculation

provided him by his client and included it in his expert report. He did not direct the analysis, he did not review any primary documents and he did not even discuss any part of the wounded bank damages, other than the CJ-4 loan, with anyone other than Mr. Byrne.184 Thus, Dr. Horvitz has simply put forth the views of Sterling's management without any critical analysis or input to the process. These claims violate standard economic theory, and are speculative, as I discuss below. CJ-4 Loan Loss: According to Mr. Byrne, the CJ-4 loan resulted in a loss of $1.241 million to Sterling.185 Mr. Byrne alleges these damages occurred because the "Government was enforcing the limitations of loans to one borrower..."186 As a result, Mr. Byrne claims that Sterling could not advance additional funds to the borrower but had to instead foreclose and repossess the property and make improvements on it. The thrift then lost money in liquidating the property later since its market value had declined.187 Mr. Byrne concludes that Sterling would have been better off had it not been forced into foreclosing the property and simply made an incremental loan. That is, Mr. Byrne assumes, and Dr. Horvitz accepts without any further analysis, that Sterling's incremental return on additional investments in the CJ-4 loan would have been greater than its average return. In the absence of solid evidence, this claim is purely speculative from an economics perspective. Moreover, by drawing such a distinction between incremental and average returns here, Dr. Horvitz contradicts the premise behind his lost profits analysis for the 1989-2006 period. As discussed earlier, in estimating Sterling's alleged lost profits during 1989-2006, Dr. Horvitz had claimed that Sterling's incremental returns earned on foregone assets would have equaled the thrifts' actual average rate of return.

184 185 186 187

Horvitz Deposition, December 7, 2001, pp. 555-556. Analysis Prepared by Daniel G. Byrne, 1/9/04 [VSG034 0074] Byrne Deposition, December 3, 2001, p. 70. Ibid, pp. 70-71.

App. 780

57

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 6 of 42

Cost of Great American Branch denied: The damages estimate for the Great American branch should have led Dr. Horvitz to question the calculation. According to Mr. Byrne, Sterling would have paid $61,000 to acquire two branches in 1989.188 According to Mr. Byrne, the thrift was prevented from acquiring these branches due to inadequate capital.189 Later, Sterling did acquire the Spokane branch through a Resolution Trust Corporation ("RTC") sale in the 1994-1995 period,190 paying a $1.892 million deposit premium.191 Mr. Byrne alleges, and Dr. Horvitz proposes, without additional analysis,192 that the difference between this deposit premium paid in 1994 and the 1989 charge of $61,000 (plus transaction costs) calculated by Mr. Byrne represents damages to Sterling.193 While an increase in value of 6,203% in five years is not impossible, on the surface it is unlikely unless there are additional factors to explain the increase. One difference that could account for the increased value of the Spokane branch is that its deposits increased from $16.2 million as of June 30, 1989 to $26.2 million at the time Sterling acquired it from the RTC. I have not been able to determine if any capital improvements were made to the branch that could have also affected its value.194 Paying a higher price for a more valuable asset is not necessarily a source of damages. Absent additional information, Dr. Horvitz's claim is not verifiable. Dr. Horvitz's claim contradicts economic logic and once again reveals the fallacy of ex post analysis in damages calculations. Asset values change over time due to changes in the supply and demand conditions for the asset. By not investing in the branches in 1989, Sterling avoided the risk of the value of these branches declining in the future. Had the

188 189 190

Byrne Deposition, December 4, 2001, p. 137. Ibid., p. 134.

Ibid, p. 134. The purchase actually occurred in March 1994. See RTC New Release [VSG015 05650571]. Ibid, p. 137. Horvitz Deposition, December 7, 2001, pp. 530-531. Analysis Prepared by Daniel G. Byrne, 1/9/04 [VSG034 0085].

191 192 193 194

Sterling Savings Association Acquisition of Two Branches from Great American, June 30, 1989 [VSG014 0577]; Great American Bid Summary [VSG014 0564].

App. 781

58

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 7 of 42

branches actually declined in value, it is unlikely that Dr. Horvitz would have included Sterling's benefit of having avoided such a poor investment due to FIRREA. Excess Supervision Costs: Sterling's claim for excess supervision costs include $33,711 for the cost of its January 31, 1990 OTS examination and $138,266 for the incremental amount of its FDIC general assessments attributable to its status as a "troubled bank" from March 1990 to September 1991.195 By including such claims as wounded bank damages, Dr. Horvitz assumes that Sterling would not have faced such costs in the but-for world. Regulators supervise thrifts to ensure the preservation of the deposit insurance fund. Excess risk taking by thrifts adversely affects the deposit insurer's liability. Therefore, it is likely that even in the but-for world Sterling would have faced what Dr. Horvitz terms "regulatory interference." In analyzing his but-for world, Dr. Horvitz makes no attempt to

disentangle such potential grounds for regulatory actions (and their associated costs to Sterling) that were related to the alleged breach, from those unrelated to the alleged breach. Cost of 1989 Units Offering: I understand that Dr. William Hamm will explain why such costs are not attributable to the breach. In summary, Dr. Horvitz performs no formal economic analysis on his own to establish the legitimacy of Plaintiffs' wounded bank damages claims. As I discussed above, his wounded bank damages claims are speculative. 6.6 Dr. Horvitz's Expectation Damages From 2006 Onward Are Conceptually Flawed According to Dr. James, Sterling would have had $1.726 million in supervisory capital remaining as of September 30, 2006, and he estimates that the "future damages" from

Sterling Savings Bank Calculation of Damages, Regulatory Supervisory Costs, prepared by Daniel G. Byrne [VSG034 0076].

195

App. 782

59

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 8 of 42

any time a corporation finances the purchase of a safe asset through the issuance of a risky security. The spread simply represents the cost of the resulting reduction of balance sheet risk. Such spread has nothing to do with FIRREA and nothing to do with damages in this case. Consider the fact that corporations often finance risky investments by issuing securities of lower risk. They expect to earn a positive spread in this case. Clearly, if Sterling had issued securities of lower risk than its investments, Dr. James would not have concluded there were negative damages. The apples-to-oranges comparison of yields on securities of different risk in Dr. James's model contradicts the central principle of financial economics, namely, the difference in expected returns between two assets must solely be attributed to the differences in systematic risk of these asset's returns for their investors.215 This principle is the bedrock of the Nobel Prize winning Capital Asset Pricing Model ("CAPM") as well as alternative asset pricing models that have been developed in financial economics.216 For example, the basic reasoning behind the Arbitrage Pricing Theory ("APT") is that if two assets with the same systematic risk have different expected returns, a well-diversified investor can make arbitrage profit by buying the asset with the higher expected return and selling short the other asset. For instance, if the lower yield on Treasury securities were not accompanied by a concomitantly lower risk (which Dr. James ignores) then investors could sell Treasuries and buy preferred stock to take advantage of this "arbitrage" opportunity, i.e., to earn higher returns without any additional risk. However, such arbitrage would be impossible in competitive capital markets since the preferred stock's price would rise (or its return

According to asset pricing models developed in finance, the expected return on an asset depends on its systematic risk or non-diversifiable risk. The main idea is that, in a competitive market, investors who hold diversified portfolios will not be concerned about the diversifiable risk of an investment and will only require a risk premium to compensate them for non-diversifiable risk. These investors will set market prices in equilibrium and asset returns will therefore depend only on systematic risk. William F. Sharpe and John Lintner independently derived the CAPM. William F. Sharpe received the Nobel Prize in Economics in 1990 for developing the CAPM. Sharpe, W. F., "Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk," Journal of Finance, September 1964. Lintner, J., "Security Prices, Risk and Maximal Gains from Diversification," Journal of Finance, December 1965.
216

215

App. 783

65

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 9 of 42

would decline) in response until any return differential between the two securities would be solely due to their different levels of risk. In equilibrium, all securities trade at their fair risk-adjusted prices. Financial economists describe this principle by saying that transactions in fairly priced securities have zero net present value (NPV). In wellfunctioning capital markets, investors in Sterling's risky preferred stock would pay a price that would be fair given the risk they would bear for holding such securities. Similarly, Sterling would also be able to retire its liabilities or invest the proceeds in buying risk-free Treasury bonds at a fair price. The sale of one fairly priced security to invest in another will not cost more than the transaction costs incurred. Dr. James fails to note the implicit risk reduction benefit associated with his proposed capital replacement financing scheme. By issuing preferred stock and using the proceeds to pay off its liabilities as Dr. James hypothesizes, Sterling would have not only have replaced its supervisory capital but would have also lowered the risk of its balance sheet. The spread component of Dr. James's damages model is simply the cost of such risk reduction. Proper measure of the cost of replacement of supervisory capital should do just that, replace capital, without an expensive reduction in risk. The proper measure of replacement cost is simply the transaction costs. If Sterling were to raise capital and make investments that result in net increase in its balance sheet risk, the spread between its financing costs and incremental income would be negative. Clearly, Dr. James would not conclude that in such a case Sterling's damages should be negative as such a spread income is simply compensation for an investment decision to bear additional risk and it should not be mixed up with the cost of replacement capital. Some examples unrelated to Sterling's replacement of supervisory capital will help to further illustrate the methodological error in Dr. James's report. Suppose a company such as GE had issued AAA-rated debt with a yield of 9% and invested the proceeds in junk bonds yielding 15%. GE would have earned a positive spread of 6% but it would not have added any value for its shareholders. This is because the increase in risk of GE's assets as a result of its investment in junk bonds would have raised GE's cost of capital by an amount sufficient to offset its additional income from the positive spread.

App. 784
66

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 10 of 42

8.3.9.5

Dr. James's Adjustments To The But-For Sterling's ROAA For Sterling's Claimed "Wounded Bank" Costs Are Not Supportable

Dr. James has adjusted the but-for Sterling's earnings for Sterling's claimed "wounded bank" costs.298 These costs include the loss on the CJ4 Loan, excess supervision costs and claimed damages relating to the price that Sterling paid for a branch in Spokane, Washington.299 As I discussed Section 6.5 above, these claims are speculative. Thus, there is no basis for these "wounded bank" adjustments to pre-tax, pre-amortization earnings. 8.4 Transaction Costs That Sterling Would Not Have Incurred Absent The Phase Out Of Supervisory Capital Are Reasonable Damages The third component in Dr. James's model is transaction costs. Dr. James includes in Sterling's damages the difference between the gross and net proceeds received from its 1991 common stock offering or $2.767 million. To this amount, he adds $0.560 million, his estimate of the value of the warrants issued to the underwriters as part of their compensation for their services. As noted above, he then makes a pro-rata adjustment to the total of these two amounts to account for the portions of the capital raised in the 1991 offering that he allocates to replacing the capital from Sterling's withdrawn 1989 offering and to replacing the Central Evergreen goodwill. I agree that to the extent that the breach caused Sterling to incur transaction costs that it would not have otherwise incurred, these costs are appropriately viewed as damages. It is my understanding that Sterling asserts that it had an obligation to raise capital based upon the Central Evergreen agreements. As I discussed above, to the extent that the loss of its supervisory capital caused Sterling to have to raise a larger amount of capital than it otherwise would have, the incremental costs associated with raising that capital are a reasonable measurement of damages.

298 299

Revised Second James Report, Exhibit E, Table 3d. Revised Second James Report, Exhibit E, Tables 3c, 3d and 3e.

App. 785

102

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 11 of 42

App. 786

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 12 of 42

App. 787

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 13 of 42

App. 788

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 14 of 42

App. 789

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 15 of 42

App. 790

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 16 of 42

App. 791

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 17 of 42

App. 792

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 18 of 42

App. 793

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 19 of 42

App. 794

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 20 of 42

App. 795

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 21 of 42

App. 796

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 22 of 42

App. 797

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 23 of 42

App. 798

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 24 of 42

App. 799

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 25 of 42

App. 800

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 26 of 42

App. 801

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 27 of 42

App. 802

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 28 of 42

App. 803

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 29 of 42

App. 804

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 30 of 42

App. 804A

Case 1:95-cv-00829-TCW

Document 222-26


Filed 02/26/2007

Page 31 of 42

Expert Report of W. Barefoot Bankhead
Sterling Savings Association, et al. v. United States of America Case No. 95829 C I. Qualifications 1. My name is W. Barefoot Bankhead, and I am a Managing Director with Navigant Consulting, which provides professional services to its clients in various areas, including financial, accounting and economic analysis. I graduated from the University of Texas in December 1976 with a Bachelor of Business Administration degree and a major in accounting. I am a certified public accountant in the State of Texas. From 1977 through 1987, I worked in public accounting, the last several years of which were with a then "big eight" firm, primarily serving clients in the financial services and real estate industries.

2. In April 1987, I began a twoyear accounting fellowship with the Federal Home Loan Bank Board, now the Office of Thrift Supervision, the primary regulatory body for the savings and loan industry. After my fellowship and prior to joining Navigant Consulting in August 1998, I worked for nine years in a variety of positions with financial institutions. During that time, I served as an audit director for one of the largest bank holding companies in the country and as chief financial officer and member of the board of directors for a savings bank. My current resume is provided as Exhibit A. My hourly billing rate is $468.

App. 805

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 32 of 42

II.

Background 3. From November 1985 through December 31, 1988, Sterling Savings

Association ("Sterling") acquired the assets and liabilities of three thrift institutions through a series of assisted acquisitions as reflected in Table 1.

Table 1. Thrift Acquisitions by Sterling Savings Association
Date November 5, 1985 April 8, 1988 December 31, 1988
Source: Sterling Savings Association, Annual Report for the years ended June 30, 1986 (pp. 2223), 1988 (pp. 3031) and 1989 (pp. 3132) [SG2005 25262527; 24592460; 23992400]

Acquired Institution Lewis Federal Savings and Loan Association TriCities Savings and Loan Association Central Evergreen Savings and Loan Association

Location Chehalis, Washington Kennewick, Washington Chehalis, Washington



4. In conjunction with the Lewis Federal and TriCities acquisitions, Sterling and the Government entered into agreements that, among other things, provided certain forbearances related to the accounting for the goodwill created in the acquisitions.1 Sterling alleged that certain documents governing the Central Evergreen transaction provided a similar goodwill accounting forbearance.

5. The Financial Institutions Reform, Recovery and Enhancement Act ("FIRREA") was enacted on August 8, 1989. Capital regulations adopted as a result of FIRREA required, among other things, thrifts to deduct goodwill from regulatory capital.


Forbearance letters from the Federal Home Loan Bank Board ("FHLBB") to Sterling dated November 12, 1985 and April 6, 1988 grant certain accounting forbearances in the Lewis Federal and TriCities acquisitions, respectively.
1



App. 806

2

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 33 of 42

6. On September 12, 2002, the Court of Federal Claims ("the Court") initially determined that various contractual documents associated with all three of Sterling's acquisitions gave it the right to include goodwill in regulatory capital and that the enforcement of the FIRREA capital regulations represented a breach of the Government's commitment under those contracts. In an effort to estimate the damages associated with those breaches, Sterling submitted expert reports authored by Dr. Paul Horvitz and Dr. Christopher James. Dr. Horvitz's report, which was dated February 20, 2004 ("the Horvitz 2004 Report"), calculated alleged lost profits that resulted from the breaches. Dr. James's report, which was also dated February 20, 2004, calculated alleged damages related to the hypothetical raising of mitigation capital as replacement for the goodwill.2

7. On August 30, 2006, the Court amended its previous ruling and determined that the enforcement of the FIRREA capital regulations did not result in a breach of the Central Evergreen contract as there was no forbearance related to the accounting for the Central Evergreenrelated goodwill. Dr. Horvitz issued his Update to the Expert Report of Paul Horvitz ("Horvitz 2006 Report") and Dr. James issued his Update to the Expert Report of Dr. Christopher James ("James 2006 Report") on December 11, 2006, to amend the plaintiffs' damages claim reflecting the Court's updated ruling.

8. The United States Department of Justice ("DOJ") has asked me to evaluate Dr. Horvitz's calculation of core capital in the Horvitz 2006 Report. Because Dr.
Dr. Horvitz and Dr. James initially issued damages reports in 2001. However, during 2003, the Court issued an opinion related to the amortization of goodwill associated with certain Federal Savings and Loan Insurance Corporation ("FSLIC") assistance. The 2004 reports represent Dr. Horvitz's and Dr. James's efforts to amend their reports to reflect the impact of the Court's 2003 opinion related to the amortization of goodwill associated with FSLIC assistance.
2



App. 807

3

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 34 of 42

James has adopted Dr. Horvitz's lost profits methodology, DOJ has also asked me to evaluate the lost profit calculations in the James 2006 report. The documents I have considered in preparing this report are listed in Exhibit B. As my work in this matter is ongoing, if necessary, I will supplement my findings based on the results of my continuing work, or as a result of any new matters raised by plaintiffs' experts.

III. Summary of Opinions 9. Based upon my review and analysis of the documents cited in Exhibit B and my relevant education and experience, I have formed the following opinions:

·

Dr. Horvitz's calculation of butfor Sterling's core capital erroneously includes goodwill in excess of amounts permitted under the regulation governing the phaseout of goodwill. Correcting for this error eliminates any lost profits as calculated by Dr. Horvitz.

·

Even calculating butfor Sterling's core capital by allocating qualifying supervisory goodwill proportionally between contractual and non contractual supervisory goodwill, Dr. Horvitz's model produces no lost profits.

·

Since Dr. James has adopted Dr. Horvitz's lost profits calculation during the period between the breach and the date of full mitigation, his analysis of lost profits suffers from the same flaws that I have identified with Dr. Horvitz's model. Once the error in Dr. James's core capital calculations is corrected, Dr. James's model produces no lost profits.

App. 808

4

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 35 of 42

IV. Dr. Horvitz's Failure to Correctly Calculate Core Capital in the ButFor Bank 10. The FIRREA capital regulation established three capital requirements ­ tangible, core, and riskbased. Thrifts were required, among other things, to deduct all intangible assets, including goodwill, in the calculation of tangible capital; limit identified intangible assets included in core and riskbased capital; and phase out supervisory goodwill from core and riskbased capital over a five year period ending December 31, 1994.3 Through December 31, 1994, thrifts were not required to deduct qualifying supervisory goodwill in calculating core capital.4 Qualifying supervisory goodwill was limited to 1.5 percent of assets (as defined in the regulation) through December 31, 1991, and the includable amount then declined as a percentage of assets until January 1, 1995, when all supervisory goodwill was required to be deducted in calculating core capital.


12 CFR 567.1(ee) defined supervisory goodwill as goodwill resulting from the acquisition, merger, consolidation, purchase of assets, or other business combination (if such transaction occurred on or before April 12, 1989) of (1) a savings association where the fair market value of assets was less than the fair market value of liabilities at the acquisition date; or (2) a problem institution.
3

12 CFR 567.1(w) defined qualifying supervisory goodwill as (1) any unamortized goodwill (FSLIC Capital Contributions as reported in the September 30, 1989 Thrift Financial Report) that existed on April 12, 1989 resulting from prior regulatory accounting practices less any amortization that would have occurred subsequent to April 12, 1989 through the current reporting period where the amortization is calculated on a straightline basis over the shorter of 20 years, or the remaining period for amortization in effect on April 12, 1989 for regulatory accounting practices; plus (2) the lesser of (i) supervisory goodwill that is included in goodwill that is reflected in the current reporting period under generally accepted accounting principles ("GAAP"); or (ii)(A) supervisory goodwill that is included in goodwill that is reflected in the current reporting period under GAAP; (B) plus any amortization of the goodwill in (ii)(A) that occurred subsequent to April 12, 1989 for GAAP reporting purposes; (C) minus the amortization of the goodwill in (ii)(A) through the current period that results when the goodwill is amortized subsequent to April 12, 1989 on a straightline basis over the shorter of 20 years, or the remaining period for amortization in effect on April 12, 1989 for GAAP reporting purposes.
4



App. 809



5

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 36 of 42

11. Sterling's supervisory goodwill exceeded the amount that could be included as qualifying supervisory goodwill in core capital under the FIRREA capital regulation. Table 2 summarizes Dr. Horvitz's determination of Sterling's supervisory goodwill as of December 31, 1989.

Table 2. Sterling's Supervisory Goodwill, December 31, 1989
Institution Lewis Federal and TriCities (determined to be contractual) Central Evergreen (determined to be noncontractual) Total Supervisory Goodwill Qualifying Supervisory Goodwill (1.5% of butfor Sterlings assets) Amount (000s) $15,557 11,357 $26,914 $10,421



12. The Court has determined that the Government breached contracts allowing Sterling to include the Lewis Federal and TriCities goodwill in regulatory capital. Consequently, in the butfor world constructed by Dr. Horvitz, Sterling can include all of the $15,557,000 of Lewis Federal and TriCities supervisory goodwill in regulatory capital as of December 31, 1989.

13. In his 2006 Report, Dr. Horvitz assumes that Sterling's Lewis Federal and TriCities contractual supervisory goodwill is exempt from the limitations on the amount of supervisory goodwill that qualifies for core capital. This assumption is inconsistent with his 2004 report. In the Horvitz 2004 Report, Dr. Horvitz included contractual supervisory goodwill in his calculation of qualifying supervisory goodwill for purposes of determining the amount of disallowed intangibles due to the breach. (See Horvitz 2004 Report, Exhibit 6, Table 4b.) As illustrated in Table 3 below, Dr. Horvitz's 2006 Report erroneously assumes that

App. 810

6

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 37 of 42

an additional $10,421,000 of butfor Sterling's noncontractual supervisory goodwill was qualifying supervisory goodwill and includable in core capital.

Table 3. Dr. Horvitz's Calculation of ButFor Sterling's Core Capital at December 31, 1989 Amount (000s) ($12,581) 15,557 10,421 5,670 10,500 (781) $28,786







Tangible Capital at 12/31/89: Adjustments Related to Unidentified Intangibles Contractual goodwill (Lewis Federal and TriCities) Qualifying supervisory goodwill (Central Evergreen) Other Adjustments Identified intangibles Capital raising Incremental butfor bank adjustments ButFor Sterlings Core Capital

14. Butfor Sterling could consider all $15,557,000 of the Lewis Federal and TriCities contractual supervisory goodwill as qualifying supervisory goodwill even though this amount exceeded the $10,421,000 of qualifying supervisory goodwill otherwise permissible under the FIRREA capital regulation. However, Dr. Horvitz has erroneously included $25,978,000, or 3.74 percent of butfor Sterling's assets, as qualifying supervisory goodwill even though the FIRREA capital regulation permits only 1.5 percent and the contractual supervisory goodwill totals only 2.24 percent.5

15. The FIRREA capital regulation does not distinguish between contractual and noncontractual supervisory goodwill. In addition, the regulation clearly limits the amount of supervisory goodwill that qualifies as core capital. It is
5

See Exhibit C.

App. 811

7

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 38 of 42

implausible to assume that butfor Sterling would have been permitted to include the noncontractual Central Evergreen supervisory goodwill as qualifying supervisory goodwill when the contractual supervisory goodwill already exceeded the 1.5 percent maximum amount of qualifying supervisory goodwill includable in core capital as of December 31, 1989. To illustrate the impact of Dr. Horvitz's error, I have recalculated butfor Sterling's core capital using the information contained in the Horvitz 2006 Report:

Table 4. ButFor Sterling's Adjusted Core Capital at December 31, 1989 Amount (000s) ($12,581) 15,557 10,500 (781) 3,174 $15,869







Tangible Capital at 12/31/89: Adjustments Related to Unidentified Intangibles Qualifying supervisory goodwill (Lewis Federal and TriCities)6 Other Adjustments Capital raising Incremental butfor bank adjustments Identified intangibles 7 ButFor Sterlings Core Capital



16. Adjusting Dr. Horvitz's model for the one change to correct the calculation of qualifying supervisory goodwill includable in core capital as of


Qualifying supervisory goodwill as calculated by Dr. Horvitz totals only $10,421. However, because of the Court's opinions, butfor Sterling can include all of the Lewis Federal and Tri Cities supervisory goodwill for purposes of calculating regulatory capital.
6 7

The FIRREA capital regulation permits certain identified intangibles to be included in core capital up to a limit of 25 percent of core capital (before inclusion of identified intangibles).



App. 812



8

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 39 of 42

December 31, 1989, results in negative lost profits, thus eliminating Sterling's entire lost profits claim.8

17. As shown below in Table 5, the contractual supervisory goodwill included as qualifying supervisory goodwill as a result of the Court's decisions exceeds the amount permissible under the FIRREA regulation at all times through December 31, 1994. Consequently, butfor Sterling would have been required to deduct all noncontractual supervisory goodwill (i.e., goodwill associated with the Central Evergreen transaction) when calculating core capital. However, by erroneously including noncontractual supervisory goodwill as qualifying supervisory goodwill, Dr. Horvitz's butfor Sterling includes supervisory goodwill from two to nearly three times the amount permitted by the FIRREA capital regulation through December 31, 1994.

Table 5. ButFor Sterling's Total Supervisory Goodwill
Year Ended 12/31/1989 6/30/1990 6/30/1991 6/30/1992 6/30/1993 6/30/1994
Source: Exhibit C

FIRREA Allowable % of Tangible Assets 1.500% 1.500% 1.500% 1.000% 0.750% 0.375%

Total ButFor Contractual Goodwill 2.239% 2.194% 1.940% 1.750% 1.471% 0.826%

Total ButFor Supervisory Goodwill 3.739% 3.694% 2.950% 2.461% 1.995% 1.075%




As a result of this correction, butfor Sterling's core capital would have been 2.21 percent, below the 3 percent core capital required by the FIRREA capital regulation, effective with the filing of the March 31, 1990 Thrift Financial Report.
8



App. 813



9

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 40 of 42

18. Even if one were to assume that allocating qualifying supervisory goodwill proportionally between contractual and noncontractual supervisory goodwill would have been appropriate, which, based upon my experiences at the FHLBB and in the thrift industry, I do not believe to be a plausible assumption, Dr. Horvitz's model would still result in no lost profits damages. To illustrate, I have recalculated butfor Sterling's core capital under this proportional allocation methodology to determine the potential impact at December 31, 1989, as shown in Table 6.

Table 6. ButFor Sterling's Core Capital at December 31, 1989 (Proportional Allocation) Amount (000s) ($12,581) 15,557 4,398 10,500 (781) 4,273 $21,366

Tangible Capital at 12/31/89:





Adjustments Related to Unidentified Intangibles Qualifying supervisory goodwill (Lewis Federal and TriCities)9 Qualifying supervisory goodwill (Central Evergreen)10 Other Adjustments Capital raising Incremental butfor bank adjustments Identified intangibles11 ButFor Sterlings Core Capital


Qualifying supervisory goodwill under the FIRREA capital regulation totals $10,421. Sterling's total supervisory goodwill from both contractual and noncontractual acquisitions was $26,914 ($15,557 from Lewis Federal and TriCities and $11,357 from Central Evergreen). Allocating the $10,421 qualifying supervisory goodwill proportionally between contractual and noncontractual results in $6,023 (57.8 percent) allocated to the Lewis Federal and TriCities contractual supervisory goodwill. However, per the Court's opinions, butfor Sterling can include all of the Lewis Federal and TriCities supervisory goodwill in calculating regulatory capital. Consequently, $15,557 of the Lewis and TriCities contractual supervisory goodwill is considered to be qualifying supervisory goodwill for purposes of calculating core capital.
9

Allocating the $10,421 qualifying supervisory goodwill proportionally results in $4,398 (42.2 percent) allocated to the Central Evergreen noncontractual supervisory goodwill.
10 11

See Footnote 7.



App. 814



10

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 41 of 42

19. I have adjusted Dr. Horvitz's model to reflect this lower level of core capital at December 31, 1989. After making this one change, the amount of lost profits generated under Dr. Horvitz's model is negative (i.e., no lost profits).

V. Dr. James's Adoption of Dr. Horvitz's Erroneous ButFor Sterling Core Capital Calculations 20. DOJ has asked me to review certain portions of the James 2006 Report that purport to calculate lost profits during the period between the breach and the date of full mitigation (February 28, 1993). Dr. James effectively adopts Dr. Horvitz's lost profits model, thus Dr. James's analysis of lost profits in the James 2006 Report suffers from the same flaws that I have identified in the Horvitz 2006 Report. As I did with Dr. Horvitz's model, I have adjusted for the error in the calculation of qualifying supervisory goodwill and determined that Dr. James's model, like Dr. Horvitz's model, results in no lost profits.

21. I hold these opinions to a reasonable degree of certainty and plan to testify to these issues at trial. As noted above, my work is ongoing.

February 15, 2007 ________________________ W. Barefoot Bankhead

App. 815

11

Case 1:95-cv-00829-TCW

Document 222-26

Filed 02/26/2007

Page 42 of 42

CERTIFICATE OF FILING I hereby certify that on February 26, 2007, a copy of foregoing "APPENDIX TO DEFENDANT'S REVISED MOTION FOR SUMMARY JUDGMENT REGARDING DAMAGES - VOL. II" was filed electronically. I understand that notice of this filing will be sent to all parties by operation of the Court's electronic filing system. Parties may access this filing through the Court's system.

/s/ Elizabeth M. Hosford Elizabeth M. Hosford