Free Reply to Response to Motion - District Court of Federal Claims - federal

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 1 of 21

IN THE UNITED STATES COURT OF FEDERAL CLAIMS No. 05-231 T (Chief Judge Damich) JZ Buckingham Investments, LLC as Tax Matter Partner of JBJZ Partners, a South Carolina general partnership, Plaintiff, v. United States of America, Defendant. _________

REPLY DECLARATION OF A. LAWRENCE KOLBE ON ECONOMIC IMPLICATIONS OF THE DECLARATION OF DON M. CHANCE, PH.D., CFA AND OF THE PLAINTIFF'S RESPONSE TO GOVERNMENT'S MOTION TO PARTIALLY EXCLUDE EXPERT REPORT AND TESTIMONY OF DON M. CHANCE Prepared for the U.S. Department of Justice

The Brattle Group 44 Brattle Street Cambridge, MA 02138-3736 617.864.7900 voice 617.864.1576 fax May 16, 2008

5
gov.reply.app2

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 2 of 21

Table of Contents INTRODUCTION AND SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Purpose of Reply Declaration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Summary of Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 CHANCE DECLARATION'S COMMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 PLAINTIFF RESPONSE'S COMMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Claim of a "Flip-Flop" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Comments on the Seven Specific Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

gov.reply.app3

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 3 of 21

I.

INTRODUCTION AND SUMMARY

My name is A. Lawrence Kolbe. My business address is The Brattle Group, 44 Brattle Street, Cambridge, Massachusetts 02138. I hold a B.S. in from the U.S. Air Force Academy (1968) and a Ph.D. from the Massachusetts Institute of Technology (1979), both in economics. I am a financial economist with three decades of experience in the field. I am co-author of three books and author or co-author of a number of articles. I have testified on financial and regulatory issues in many forums in the U.S. and other countries, including U.S. District Courts and the U.S. Court of Federal Claims. The appendix to this reply declaration provides more detail on my professional

qualifications. My Expert Report in this matter was submitted on June 1, 2007 ("Kolbe Report"), and my Rebuttal Report on July 2, 2007 ("Kolbe Rebuttal"). My deposition in this matter was taken on August 23, 2007. I prepared a Declaration in this matter on the present topic that was filed on March 7, 2008 ("Kolbe Declaration"). A. Purpose of Reply Declaration

I have been provided with two documents, the "Declaration of Don M. Chance, Ph.D., CFA", dated April 21, 2008 ("Chance Declaration"), and the "Plaintiff's Response to the United States' Motion to Partially Exclude the Expert Report and Testimony of Don M. Chance,"dated April 21, 2008 ("Plaintiff Response"). The Chance Declaration purports to respond to statements in the Kolbe Declaration, and aspects of the Plaintiff Response offer additional responses to statements in the Kolbe Declaration. I have been asked by counsel to the Tax Division of the U.S. Department of Justice ("DOJ") to analyze whether anything in the Chance Declaration or Plaintiff Response alters my conclusion that the Chance Report's methodology of comparing profit probabilities is not an economically reliable methodology. The Chance Report compared the profit probabilities of the COBRA paired options to those of Google and the S&P 500 index to determine whether the COBRA options had a reasonable
gov.reply.app4

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 4 of 21

probability of profit. In the Kolbe Declaration, I concluded that such a methodology has no foundation in standard economic principles. I have now been asked to determine and to comment on as warranted whether any of the statements in the Chance Declaration and Plaintiff Response directly address this precise issue, and if so, what the economic merits of such statements are.1 B. Summary of Findings

Chance Declaration: Please recall that there are three parts to the Chance Report's profit probability comparison methodology: (1) calculate profit probabilities on a selection of publicly traded options, (2) calculate profit probabilities for the options at issue in this dispute, and (3) conclude that a probability of profit for an option in this dispute is reasonable if it lies within the range observed for publicly traded options.2 The first two parts jointly constitute a profit probability calculation step, and the last part a profit probability comparison step. The Chance Declaration misses the point of my declaration (and of my original rebuttal). The fundamental problem with the Chance Report's methodology is not how the profit probabilities that underlie its analysis were calculated (the probability calculation step), but to what use they were put (the probability comparison step). The probability comparison step is the part of the Chance Report's methodology without foundation in economic principle. The flaw arises when the Chance Report compares probabilities from options priced in the market to the probabilities of the digital options in the present dispute without consideration of whether those digital options were priced in

1

This narrow assignment requires that this reply declaration not address several aspects of both the Chance Declaration and the Plaintiff Response with which I disagree. Therefore, my silence here on any statement made in those documents should not be construed to mean that I accept or agree with it. The statements in this section are summaries of detailed discussions in the body of the reply declaration. It is prepared only for the convenience of the reader, and this section cannot and is not intended to replace the subsequent, more detailed discussions. For this reason, this summary does not contain references to documents or data sources. Footnotes that provide the sources of the information on which I rely appear in the body of the declaration.

2

2

gov.reply.app5

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 5 of 21

a way that was consistent with their probabilities. That step is the novel part of the Chance Report's methodology, the step unsupported by standard economic principles. It is without foundation in economic principle to assume that because publicly traded options have a particular range of profit probabilities, any profit probability within that range on some other option is per se "reasonable" regardless of how the price at which that other option is bought or sold relates to its market value. Nor does anything in the Chance Report or the Chance Declaration provide economic justification for the assumption that such a comparison can assess whether the profit probability is "reasonable." The Chance Declaration focuses instead on defending the step the Kolbe Declaration does not criticize, the calculation of the probabilities, not the comparison step. The Chance Declaration nowhere addresses the fundamental problem of the probability comparison step in Chance Report's methodology. Plaintiff Response: Parts of the Plaintiff Response reflect the mistaken focus in the Chance Declaration summarized above, and the above discussion applies to these statements, too. However, the Plaintiff Response goes beyond the Chance Declaration to make statements about my reports and/or declaration that are economically or factually incorrect. My reply declaration discusses six of them, one concerning a claimed inconsistency among my reports and five concerning comments on the seven specific factors I was asked by DOJ to consider in my original declaration. The Plaintiff Response claims that my view of the price charged for the digital options in this dispute "flip-flops" between the Kolbe Report and the Kolbe Declaration. The claimed reason is that (1) some of the discussion in the Kolbe Report explains that part of the price charged for the COBRA options reflects a fee to Deutsche Bank A.G. ("DB"), while (2) the Kolbe Declaration (and the Kolbe Rebuttal) focus on the problems for the Chance Report's methodology arising from the material excess of the prices charged for the COBRA options over their market values. But there

3

gov.reply.app6

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 6 of 21

is no "flip-flop," because DB's material fees, which are embedded in the prices paid for the options, are the reason that the prices paid for the options materially exceed their value. The first of the comments on the seven factors addresses Factor 1, whether the method is susceptible to testing. Recall that the Kolbe Declaration found that Chance Report's test effectively cannot be failed because financial markets are so large that some publicly traded option can be found with almost any profit probability. The Plaintiff Response states in reply that the Chance Report put some limits on the options it selected. There is no explanation for why the particular limits cited are necessary for identification of a sample of options for testing profit probability "reasonableness," nor is there an economic literature on this subject. Moreover, even accepting these limits as stated, they are not meaningful in this context because the part of the market that remains even after the cited limits are imposed is still so large. Second, regarding the potential error rate (Factor 3), the Plaintiff Response states that "[a] test is not rendered unreliable on the basis that it does not produce erroneous results." This is simply wrong, because a test can have false negatives -- here, failure to detect an unreasonable profit probability -- as well as false positives -- here, incorrectly identifying a reasonable profit probability as "unreasonable." The Chance Report's error rate for detection of "unreasonable" probabilities will effectively be 100 percent. Third, regarding standards for the method's operation (Factor 5), the Plaintiff Response claims that there are standards in economics that govern the Chance Report's methodology for assessing whether the probability of profit is "reasonable." However, there are no such standards, because nothing in economics justifies the probability comparison step.

4

gov.reply.app7

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 7 of 21

Fourth, the Plaintiff Response claims that the Chance Report's methodology is based on a reliable and widely accepted technique (Factor 6). This is incorrect, because there is no such widely accepted technique in economics for comparing the "reasonableness" of option profit probabilities. Finally, regarding the uses to which the method has been put (Factor 7), the Plaintiff Response claims that the Chance Report's methodology is premised on a model that was not developed solely for the purpose of testifying. This is incorrect, because there is no accepted economic model for comparing the "reasonableness" of option profit probabilities. The remainder of this reply declaration first reviews the Chance Declaration, then the Plaintiff Response. II. CHANCE DECLARATION'S COMMENTS

The Chance Declaration simply misses the point of my own declaration, so that much of its content is irrelevant. Please recall that there are three parts to the Chance Report's profit probability comparison methodology: (1) calculate profit probabilities on a selection of publicly traded options, (2) calculate profit probabilities for the digital options at issue in this dispute, and (3) conclude that a probability of profit for an option in this dispute is reasonable if it lies within the range observed for publicly traded options.3 The first two parts jointly constitute a profit probability calculation step, for the publicly traded and the COBRA options, respectively, and the last part a profit probability comparison step. The Chance Declaration's initial summary claims that4 The Government's essential point is that my methodology in determining the reasonableness of profit probabilities cannot be used for the digital options in this case because they did not trade in well-functioning capital markets with liquidity,
3

Kolbe Declaration, Section II, based on quotations from the Chance Report. The Chance Declaration does not dispute this description of the Chance Report's methodology. Chance Declaration, Section III.

4

5

gov.reply.app8

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 8 of 21

transparency, and other desirable factors that are the basis for most valuation models in finance. The Government's objection is based on their further allegation that the options did not trade at market values. I did not value all of the options (only the short options on the transfer dates), so it is not possible for me to affirm that these options traded in excess of market value. If we accept for the sake of argument that the Government's valuation of the digital options is correct, one would conclude that the investors paid more than market value. Their objection to my probability calculations is groundless, however, because the prices at which the options were actually traded were incorporated into my calculations. Recall that the digital options have only three outcomes. To find the probability of profit, it is necessary only to identify the outcomes that are profitable. For a call, profitability would be achieved if the low strike were hit because then the payoff would be in the sweet spot (clearly enough to be profitable). If the currency passes the high strike, profitability is still achieved though the profit is not as high. Calculation of the probability is merely an exercise in determining whether the exchange rate reaches the required level to be profitable. My calculations are therefore not invalidated even if the digital options did sell at a premium. In fact, the options could have even been purchased at higher prices, prices up to but just below the payoff from passing the higher strike without affecting the probability. And even if the Government were correct in challenging my methodology on this basis, the calculations of Drs. Kolbe and DeRosa would be invalidated as well because their calculations are based on the same model that I used. This response misses the point of my declaration (and of my original rebuttal report).5 The fundamental problem with the Chance Report's methodology is not how the profit probabilities that underlie its analysis were calculated, but why they were calculated, or more particularly, to what use they were put, which is the part of the Chance Report's methodology without foundation in economic principle.6 The flaw comes not in the step where the Chance Report computes the probabilities, it comes in the step in which it compares probabilities from options on Google and S&P 500 index options to the probabilities of the paired digital options in the present dispute without consideration of whether the COBRA options were priced in a way that was consistent with their

5

See the Kolbe Declaration, Sections III and IV, and the Kolbe Rebuttal, Section III. Indeed, the Kolbe Declaration assumes that the profit probability calculations themselves were perfect. (Kolbe Declaration, p. 5 and footnote 13.)

6

6

gov.reply.app9

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 9 of 21

probabilities.7 That step is the novel part of the Chance Report's methodology and the one without support in standard economic principles. It is fallacious and without foundation in economic principle to assume that because publicly traded options have a particular range of profit probabilities, any profit probability within that range on some other option is per se "reasonable" regardless of how the price at which that other option is bought or sold relates to its market value. This can be illustrated by using an example from the Chance Declaration. The example postulates a wager in which a roll of a die that comes up 1, 2 or 3 yields nothing while a roll of 4, 5 or 6 receives a payment of $10.8 It observes that the profit probability is the same regardless of how much you pay for the right to play this game, as long as you pay less than $10. This is perfectly true, but it begs the question of whether that probability is properly classified as "reasonable." The classification step is where the novel, and economically unsupportable, part of the Chance Report's methodology arises. That is, suppose you pay $9 for the right to pay this game, a possibility the Chance Declaration's example explicitly mentions. You have a 50-50 chance of winning $10, so your expected profit is [(0.5 × $10) ! $9] = ($5 ! $9) = !$4. There is nothing in economics that says that such a wager provides a "reasonable" probability of profit. In what sense could it be economically "reasonable" to overpay by 80 percent ($4/$5) of the fair price of a bet? To address fully the Chance Report's comparison methodology and why it is without foundation in economic principle, we need to extend the Chance Declaration's example to add the

7

Thus, the above quotation from the Chance Declaration is also incorrect to assert that the criticism in the Kolbe Declaration arises because the COBRA option positions traded over the counter, directly between a bank and its customers, instead of on a public exchange. To the contrary, over-the-counter trades are very common. The problem with the Chance Report's methodology is not with where the options traded, it is with the fact that they were materially overpriced relative to their profit probabilities, a factor that the Chance Report's profit probability comparison methodology ignores. Chance Declaration, Section IV, part 1.

8

7

gov.reply.app10

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 10 of 21

equivalent of the public option markets for Google stock and the S&P 500 index, on which the Chance Report relies. Suppose that the right to win $10 from the roll of a fair die were issued as a publicly traded security in financial markets. To consider the Chance Report's methodology, we need to imagine there are different such securities with different probabilities of winning. For example, suppose one type of bet, "Wager A," pays off if the die comes up with a 5 or 6, that "Wager B" pays off with a 4, 5 or 6 (as in the original example), and that "Wager C" pays off with a 3, 4, 5 or 6. The payoff probabilities of these three options are a, ½ and b, respectively. One would expect Wager A to trade for about ($10 × a) = $3.33, Wager B for about ($10 × ½) = $5, and Wager C for about ($10 × b) = $6.67.9 It would be unsurprising for a financial economist to accept these profit probabilities as "reasonable" at these prices, since each probability is consistent with the price paid and the potential payoff. Thus for each wager, the economic conclusion as to whether the probability of profit is "reasonable" is inextricably linked to the price paid for the wager. However, in this analogy, the Chance Report's methodology states that because the range of probabilities from the publicly traded wagers goes as low as a, the profit probability of any wager as low as a is reasonable no matter what the price paid for it, just as long as the price is less than the potential payoff. For example, if one paid $9 for a 50-50 chance at $10 (a wager that has a profit probability of ½), the Chance Report's methodology would characterize this wager as "reasonable," even though the price paid for the option exceeded an economically fair price by 80

9

The word "about" is required in case there is a delay in payment or some other factor that might cause investors to require a positive rate of return for undertaking the wager. Absent such factors, the prices should be close to those stated, because any investor would be able to lock in an effectively guaranteed return by placing enough such bets.

8

gov.reply.app11

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 11 of 21

percent.10 This probability comparison step is the novel part of the Chance Report's methodology, not its calculation of the probabilities, and the comparison step is without foundation in economic principle. On its face, it is plainly "unreasonable" and unsupported by standard economic principles to pay $9 for a 50-50 chance at winning $10, and the fact that other wagers exist for which even worse odds are reasonable (such as Wager A above), is simply irrelevant to that conclusion.11 In fact, the Chance Report's profitability comparison methodology would go further still. Imagine a fourth type of publicly traded wager existed, Wager D. Wager D would pay $10 only if a throw of two dice both came up 6. The odds of that outcome, which is the only way for the wager's owner to make a profit, are one in 36, or 2.8 percent. This percentage is higher than the probability found in some of the analyses of publicly traded options that underlie the Chance Report's methodology.12 Thus, under the Chance Report's methodology, a "reasonable probability of profit" would exist if someone paid $9 for a $10 payoff with a one-in-36 chance of getting the $10, which implies an expected profit of [(0.028 × $10) ! $9] = [$0.28 ! $9] = ! $8.72 on a $9 bet (which is an expected loss of 97 percent of the amount bet). Nor does the Chance Report or the Chance Declaration provide any economic justification for the assumption that such a comparison can assess whether the profit probability is "reasonable." To the contrary, as noted above, the Chance Declaration focuses instead on defending the step the

10

For clarity, note that the Chance Report does take account of the price paid in calculating the profit probability itself, as noted in the quotation from the Chance Declaration. However, it fails to consider whether the expected profit that flows from its calculated probability is even positive, let alone "reasonable." See the Kolbe Rebuttal at Section III and the Kolbe Declaration, footnote 3 and Section IV. The cited numerical example from the Chance Declaration further verifies this failing. No economically rational investor would accept such a bet, for example. In fact, even casinos and lottery tickets offer better gambles than this, yet such a tradeoff is cited by the Chance Declaration itself as consistent with the Chance Report's methodology. See the Kolbe Declaration, footnote 15.

11

12

9

gov.reply.app12

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 12 of 21

Kolbe Declaration does not criticize, the calculation of the probabilities, instead of the comparison step. The Chance Declaration nowhere addresses this fundamental problem with the probability comparison step in the Chance Report's methodology. III. PLAINTIFF RESPONSE'S COMMENTS

Parts of the Plaintiff Response reflect the failure of the Chance Declaration to address the probability comparison step, and the above discussion applies to these statements, too.13 However, the Plaintiff Response goes beyond the Chance Declaration to make other statements about my reports and/or declaration that are economically or factually incorrect. This section discusses six of them, one regarding a claimed inconsistency and the rest regarding the Plaintiff Response's comments on five of the seven specific factors I was asked to address in my original declaration: · A claim that my view of the price charged for the digital options in this dispute "flip-flops" between the Kolbe Report and the Kolbe Declaration; and Statements regarding five of the seven factors: (1) that the Kolbe Declaration was incorrect that the Chance Report's comparison methodology is essentially impossible to fail; (2) that a test that produces no erroneous results cannot be unreliable, (3) that there are standards in economics that govern the Chance Report's comparison methodology for assessing whether the probability of profit is "reasonable,"(4) that the comparison methodology is based on a reliable and widely accepted technique, and (5) that it is premised on a model that was not developed solely for the purpose of testifying. A. Claim of a "Flip-Flop"

·

The Plaintiff Response contains a section that claims that "the government contradicts itself on the treatment of the alleged excess amount paid for the digital options."14 This claim is wrong because it is based on a misunderstanding of the economics of price and value.

13

Again, this reply declaration's silence on a point made in the Plaintiff Response does not indicate that I agree with it. Plaintiff Response, Section III, part B, number 1; the quotation is from the title to the section.

14

10

gov.reply.app13

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 13 of 21

Specifically, when an investment bank trades an over-the-counter option with an investor, two things happen. A sum of money is transferred from one of the parties to the other, and an option is given in exchange. The bank's fee for this service is the difference in value between the two: Bank Fee = Value of what the bank gets ! Value of what the bank gives. (1)

If the bank buys an option written by the investor (so the investor is short the option), Bank Fee = Value of the short option ! Price paid by the bank. (1a)

If the bank writes and sells an option to the investor (so the investor is long the option), Bank Fee = Price paid to the bank ! Value of the long option. (1b)

For a bank to earn a positive fee when buying an option written by the investor (Equation (1a)), the value of the option the bank receives must exceed the price the bank pays for it. For a bank to earn a positive fee when writing and selling an option to the investor (Equation (1b)), the price paid by the investor to the bank must exceed the value of the option the bank gives in exchange. In short, the price of an over-the-counter option routinely includes a fee to the bank involved, fundamentally based on the fact that an investor buys from the bank at an "ask" price and sells to the bank at a "bid" price. Therefore, there is no inconsistency between the Kolbe Report and the Kolbe Declaration in this regard. No "flip-flop" is involved when (1) the Kolbe Report finds that part of the price paid for the option positions is a material fee to DB and (2) the Kolbe Rebuttal15 and the Kolbe Declaration16 observe that the prices paid for the option positions materially exceeded their values, creating a mismatch between the price paid and the probability of earning a profit. Both statements are true, because DB's material fees are the reason that the prices paid for the options materially exceed their values.

15

Kolbe Rebuttal, Section III. Kolbe Declaration, Sections III and IV.

16

11

gov.reply.app14

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 14 of 21

B.

Comments on the Seven Specific Factors

The Plaintiff Response also addresses the findings of the Kolbe Declaration regarding five of the seven specific factors identified in the DOJ letter attached to that declaration. First, regarding Factor 1, whether the method can be tested, the Kolbe Declaration concluded that the Chance Report's method effectively cannot be failed because financial markets are immense and options can be found with almost any probability of profit.17 The Plaintiff Response states that the Chance Report put some limits on the options it selected, which would restrict the test from considering some traded options.18 There is no explanation in Dr. Chance's reports or the Plaintiff Response as to why these particular limits are necessary for selecting a sample of options for testing profit probability "reasonableness."19 Moreover, even if these particular limits are accepted as stated, they are not meaningful in this context because the part of the market that remains even after the stated limits are imposed is still so large.20 A sample of options that conformed to the limits

17

Kolbe Declaration, Section V, cited in the Plaintiff Response in Section III, part B, number 3. Plaintiff Response, Section III, part B, number 3. The statement is that the Chance Report selects options on "well recognized" securities that trade at least 100 contracts per day and that have similar times to expiration. Nor do any of these documents cite a body of economic literature in support of these particular limits in profitability "reasonableness" tests, because no such literature exists. Recall that Dr. Chance himself could not reject the possibility that a profit probability as low as 0.2 percent might be found, for example. (See Dr. Chance's September 6, 2007 deposition, pp. 243-45.) More generally, options abound that satisfy the stated criteria. For example, in July 2006, the month used in the Chance Report, 59 companies had average daily volumes of at least 5,000 equity option contracts on the Chicago Board Option Exchange; 25 companies had average daily volumes of at least 10,000 contracts, and 5 of at least 20,000 contracts. (These data are from the Chicago Board Option Exchange website: http://www.cboe.com/data/AvgDailyVolArchive2006.aspx.) With total volumes this high, many individual option contracts will hit the 100/day standard used in the Chance Report. In fact, Google, the stock selected in the Chance Report, was only third on the list, well behind Apple Computer and the Altria Group. Additionally, even for the specific options selected in the Chance Report, an analyst could always calculate their probabilities on multiple days.

18

19

20

12

gov.reply.app15

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 15 of 21

cited in the Plaintiff Response would always exist that still provided a test that was essentially impossible to fail.21 Second, the Plaintiff Response argues against the Kolbe Declaration's findings on the third factor, the potential error rate. (Recall that the Kolbe Declaration finds this factor relevant because the test will effectively never find an option to have an unreasonable probability of profit.) In reply to this point, the Plaintiff Response states that, "[a] test is not rendered unreliable on the basis that it does not produce erroneous results."22 That statement is simply wrong. A test may have two types of errors, finding something to be true when it is actually false and finding something to be false when it is actually true. It is correct that the Chance Report's comparison test will essentially never erroneously find an option's profit profitability to be unreasonable when it is actually reasonable, because the methodology will essentially never find any profit probability to be unreasonable. But this means it will essentially always erroneously find profit probabilities to be reasonable when they are actually unreasonable, unless the definition of "unreasonable" is such that there actually are no "unreasonable" profit probabilities (in which case, why waste resources conducting a test at all?). If "unreasonable" profit probabilities exist, the Chance Report's error rate in detecting them will effectively be 100 percent.23 Third, the Plaintiff Response says that the Kolbe Declaration is incorrect to say that Factor 5, whether standards exist for the methodology, is not relevant because the Chance Report's

21

This part of the Plaintiff Response and the next (part B, number 4) both also claim that the government's real quarrel is with the results of the methodology, not the methodology itself However, the conclusions of the test are effectively predetermined by the nature of the test, so the test itself is the problem. Plaintiff Response, Section III, part B, number 4. This point was noted in the Kolbe Declaration in Section V, but it is not addressed in the Plaintiff Response.

22

23

13

gov.reply.app16

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 16 of 21

calculations of the probabilities of profit use well-recognized techniques such as the Black-Scholes model.24 However, as discussed above, the fundamental problem with the Chance Report's methodology is not with the step in which it calculates the probabilities, it is with the probability comparison step, and in that area Factor 5 is irrelevant. There is no well-established standard in economics that would govern whether an option should be classified as having a "reasonable probability of profit." That question has not been generally addressed in economics, but rather has arisen only in the present circumstances. Fourth, the Plaintiff Response says that the Kolbe Declaration's comments on the sixth factor, how the method relates to other methods, are incorrect because the Chance Report's method relies on the Black-Scholes model.25 Again, the fundamental problem with the Chance Report's method is not with the step in which the profit probabilities are calculated, which is the part that relies on the techniques cited in this part of the Plaintiff Response. Instead, it is with the step that identifies the profit probabilities of options that trade at prices far above their market value as "reasonable" simply because the probabilities fall within the wide range observed in the market. This second step has all the problems discussed above. The Plaintiff Response mentions this step only briefly, as "a simple comparison," without any attempt to justify the economically unwarranted assumption that such a comparison is meaningful.26 Finally, regarding Factor 7 (the uses to which the method has been put), the Plaintiff Response states that the Chance Report's method has not been developed for purposes of testifying

24

Plaintiff Response, Section III, part B, number 5. Plaintiff Response, Section III, part B, number 6. Plaintiff Response, Section III, part B, number 6.

25

26

14

gov.reply.app17

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 17 of 21

gov.reply.app18

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 18 of 21

Appendix: QUALIFICATIONS OF A. LAWRENCE KOLBE EDUCATION B.S. in International Affairs (Economics), U.S. Air Force Academy, 1968. Ph.D. in Economics, Massachusetts Institute of Technology, 1979.

EMPLOYMENT U.S. Air Force, 1963-1977 Charles River Associates, 1978-1985 Putnam, Hayes and Bartlett, 1985-1990 The Brattle Group, 1990-

HONORS AND AWARDS Sears Foundation National Merit Scholarship, 1963 (declined). Fairchild Award, U.S. Air Force Academy, 1968 (for standing first in his class, academically). National Science Foundation Graduate Fellowship in economics, MIT, 1968-1971. Joint Service Commendation Medal, 1975.

PROFESSIONAL AFFILIATIONS American Economic Association American Finance Association The Econometric Society Served as Referee for The Rand Journal of Economics, Land Economics, The Journal of Industrial Economics

PAPERS AND PUBLICATIONS IN THE LAST TEN YEARS "How to Value a Lost Opportunity: Defining and Measuring Damages from Market Foreclosure," (with William B. Tye and Stephen H. Kalos), The Economics of Antitrust Injury and Firm-Specific Damages, Kevin S. Marshall, ed., Tucson, Arizona: Lawyers & Judges Publishing Company, Inc. (2008). (Republication of a 1995 article.) "Measuring Return on Equity Correctly: Why current estimation models set allowed ROE too low," Public Utilities Fortnightly (with Michael J. Vilbert and Bente Villadsen), August 2005. "The Effect of Debt on the Cost of Equity in a Regulatory Setting," (with Michael J. Vilbert and Bente Villadsen, and with "The Brattle Group" listed as author), published by the Edison Electric Institute (dated January 2005, issued April 2005)

gov.reply.app19

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 19 of 21

Capital Investment and Valuation, (with Richard A. Brealey and Stewart C. Myers, with "The Brattle Group" listed as third author), New York: McGraw-Hill/Irwin (2003). "The True Hourly Rate for Private Counsel in the State of Louisiana Tobacco Lawsuit," (with August J. Baker and Bin Zhou), Brattle report prepared for private counsel to the Louisiana Attorney General in the state's lawsuit to recover health care costs from the tobacco industry (July 2000). "The Cost of Capital for the Dampier to Bunbury Natural Gas Pipeline," (with M. Alexis Maniatis and Boaz Moselle) Brattle report submitted to the Office of Gas Access Regulation, Western Australia (October 1999). "Compensation for Asymmetric Risks," (with others) Brattle report prepared for GPU PowerNet, Melbourne, Australia (October 1999). "A Non-Practitioner's Guide to the State of the Art in Cost of Capital Estimation," (with others) Brattle report prepared for GPU PowerNet, Melbourne, Australia (June 1999). "A Note on the Pre-tax Weighted Average Cost of Capital in a Regulatory Context with Australian Dividend Tax Credits and Alternative Debt Refinancing Policies" (with M. Alexis Maniatis), Working Paper in Progress. "The Impact of Stranded-Cost Risk on Required Rates of Return for Electric Utilities: Theory and An Example" (with Lynda S. Borucki). Journal of Regulatory Economics Vol. 13 (1998), 255-275.

TRIAL TESTIMONY AND DEPOSITIONS IN LAST FOUR YEARS 2004 Case: Party: Forum: Case: Party: Forum: Case: Party: Forum: Case: Party: Forum: Coltec Industries, Inc. v. The United States of America, Fed. Cl. No. 01-72T The United States of America The United States Court of Federal Claims Liberty Digital, Inc. v. AT&T Broadband, LLC and Comcast Corporation, Civil Action No. 03-CV-95 AT&T Broadband, LLC and Comcast Corporation Colorado District Court, County of Arapahoe, 18th Judicial District West Virginia-American Water Company, Case No. 04-0373-W-42T West Virginia-American Water Company Public Service Commission of West Virginia Koch Industries, Inc., and Subsidiaries v. The United States, Fed. Cl. No. 028T The United States The United States Court of Federal Claims A-2

gov.reply.app20

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 20 of 21

Case:

Party: Forum:

United States of America v. Anne (Sandy) Batchelor-Robjohns, a/k/a/ Anne O'Neill Batchelor; Daniel J. Ferraresi; Marvin C. Gutter; and Father Patrick O'Neill, Co-Personal Representatives of the Estate of George E. Batchelor, Case No. 03-20164-CIV-UngaroBenages/Brown United States of America United States District Court for the Southern District of Florida, Miami Division

2005 Case: Party: Forum: Case: Party: Forum: Case: TransCanada PipeLines Limited 2004 Mainline Tolls and Tariff (Phase 2), RH-2-2004 TransCanada PipeLines Limited [Canadian] National Energy Board Polymer Dynamics, Inc. v. Bayer Corporation Polymer Dynamics, Inc. U.S. District Court for the Eastern District of Pennsylvania Jade Trading, LLC, by and through Robert W. Ervin and Laura Kavanaugh Ervin on Behalf of Ervin Capital, LLC, Partners Other than the Tax Matters Partner, v. The United States, Case No. 03-2164T The United States The United States Court of Federal Claims

Party: Forum: 2006 Case: Party: Forum: Case:

Arizona-American Water Company, Inc., for Utility Service by its Paradise Valley Water District, Docket No. WS-01303A-05-0405 Arizona-American Water Company, Inc. Arizona Corporation Commission K2 Trading Ventures, LLC, New Vista, LLC, Tax Matters Partner, v. The United States of America, by and through the Internal Revenue Service, Case No. 04-1419T The United States of America The United States Court of Federal Claims Garco Investments LLP, et al. v. Sprint Corporation, Case No. 04 CV 01714 Garco Investments LLP, et al. District Court of Johnson County, Kansas, Civil Court Department

Party: Forum: Case: Party: Forum:

A-3

gov.reply.app21

Case 1:05-cv-00231-EJD

Document 173-3

Filed 05/16/2008

Page 21 of 21

2007 Case: JZ Buckingham Investments, LLC as Tax Matters Partner of JBJZ Partners, a South Carolina general partnership, v. United States of America, Case No. 05-231 T United States of America The United States Court of Federal Claims Gary Woods, as Tax Matters Partner of Tesoro Drive Partners, a Texas general partnership, and Gary Woods, as Tax Matters Partner of SA Tesoro Investment Partners, a Texas general partnership, v. United States of America, Case Nos. SA-05-CA-0216-XR. United States of America United States District Court for the Western District of Texas, San Antonio Division MURFAM FARMS, LLC, By and Through Wendell H. Murphy Jr., a Partner Other Than the Tax Matters Partner; PSM FARMS, LLC, By and Through Stratton K. Murphy, a Partner Other Than the Tax Matters Partner; MURPHY PORK PARTNERS, LLC, By and Through Wendell H. Murphy, Jr., a Partner Other Than the Tax Matters Partner, v. United States of America, Case Nos. 06-245 T, 06-246 T, and 06-247 T (Consolidated). United States of America The United States Court of Federal Claims Fidelity International Currency Advisor A Fund, L.L.C., by the Tax Matters Partner, v. United States of America, Case No. 05-40151-FDS. United States of America United States District Court for the District of Massachusetts

Party: Forum: Case:

Party: Forum:

Case:

Party: Forum: Case: Party: Forum: 2008 Case: Party: Forum:

Northwestern Energy, Docket No. 2007.7.82 Northwestern Energy Department of Public Service Regulation, Montana Public Service Commission

A-4

gov.reply.app22