Opinion
Case No. 98-13205
March 25, 2002
William A. Favreau, Esq., Michael D. Assaf, Esq, O'Connell Aronowitz, Plattsburgh, NY, for Debtor.
Kevin Purcell, Esq., Assistant U.S. Trustee, Albany, NY, United States Trustee.
Adam H. Friedman, Esq., Traub, Bonaquist Fox, LLP, New York, NY, for Unsecured Creditors' Committee.
E. Lisa Tang, Esq. Thuillez, Ford, Gold JOhnson, LLP, Albany, NY, for Barrie Guibord.
Robert J. Rock, Esq., Albany, NY, Attorney for Peter Guibord.
MEMORANDUM-DECISION AND ORDER
The matter before the court is the valuation of the Debtor's business enterprise in the context of a proposed modified plan's confirmation. Because it involves a core proceeding under 28 U.S.C. § 157(b)(2)(L), the court has jurisdiction pursuant to 28 U.S.C. § 157(a), 157(b)(1) and 1334(b).
Facts
On May 7, 1998, the Debtor filed a Chapter 11 petition. After several requests for an extension of time, it filed its initial plan on February 8, 1999. On the same date, it also filed an application to accelerate a hearing on its disclosure statement and plan. The court set a hearing on that application for February 25, 1999; it was adjourned to April 9, 1999 and then rescheduled to June 4, 1999 at which time the court confirmed the plan after resolving the objections of a few parties in interest. Consistent with its local practice, the court also issued a show cause order which set a hearing for December 16, 1999 if the Debtor failed to file a report of substantial consummation pursuant to Local Rule 3022-1 and an application for final decree pursuant to F.R.Bankr.P. 3022.
According to the Debtor, it obtained debtor-in-possession financing with First International Bank ("FIB") in the original principal amount of $2,800,000 ("loan"), with the intent that loan proceeds would be used to fund the first plan payments and ongoing operations would fund the balance of claims. When the Debtor failed to file the report of substantial consummation, the December 16, 1999 hearing was triggered. The events that took place on December 16th led the court to conclude that hearing and issue another show cause order which scheduled a hearing on March 2, 2000 for conversion or dismissal of the case based on the Debtor's failure to attain substantial consummation. The Office of the United States Trustee ("Trustee") filed a statement in support of dismissal or conversion.
The court conducted phone conferences regarding the conversion or dismissal of the case on March 2, 2000, March 9, 2000 and March 16, 2000. On April 3, 2000, the Internal Revenue Service ("IRS") filed a statement in support of dismissal or conversion, stating the Debtor had been notified of post-petition accruals totaling $500,664.03 and had not made any attempt to pay them.
On April 6, 2000, the parties agreed to an order appointing Trimingham Advisors, Inc. ("Trimingham") as responsible officer of the Debtor and delegating certain responsibilities to it.
That order covered a 30 day period. At the end of that period, another consent order employing Trimingham as responsible officer was entered effective through July 7, 2000. Numerous consensual extensions followed and, eventually, the Debtor filed an application for an order to show cause to appoint Trimingham as responsible officer of the Debtor. On May 29, 2001, the court ordered the continued appointment of Trimingham as responsible officer.
Following Trimingham's appointment, the Debtor received several offers to purchase property. Olsen Technologies made three offers; its highest offer, $4.2 million, sought to obtain the Debtor's intellectual property and some of its other assets and equaled, dollar for dollar, the Debtor's accounts receivable. Detroit Stoker made a purchase offer of $3.2 million for intellectual property, inventory and some equipment. CoBe Capital was the third offeror; it sought the Debtor's balance sheet and promised to fund a modified plan, including a $250,000 cash infusion. Although Trimingham supported the CoBe Capital offer, none of the offers ever went to closing.
The information contained in this paragraph is based largely on trial testimony at Tr. 3/13/01 pp. 29-40.
The transcriptionist inaccurately spelled this entity's name as "Coby Capital." E.g., Tr. 3/13/01 p. 37.
On August 31, 2000, the Committee of Unsecured Creditors ("Committee") filed a motion to compel the production of documents and an examination of the Debtor's principals and other insiders pursuant to F.R.Bankr.P. 2004; it later filed an adversary proceeding against them and entities related to them seeking recovery of estate property. On December 14, 2000, the Debtor filed a post-confirmation disclosure statement and modified Chapter 11 plan, together with an application to shorten time for a hearing on both. One of the modified plan's more controversial provisions is the treatment of the Debtor's shareholders. Section 4.9 of the modified plan provides for no distribution to equity holders and the cancellation of all shares. Section 5.1(b) provides for the issuance of new common stock and, as best as the court can determine at this time, if a shareholder is also an insider creditor, he or she will receive a payment of 15% of Class B shares. The Class B shares are characterized as "Participating/Non-voting" shares in Section 5.1(b) of the modified plan.
Not surprisingly, the shareholders objected to the modified plan as did a number of other parties in interest. After considering the objections and the Debtor's responses to them, the court issued a scheduling order which set an evidentiary hearing to specifically address the initial issue of the value of the Debtor's business enterprise. On consent of the parties, specifically the Debtor and two of its shareholders, Barrie Guibord and Peter Guibord, the court set an initial, expedited trial date of February 12, 2001. The trial date of February 12th became the 16th and ultimately became March 13th due to matters involving discovery and the expert witnesses.
On March 8th, the last day to file an objection to a pretrial statement and five days before the Tuesday trial date, the Debtor filed an objection to the Guibords' pre-trial statement, seeking to preclude the valuation report prepared by ACR International, Inc. ("ACR") and the testimony of Rod Garrett ("Garrett"), the principal of ACR. The Debtor argued the ACR report and Garrett's proposed testimony completely failed to pass the standards for expert testimony set forth in Daubert v. Merrill Dow Pharmaceuticals, Inc. 509 U.S. 579 (1993) ("Daubert") and Kumho Tire Company, Ltd. v. Carmichael, 526 U.S. 137 (1999) ("Kumho").
On the first day of trial, the court informed the parties that it initially viewed the Debtor's objection to the admission of the ACR report and Garrett's testimony as a matter more appropriately raised during trial as an objection to the credibility, methodology, etc., of the report and/or expert witness. (Tr. 3/13/01 pp. 8-9.) After stating its belief that the parties were well-versed and quite capable of arguing whatever appropriate objections might exist regarding the credibility of each expert witness and the methodologies used by them, the court ruled that the Debtor's Daubert objection would be heard in conjunction with Garrett's actual testimony, in effect, leaving the issue of the admissibility of the report and the attendant testimony as a matter for the parties to brief post-trial. (Tr. 3/13/01 p. 9.) Debtor's counsel renewed his Daubert objection when Garrett took the stand to testify and when the ACR report was offered into evidence. (Tr. 3/26/01 p. 117.) He withdrew his objection to the admission of the ACR report, however, noting for the record, "Your Honor, subject to my Daubert objection, we'll allow it in." (Tr. 3/26/01 p. 117.) The court received the document into evidence in that same manner. (Tr. 3/26/01 p. 117.)
The Debtor retained Collar City Auctions and Realty ("Collar City") to perform valuations of its real and personal property. The Collar City report contains the following values:
$120,000 is Fair Market Value
The Collar City Report notes that under a Chapter 7 liquidation, "auction values" would prevail and would be subject to a 10% commission payable to the auctioneer. The Guibords did not submit a competing expert report on real or personal property valuations.
The Debtor also had Marin Environmental, Inc. ("Marin") prepare a report; it contains an estimate of the overall cost to remedy certain existing environmental problems to a condition acceptable for development if it ever ceased operations as a foundry and machine shop. Marin's estimate for that remediation ranges from $350,000 to $600,000. Although its expert testified that the environmental work necessary for the Debtor to continue to operate as a foundry would be less expensive, he also admitted that the New York State Department of Environmental Conservation ("DEC") had not indicated any remediation was necessary. (Tr. 3/13/01 p. 123.)
The valuations contained in both the Urbach, Kahn and Werlin ("UKW") report and the ACR report did not take into account the cost of environmental remediation. The Guibords did not submit a competing expert report covering the cost of any environmental remediation although Garrett admitted at trial that his value would be reduced "dollar for dollar" by the cost of environmental remediation. (Tr. 3/27/01 p. 149.)
During trial the court also heard from numerous Trimingham employees and various appraisal experts. One of the key people at Trimingham who testified was Donald F. Fleischer. He provided relevant information about the Debtor's business, including the following: a gross profit margin ("GPM") of 40-41% for the year 1998, a GPM of 38% for the year 1999, a GPM of 41.9% for the year 2000, projected sales of $5,596,800 for the year 2000, the same amount of projected sales for the year 2001, $817,700 in net income (before reorganization expenses and income taxes) for the year 2000 according to a draft audit statement, reorganization expenses of $878,573 for the year 2000 and a reserve of $1.3 million over the next three years for capital expenditures. (Tr. 3/13/01 pp. 47-49, 68, 104-107, 109-110.) Regarding the capital reserve, on recross, Fleischer testified that capital expenditures would increase the Debtor's efficiency as a manufacturer, which, in turn, would improve its profit margin. (Tr. 3/13/01 p. 112.) On further redirect, when Debtor's counsel asked, "Is the projection with respect to that — is the favorability that increased efficiency would have as a result of the cap ex built into management's current projections with respect to gross profit?" Fleischer answered, "No, it's not." (Tr. 3/13/01 p. 114.) According to his earlier testimony, the GPM management projected for 2001 was 44%. (Tr. 3/1301 p. 15.)
Early on, it became apparent that the focus of the trial, the "battle of the experts," would center on the reports of Garrett and of David Evans ("Evans"), the Debtor's expert and a partner at UKW. Pursuant to the scheduling order and unlike the ACR report, the UKW report was received into evidence prior to trial because the court did not receive any objection to its admission. It presented an "indicated" value of $0 for the Debtor's business enterprise and a "normalized" value of $3,000,000. It was the second scenario, the normalized value, that Evans was examined and cross-examined on most extensively.
Using his own definition of "normalization," Evans used data compiled by Risk Management Associates ("RMA") to normalize the Debtor's actual performance numbers. (Tr. 3/13/01 pp. 155, 189.) The data he used was based on companies with a similar, but not identical, Standard Industrial Classification code ("SIC code") as the Debtor's, companies in the valve industry comparable to the Debtor in sales volume. (Tr. 3/13/01 pp. 157, 159-160; UKW report p. 14.) The specific SIC code he used was 3494, the code for manufacturing — valves and pipe fittings; its description included metal valves and pipefittings not elsewhere described (in the other codes), such as plumbing, heating valves and flanges. (Tr. 3/13/01 p. 158; Tr. 3/14/01 p. 13.) Evans stated on cross examination that SIC code 3491 was manufacturing — industrial valves and that its description does not include fluid power valves, plumbing fixture fittings and trim or plumbing and heating valves. (Tr. 3/14/01 p. 14.) During cross examination by Lisa Tang ("Tang"), attorney for Barrie Guibord, redirect examination and examination by the court, Evans testified that he used companies under SIC code 3494 as comparables instead of the closer match of SIC code 3491 because he used "sales" as his criteria for comparison and while there was data for 18 companies under SIC code 3494, there was no data under SIC code 3491 for companies with sales between $5 million and $10 million. (Tr. 3/14/01 pp. 31, 49-50, 54-55.) He further explained that he always ignored the data available for comparable "asset" companies because he believed comparing a company's volume of sales involves comparing reliable, fixed numbers, not a "manufactured number" like assets. (Tr. 3/14/01 p. 56.) He also testified that if it had been available, he would have used the empirical data reported under SIC code 3491 for sales between $5 and $10 million. (Tr. 3/14/01 p. 56.)
Although never directly asked during trial and never offering his own explanation as to why he chose his particular methodology, Evans used an apple tree analogy to describe it. His "capitalization of normalized earnings" method, using that analogy, involved counting the number of apples on the tree and then valuing them by applying what he called a capitalization rate. (Tr. 3/13/01 p. 163; Tr. 3/14/01 p. 20.) His "normalization" method involved two sets of numbers. The first set was the Debtor's actual numbers for the years 1998, 1999 and eleven months of 2000. Using those numbers, he projected the Debtor's income for 2001. (Tr. 3/13/01 pp. 154-155.) He took that historical data and the Debtor's balance sheet for those years and converted the raw numbers to percentages. Using the RMA compilation for the year 2000, Evans obtained a debt ratio of about 20% for the comparable companies he chose. (Tr. 3/13/01 p. 160.) He calculated percentages for the companies he believed were comparable and then applied those percentages to the Debtor's debt numbers, thereby attaining "debt normalization." (Tr. 3/13/01 pp. 158-59; UKW report Ex. D.1.)
Evans actually used three sets of numbers. The third set, the Debtor's actual numbers without capital expenditures, was not considered by the court in this decision.
Using the Debtor's actual sales numbers but normalized values for cost of goods sold, operating expenses and miscellaneous expenses, Evans obtained a number reflecting normalized net profit before taxes for each of the four years. He then added the Debtor's actual depreciation and amortization, normalized interest (by calculating total normalized debt as 20% of the assets shown in the Debtor's unaudited balance sheet for December 31, 2000 and multiplying that number by 11.5%, the Debtor's actual interest rate) and a normalized change in working capital, obtaining a number he called "pre tax cash flow to invested capital — normalized" for each year. (UKW report Ex. D.1.) Using those numbers, he subtracted taxes at 40% to obtain an annual "after tax normalized cash flow" that averaged $462,185 over the four years. (Tr. 3/13/01 p. 162; UKW report Ex. E.1.) That was how he determined income "on a normalized basis." (Tr. 3/13/01 p. 162.)
Evans testified that his normalized operating expenses did not include any Chapter 11 professional fees. (Tr. 3/14/01 pp. 34-35.)
Regarding his capitalization rate, Evans began by computing a cost of capital, starting with the riskless 20-year Treasury bond rate (at 5.6% as of December 31, 2000) and then adding three additional premiums. (Tr. 3/13/01 p. 163.) The three additional premiums included the common stock premium (at 7.8% according to Ibbotson Associates' 2000 compilation entitled "Stocks, Bonds, Bills and Inflation — 2000 Yearbook — Market Results for 1926-1999" ("Ibbotson's")), the small stock premium (at 4.3% according to Ibbotson's), and a company specific premium (at 4.0% based on Evans's judgment). (Tr. 3/13/01 pp. 164-166; UKW report pp. 18-19 and Ex. F.)
Using the Debtor's actual, historical cost of borrowing money at 11.5% (prime plus 2%) and subtracting 4.6% as the tax benefit (i.e., 40% of 11.5%), Evans calculated a weighted average cost of capital ("WACC") of 18.4%. In doing this calculation, he used a 22.5% to 77.5% debt to equity ratio, although neither side explored his reason(s) for using that ratio. His WACC calculation follows: .225 x (11.5% — 4.6%) + .775 x (5.6% + 7.8% + 4.3% + 4%) = 18.4%. (Tr. 3/13/01 p. 168; UKW report Ex. F.1.) He arrived at his 15.4% capitalization rate by subtracting a long-term growth rate of 3% from the 18.4% WACC. (Tr. 3/13/01 pp. 170-171; UKW report Ex. F.) Dividing the after tax normalized cash flow figure of $462,185 by the 15.4% capitalization rate, Evans arrived at his income approach valuation of $3,007,059. (UKW report Ex. F.1.)
Unless that ratio is the same "debt ratio of about 20%" Evans obtained from RMA compilation of comparable companies.
Of course, if those actual numbers were used the value would have been $3,001,201. Although he "rounded up" to arrive at the capitalization rate of 15.4%, he used 15.37% when performing the final part of his calculation, resulting in the slightly higher value.
The direct examination of Garrett did not provide the court with the same step by step analysis Debtor's attorney achieved by his direct examination of Evans. Garrrett testified that his primary income valuation method was called "discounted cash flow" ("DCF"). After reviewing all of the available preliminary reports Trimingham provided for the year 2000, Garrett averaged May through December's revenue and then annualized it, arriving at an amount of $6,513,500. (Tr. 3/27/01 pp. 160, 194.) Garrett further testified that he had also reviewed the Debtor's financial statements for the years 1996 through 1999, but did not use or rely on the numbers contained in them because the Debtor's history was so "disrupted," rendering, in effect, much of the historical information the Debtor's management provided him with "irrelevant" because of all of the "accounting issues" that would have had to have been resolved. (Tr. 3/26/01 pp. 160-161)
He also used a capitalization of earnings method.
Garrett projected approximately the same $6.5 million in revenue for 2001 with a 4% growth rate over the next five years. (ACR report p. 10.) Once again using Trimingham's May through December 2000 numbers where available, he calculated cost of sales, gross profit, general expenses, administrative expenses and selling expenses and then determined operating income for the year 2000 on an annualized basis. (Tr. 3/26/01 p. 170; ACR report pp. 10-11.) After calculating percentages for those figures, similar to how he calculated revenue figures, he was also able to calculate operating income for the years 2001 through 2005. (ACR report pp. 10-11.)
To complete the cash flow analysis (as shown in Table 2-C of the ACR report), Garrett took operating income for each year and added back "cash items," i.e., accounting items like depreciation and amortization. (Tr. 3/26/01 p. 137.) After subtracting interest at 11.5% and income taxes at 42%, he arrived at "cash available to pay debt" figures for each of the five projected years. (ACR report p. 14.) Garrett testified these figures represented the "free cash flow" available to the Debtor. (Tr. 3/26/01 p. 174.)
Garrett testified that another key element of the discounted cash flow valuation method is the WACC. (Tr. 3/26/01 p. 175.) He began his cost of debt calculation at 11.5%, the historical cost of the FIB loan, and subtracted a 42% tax rate, arriving at 6.67% for cost of debt. (Tr. 3/26/01 p. 176; ACR report pp. 16-17.) As for his calculation of the Debtor's cost of equity, it was during cross examination that Garrett revealed he had used Ibbotson Associates' Cost of Capital Quarterly, December 2000, specifically, the data reported for the three smallest of the ten companies with SIC code 349 and $6 million average revenue. (Tr. 3/27/01 pp. 61, 65; ACR report pp. 16-17.) It was also during that particular part of his cross examination that the court learned of his lack of knowledge regarding the three companies whose data he used, specifically, what valves or products they made.
After averaging the four calculations Ibbotson's had performed for the three companies, Garrett arrived at a cost of equity of 14.7%; he then testified, again on cross, that he added another 10% as an "additional risk premium." (Tr. 3/27/01 p. 70; ACR report p. 17.) In obtaining his debt to equity ratio of 2.7 (i.e., 72.9% debt to 27.1% equity), although his report states he used RMA data for the year 1999-2000, Garrett admitted he actually used RMA data for 1998-1999 for companies that qualified under SIC code 3491 with sales between $5 and $10 million. (Tr. 3/26/01 pp. 179-180; ACR report p. 17.) Applying the 2.7 ratio, he calculated a WACC of 11.55%. (ACR report p. 17.)
Once he had the cash flow figures and the WACC, he discounted the cash flow figures and obtained their present value. (Tr. 3/26/01 p. 182; ACR report p. 19.) In this discounting calculation, he used what he called a "standard mathematical formula" involving the WACC to arrive at a net present value ("NPV") factor for the years 2001-2005. (Tr. 3/26/01 p. 183.) Multiplying the NPV factor for each year by the cash flow figure for that year enabled him to arrive at present value figures for the years 2001 through 2005.
Regarding the final component of his DCF methodology, calculating terminal value, Garrett testified that he began the calculation by dividing an amount equal to the net operating profit after tax ("NOPAT") for the year 2005 by 9.55%, resulting in an amount of $7,625,500. (Tr. 3/26/01 pp. 180-181; ACR report p. 19.) He obtained the NOPAT figure ($728,600) by adding the total interest expense he calculated for 2005 (i.e., $123,500) to the net income he calculated for that year (i.e., $605,100). The 9.55% figure resulted when he subtracted a 2% growth factor from the WACC. According to Garrett's testimony, $7.6 million represented the Debtor's worth in the year 2005, therefore, he used the NPV factor for 2005 (i.e., 0.57884) and discounted the future value to arrive at a present day, terminal value of approximately $4,414,000. (Tr. 3/26/01 p. 181.) By adding the total of the present value figures (i.e., $2,169,200) to the terminal value (i.e., $4,414,000), he arrived at his income approach valuation of $6,583,100. (Tr. 3/26/01 p. 184; ACR report p. 19.)
As a final step, Garrett subtracted the amount of debt Tang told him existed, resulting in a value for equity holders. Because the amount of debt the Debtor owes is an issue in the remaining part of the contested confirmation matter, his calculation of the equity holders' value need not be discussed further.
Debtor's counsel's cross-examined Garrett extensively and continually raised his Daubert objection; he conducted a thorough cross examination on Garrett's use of seven versus eight months of the Debtor's actual numbers. (E.g., Tr. 3/27/01 pp. 118-136.) According to that part of his testimony, Garrett only used seven months of the Debtor's historical figures in calculating cash flow (as shown in Table 2-C) and eight months for sales and revenue calculations (as shown in Table 2-A). He explained that in his meeting with Doug Taff and Andre Labranche, two Trimingham employees, they discussed the Debtor's financial statements and, as a result of that discussion, it was clear to him that the adjustments Trimingham was going to be make would be on the expense and not the revenue side. (Tr. 3/27/01 p. 130.) As a result, Garrett was more comfortable using the December revenue number (i.e., the "eighth" month) when he did his sales and revenue averages. (Tr. 3/27/01 p. 130.)
Garrett did not fair so well during most of his remaining cross examination. The transcript shows vigorous questioning on the following topics: his limiting condition of "time and money"; his definition of "analysis"; his educational background and general understanding of statistics; his calculation of the Debtor's cost of equity, cost of debt and WACC; his use of a "control premium"; and the effect of the Debtor's backlog on its numbers during the "seven month period." (Tr. 3/27/01 pp. 26-29, 30-34, 47-59, 66-79, 99-107, 117-129, 160-166.) In addition to going over the typographical errors he had already testified about during direct examination, Debtor's counsel elicited during cross examination that Garrett had used the Securities and Exchange Commission's ("SEC") website to verify the debt to equity ratios of the three companies he used in his cost of equity determination, but only listed that source of information site as "internet" in his report. (Tr. 3/27/01 pp. 96-97.) It was during cross examination that the court became very aware of Garrett's demeanor in handling questions from the Debtor's attorney; it is also when the court overheard several remarks he made that, after trial, were not contained in the transcript.
Cross examination of Garrett's findings occurred in another sense: during the direct examination of the Debtor's rebuttal expert, Robert Jones ("Jones"). Jones testified that he disagreed with not only with Garrett's conclusion but with two other areas of his report: his reliance on what Jones called "some relatively invalid statistical data" and his failure to take into account the effects of financial leverage on the return on equity. (Tr. 3/27/01 p. 230.) Based on his own review, he found that the data of approximately 850 companies was reported under SIC code 349 and concluded that because Garrett's use of a sample of three as representative of the industry was "not even reasonable," it rendered the reliability of his conclusion as to value "very unreliable." (Tr. 3/27/01 pp. 230-233.) As for Garrett's debt to equity ratio of 72.9% to 27.1%, Jones opined that such leverage would require an upward adjustment to the return on equity, which, in turn, would increase the WACC and decrease the present value of the Debtor's enterprise. (Tr. 3/27/01 p. 234; See Tr. 3/27/01 pp. 235-237.)
Jones testified extensively regarding his own calculation of cost of equity at 36.3% plus the additional 10% premium Garrett had used. In arriving at the 36.3% figure, he began with the debt to equity ratio of zero as shown in the "latest data" for the same three companies Garrett used and, using Garrett's figure from Table 3 of the ACR report, a cost of equity of 14.7%. (Tr. 3/27/01 p. 240; Ex. M p. 5.) Using the zero ratio, he calculated the WACC for those three companies at 14.7%, testifying that if there was zero debt in the capital structure there would have to be 100% equity, thus, doing the arithmetic: zero times the cost of debt (i.e., 6.67%) plus one times the cost of equity (i.e., 14.7%) equals 14.7%. (Tr. 3/27/01 pp. 240-241.) Then, taking Garrett's 27.1% to 72.9% debt to equity ratio and using the same mathematical formula (i.e., weight of debt x cost of debt + weight of equity x cost of equity = 14.7%), he calculated a cost of equity of 36.3%. (Tr. 3/27/01 pp. 241-244.) According to Jones, 36.3% equaled the return on equity needed for the three companies Garrett used; it did not, according to him, reflect any company specific risk for Plattco. (Tr. 3/27/01 p. 244.) To obtain the company specific risk, he added the same 10% premium Garrett had used, resulting in a cost of equity of 46.3%. (Tr. 3/27/01 p. 245.) Using Garrett's debt to equity ratio, Garrett's 6.67% cost of debt and his own, newly calculated 46.3% cost of equity, he obtained a WACC of 17.41%. (Tr. 3/27/01 p. 245.)
After replicating Garrett's numbers from Table 5 of the ACR report, Jones used the 17.41% WACC he calculated and then discounted Garrett's cash flow figures and terminal value figure to present value in the same manner Garrett had, arriving at a DCF valuation of $3,980,800. (Tr. 3/27/01 pp. 250-252; Ex. M.) After subtracting out total debt of approximately $4.3 million and making the same "control premium" and "marketability discount" adjustments Garrett had made (as shown on Table 5 of the ACR report), Jones arrived at a negative equity value of $296,500. (Tr. 3/27/01 pp. 251, 253; Ex. M.) Jones further testified that by deducting a 2% growth figure from his 17.41% WACC and using Garrett's capitalization of earnings methodology and his NOPAT figure for the year 2000 as shown on Table 6 of the ACR report, a capitalization of earnings value of approximately $3,838,000 resulted. (Tr. 3/27/01 p. 253.)
Direct examination of Jones ended with his opinion of how much additional cost of equity would drive the value of equity to zero. Starting with the 36.3% cost of equity calculated from the "sample of three," Jones determined that an additional 1.8% would drive the value of equity to negative $5,000, "a pretty close approximation to zero." (Tr. 3/27/01 pp. 255-256.)
According to him, any investor who demanded a premium higher than 1.8% would only "drive the value of equity further into zero." (Tr. 3/27/01 pp. 256-257.) His testimony not only left the court favorably impressed with his knowledge and expertise, it also gave it a greater understanding of how Garrett performed his income approach valuation.
On cross examination, Tang asked Jones about the rate of return for venture capital; he replied, "in the neighborhood of fifty percent." (Tr. 3/27/01 p. 258.) Tang also questioned him about his direct testimony regarding risk, the weighted average cost of capital and the Debtor's assets. Specifically, she wanted to know how the intangible assets of the Debtor could impact the rate of return equity holders would demand, especially when Jones had not looked at the Debtor's assets and when certain intangible ones, like its trade name, had never been valued.
Jones responded that perhaps an asset of the Debtor would, in some small way, impact what an equity holder would want as a rate of return. (Tr. 3/27/01 p. 263.) Countering that relatively small point on redirect, Jones testified that Garrett's projections, presumably, took into account all of the Debtor's assets because those that had value would contribute to the cash flow of the business. (Tr. 3/27/01 p. 264.)
At the close of trial, the court informed the parties that it did not find Garrett credible, leaving the parties to brief the issue of the credibility of Evans's testimony. (Tr. 3/28/01 pp. 7, 26-28.) Further elaborating, the court stated Garrett's lack of credibility was based on a number of factors, including the court's experience as a trier of fact, the witness's demeanor and, most importantly, specific remarks he made — remarks that perhaps he did not want on the record but were heard nonetheless. In response to the court's statement, Tang informed the court that Mr. Garrett had the flu for the last three days of trial, a fact now in the record. (Tr. 3/28/01 p. 27.)
In a chambers' conference, the court also informed the parties that it was not inclined to merely accept Evans's valuation number.
Argument
The Debtor's argument supporting its Daubert objection is largely, if not entirely, factual. Similar to its pre-trial objection, it points out flaws Garrett allegedly made in applying his methodology, the mathematical errors he allegedly made and his alleged undisclosed/incorrectly disclosed/incompletely disclosed sources. Other than providing complete cites to the two controlling Supreme Court cases involving the admissibility of expert testimony, the Debtor's post trial brief does not contain any case law analysis.
The Unsecured Creditor's Committee post trial brief is similar to the Debtor's; it extols the merits of Evans's valuation and attacks Garrett's. It also contains legal argument regarding the various provisions of section 1129(b), argument the court will be addressing at a later date.
The remaining portion of the Debtor's post trial brief discusses the merits of the experts' valuations and is entirely factual. Asserting Evans used the DCF method of valuation, the Debtor argues its expert was correct in the four respects his report differed from Garrett's, i.e., cash flows, debt structure, WACC and ultimate valuation. Beginning with cash flow, the Debtor argues the Guibords did not dispute its historical "negative net income," thus, they did rebut the "zero equity" valuation. In his attempt to dodge the zero value, the Debtor asserts Garrett used only seven months of data, distorting its historical numbers.
Moving on to Garrett's calculation of the WACC, it propounds that he "manipulated" it in order to generate value, contrary to what the Debtor's actual performance shows. The Debtor reiterates what it calls Garrett's "fatally flawed attempt" to value a bankrupt company using a capital structure based on more debt than it has now. The balance of the Debtor's argument focuses on Evans's "normalization value" methodology, Garrett's attempt at what the Debtor calls "normalization" by only using seven month of cash flow in his DCF calculation and the fact that a great deal of Evans's testimony was uncontroverted.
The Debtor also discounts Garrett's "market approach to value." As the court has already decided, this valuation technique is not being considered.
Seemingly with a focus on the court's initial impression of Garrett's valuation, Barrie Guibord's attorney concentrated her post trial legal argument on discounting Evans's valuation methodology, particularly his use of "projected" gross sales for the year 2000 that were approximately $1 million lower than "actual" sales and his "normalization" approach using RMA data for the Debtor's operating and debt expenses. As for the latter, Tang characterizes Evans's application of the RMA-compiled 29.6% gross profit margin instead of the Debtor's historical one of around 40% as arbitrary and his similar application of his debt to capital ratio of 22.5% to 77.5% as mere application of RMA compiled data without any analysis of the Debtor's actual operations.
Robert Rock, Peter Guibord's attorney, did not file a post trial brief.
As for the Debtor's Daubert objection, Tang asserts both experts used the same "approach" although differences "between (their) opinions and methadology [sic]" exist. She argues whether "errors and contradictions" render a report inadmissible is the nature of the court's gatekeeper function under Daubert and Kumho; she asserts Garrett's mistakes "are not so fatal as to render the (opinion he) reached wholly unreliable."
Discussion
I. Admissibility of the ACR Report and Garrett's Testimony
In its decision In re Valley-Vulcan Mold Co., the Bankruptcy Appellate Panel for the Sixth Circuit reviewed various Supreme Court cases involving the issue of admissibility of expert testimony under Fed.R.Evid. 702, including Daubert, Kumho and General Elec. Co. v. Joiner, 522 U.S. 136 (1997). In re Valley-Vulcan Mold Co., 237 B.R. 322 (B.A.P. 6th Cir. 1999). The maxims of the highest court of the land on the issue of expert testimony are well known, perhaps due in large part to the amount of litigation surrounding them, and this court need not repeat all of them here. To summarize, a Rule 702 inquiry into an expert's reasoning or methodology is a flexible one where the trial judge has both considerable leeway in deciding how to go about determining the reliability of an expert, including whether the Daubert factors are reasonable measures of reliability in the case at hand, and broad discretion to determine the admissibility of the expert testimony. See Kumho, 526 U.S. at 138; Daubert, 509 U.S. at 594; Valley-Vulcan, 237 B.R. at 335.
As far as the necessity of a formal Daubert/Kumho hearing is concerned, the Supreme Court has stated, "The trial court must have the same kind of latitude in deciding how to test an expert's reliability, and to decide whether or when special briefing or other proceedings are needed to investigate reliability, as it enjoys when it decides whether or not that expert's relevant testimony is reliable." Kumho, 526 U.S. at 152; See In re Syed, 238 B.R. 133, 142 (Bankr. N.D. Ill. 1999). Of course, failure to raise a Daubert/Kumho issue waives it. Syed, 238 B.R. at 142 (citing International Surplus Lines Ins. Co. v. Fireman's Fund Ins. Co., 998 F.2d 504, 507, n. 2 (7th Cir. 1993)). Because neither side fully briefed the Daubert/Kumho issue, the court conducted its own research. It found one case in the Northern District of New York that involved arguments very similar to the Debtor's, but its facts are not entirely on point.
In Cayuga Indian Nation of New York v. Pataki, 83 F. Supp.2d 318 (N.D.N.Y. 2000), prior to the commencement of the jury trial, the district court dealt with the admissibility of the testimony of a real estate appraiser proffered as an expert on the value of ancestral land from which the plaintiff was dispossessed in violation of the Indian Trade and Intercourse Act. In reviewing the reliability consideration and the factors outlined in Daubert (i.e., testing, peer review, error rates and acceptability), the district court decided those factors were not particularly helpful in terms of ascertaining the reliability of the proposed appraisal method given the novel valuation issue. Id. at 322-323. In reviewing the expert's methodology, the court found his application of what was otherwise a recognized appraisal method (i.e., the sales comparison approach) problematic because he did not comply with established appraisal practices in collecting his sales data, he did not follow his own procedure when selecting representative sales, he relied solely on his subjective "feeling" and was unable to articulate with any degree of specificity the basis for his comparable sales selections, his report contained many reporting inaccuracies and he failed to make any adjustments whatsoever despite recognizing the possibility of the need for them. Id. at 323-25. All of these problems made it impossible for the district court "to ascertain with any degree of confidence the reliability of [the expert's method] — both in terms of the underlying data and in terms of its application." Id. at 324-25. In determining the expert's report was not admissible, the district court pointed out, however, that the focus of a Daubert inquiry is on principles and methodology, not on conclusions. Id. at 322 (citing Daubert, 509 U.S. at 594).
Relevancy is also a consideration, but it is not at issue here.
As already indicated above, despite Debtor's counsel's similar arguments regarding Garrett's errors and lack of complete disclosure, the facts of Cayuga Indian Nation are distinguishable from the instant case. Here, the trier of fact is not a jury. The fast track the valuation hearing the parties asked for and received apparently prevented depositions of the experts, inquiries that might have flushed out the ACR report's typographical mistakes or inadequately disclosed sources. To the court, the shortened notice valuation hearing not only precipitated the Debtor's eve of trial Daubert objection, it played a role in causing the errors in Garrett's report. As the court sees it, virtually all of the errors alleged in the Debtor's pretrial objection could have been sufficiently covered during a pretrial deposition by Debtor's counsel as revealed by the extensive cross examination he conducted at trial. Moreover, the valuation issue before this court, the enterprise value of the Debtor's business, is nowhere near the novel valuation issue the district court faced. Garrett did apply a valid, although admittedly "typographically flawed," methodology, however, the dictate of Daubert is on methodology, not conclusions. Daubert, 509 U.S. at 594. For all of these reasons, the court overrules the Debtor's Daubert objection and receives both the ACR report and Garrett's testimony into evidence.
II. The Experts' Opinions
Valuation is not an exact science; it often requires compromising the opinions advanced by the two battling experts. See In re Smith, 267 B.R. 568 (Bankr.S.D.Ohio 2001) and cases cited therein. Courts generally evaluate a number of factors in weighing conflicting appraisal testimony, including, "`the appraiser's education, training, experience, familiarity with the subject of the appraisal, manner of conducting the appraisal, testimony on direct examination, testimony on cross examination and overall ability to substantiate the basis for the valuation presented.'" Id. at 572 (quoting Buckland v. Household Realty Corp., 123 B.R. 573, 578 (Bankr. S.D. Ohio 1991)). A bankruptcy court may render its own opinion regarding value; it need not adopt the value contained in either expert's report, even if uncontroverted. See In re Smith, 267 B.R. at 572-73 (citing In re Abruzzo, 249 B.R. 78, 86 (Bankr.E.D.Pa. 2000)); In re Mussa, 215 B.R. 158, 166 n. 2 (Bankr.N.D.Ill. 1997); In re Opelika Mfg. Corp., 66 B.R. 444, 450 (Bankr. N.D. Ill. 1986) (citing Security First Nat'l Bank of L.A. v. Lutz, 322 F.2d 348, 355 (9th Cir. 1963)); Buckland, 123 B.R. at 578-579. The "own opinion" route is the path the court will take in the valuation matter at hand.
The court finds that despite the parties' request to have the valuation trial fast-tracked, their respective experts were not prepared to undertake a thorough analysis of the Debtor's business records in such a short period of time, even with the additional time they requested and received for discovery. Although not quite what might otherwise be characterized as "rush jobs," the court finds neither report and neither witness's testimony persuasive enough to adopt either expert's findings in full.
Furthermore, despite Tang's decision not to pursue discovery violations against the Debtor or Trimingham, the court is not convinced Garrett was not "stone-walled" in a manner that thwarted his efforts, to some degree, to obtain the same financial and accounting information Evans seemed to have no trouble accessing.
The parties and their experts have provided the court with little, if any, guidance on the various valuation approaches and methods used, particularly the generally accepted principles and the formulas behind each and their relative strengths and weaknesses. Based on its own research, the court has learned the following:
Regarding discounted income method and capitalization of earnings method, the two valuation methodologies used by the experts, noted valuation expert Shannon Pratt ("Pratt") explains the essential difference between the two methods: "A discount rate converts all of the expected future return on investment (however defined) to an indicated present value. In contrast to the more comprehensive method of discounting all of the expected returns, a capitalization rate converts only a single return flow number to an indicated present value." SHANNON P. PRATT ET AL., VALUING A BUSINESS: THE ANALYSIS AND APPRAISAL OF CLOSELY HELD COMPANIES ("VALUING A BUSINESS"), p. 204 (4th ed. 2000) (emphasis in original). He further explains, "The capitalized economic income method is used as frequently as the discounted income method, and probably even more frequently in the valuation of smaller businesses . . . [T]he capitalized economic income method is simply an abridged version of the discounted economic income method." Id. at 205.
The transcripts contain numerous references to Pratt as "an expert" or as "reliable in the [valuation] industry" by both parties as does the post trial brief submitted by the Debtor. The court views these references as both parties' recognition of Pratt's authority in the valuation arena; its independent research confirms that recognition.
The discounted income method depends on three factors: cash flow projections, a discount rate and a terminal value; the capitalized economic income method relies on two: a normalized level of earnings for a given period and a capitalization rate. See In re Mahoney, 251 B.R. 748 (Bankr.S.D.Fla. 2000); In re Fiberglass Indus., Inc., 74 B.R. 739 (Bankr.N.D.N Y 1987). In its simplest terms, value under the capitalization of earnings approach equals net annual operating income divided by a capitalization rate and, under the discounted income approach, it equals cash flow projections for a given period discounted by the discount rate plus the cash flow projection for the terminal year discounted by the difference between the discount rate and the growth factor. See Eugene D. Lanier, Inc. v. Commissioner, T.C. Memo. 1998-7; In re 203 North LaSalle Street Ltd. Partnership, 190 B.R. 567, 574 (Bankr.N.D.Ill. 1995). Although both methods involve converting anticipated income to a present value, they treat anticipated changes in future income over time differently. VALUING A BUSINESS, pp. 204, 209.
According to Pratt, in discounting, expected economic income changes are treated specifically in terms of the numerator whereas in capitalizing, changes over time in the expected economic income are treated in the denominator, specifically as a single average percentage change subtracted (assuming it is positive) from the cost of capital. Id. at 204. Pratt elaborates, "The important conceptual underpinning of the capitalized economic income valuation model is that there is either a constant annual income stream in perpetuity or a constant annualized rate of growth (or decline) in the economic income variable being capitalized in perpetuity. Obviously, this constant growth rate projection is rarely met in the real world." Id. at 205. He further states, "Unlike the discounted economic income model, the capitalization model does not take into consideration the timing of future changes in expected economic income. The greater the differences in the anticipated changes over time, especially in the early years, the more the analyst is encouraged to apply the discounted economic income method rather than the capitalized income method." Id.
Courts and commentators alike regard the income approach to valuation as the most accurate of the three approaches to use when valuing a business as a going concern, particularly in a bankruptcy context. See In re Consolidated Rock Prod. Co. v. DuBois, 312 U.S. 510, 525-526 (1941); Mahoney, 251 B.R. at 752 n. 4 (Bankr.S.D.Fla. 2000); Peter V. Pantaleo and Barry W. Ridings, Reorganization Value, 51 Bus. Law. 419 (1996); VALUING A BUSINESS, p. 40. Both the capitalization of earnings method Evans used and the DCF method Garrett used are income approaches. VALUING A BUSINESS, pp. 154, 204. In reaching this decision, the court has focused only on those methods, together with the rebuttal testimony offered by Jones.
Of course, if the other approaches (i.e., market and asset-based) are also used and the valuations using those approaches closely match the one obtained using the income approach, such valuations provide additional, persuasive evidence that the income approach rendered an accurate valuation of the business.
A. Evans's Capitalization of Earnings Method
Evans offered no explanation why he chose a capitalization of earnings method versus the more complex DCF method, the method of choice when changing conditions impact a company's financial structure, revenue sources and profits such that historical earnings and cash flow do not provide an accurate picture of a company's future prospects and, as best as the court can tell from its own limited research, the income method used by many, if not most, bankruptcy courts when conducting valuation hearings in a reorganization context. Mahoney, 251 B.R. at 753, n. 7; Jay W. Eisenhofer and John L. Reed, Valuation Litigation, 22 Del. J.C.L. 37, 99-101 and 115 (1997). Evans also used his own particular way of "normalizing" the Debtor's expenses and debt and not the generally accepted way of adjusting for non-recurring items. Furthermore, his normalization was largely based on RMA data he obtained from compilations under "plumbing and heating valve" SIC code 3494 not "industrial valve" SIC code 3491.
By using the RMA information for SIC code 3494, Evans obtained a GPM "compilation" that did not come close to matching the Debtor's quite consistent historical figures, possibly because the margins for the two different categories of the valve industry are very different; he did not explain that his use was largely due to the similarities between the two categories nor did he make any adjustment for the apparent differences. Instead, his testimony was that the companies under SIC code 3494 provided a sampling size of 18 whereas SIC code 3491did not have any compilation for similar sales, leaving the court with the impression that he was more interested in performing his calculation using a large sample than performing some analysis using data from companies with a closer match. The court would have preferred to see what valuation would have resulted had he used the same normalization/capitalization of earnings method together with the RMA compilations under SIC code 3491, using similar assets as the Debtor instead of the similar sales criteria. If a similar valuation had resulted, he would have had support for his use of SIC code 3494. Worst case scenario, because of his credible explanation that he prefers "similar sales" as a criterion versus "assets," a dissimilar result would have been evidence of nothing at all. The court views his failure to perform that calculation as further evidence that time was a significant factor in terms of the amount of evaluation and analysis each expert performed.
As for Debtor's attorney's assertion that Evans did, in fact, use the DCF method, he either misstates his position, did not carefully read Evans's testimony or does not understand all of the differences between the two methods as Pratt, an expert he cites in his post trial brief and referred to at trial, and other courts have presented them. His proposition is intriguing if one considered which half of the DCF calculation Evans's valuation left out, the projection period or the terminal value? Since terminal value is often 50% or more of the total value when using the DCF method, the resulting valuation would be very close to the one Garrett obtained, if not more than $6 million. VALUING A BUSINESS, p. 216. Additionally and more to the point, the court has already found that Evans testified he used the capitalization of earnings method.
Evans's capitalization of earnings method does not leave the court with sufficient comfort to merely adopt his valuation of $3 million. Taking the unusual normalization route he took, it does not appear that Evans followed the formula the capitalization of earnings method generally requires, i.e., taking a projection of income over a period of time based on historical information and dividing it by a capitalization rate. Instead, Evans gave the court what the Debtor would be if it mimicked some possibly very dissimilar RMA compilation-derived, non-industrial valve company with $5 to $10 million in sales.
B. Garrett's DCF method
As found above, Garrett made many off the record remarks. To the court, they undermined the confidence he had in his own findings. He did, however, use the DCF method, and, unlike Evans, he used the Debtor's historical figures for debt and expenses and the generally accepted way of normalizing those figures by subtracting non-recurring items. His use of only seven months of historical sales figures is arguably inconsistent with Rev. Rul. 59-60, but, as he testified, he did in fact review more than just those numbers and, after considering what information he was able to obtain from the Debtor's management, chose the seven-month period based on statements made directly by one, and possibly two, Trimingham employees.
The court's problems with his valuation, however, do not end with his unease at trial nor his use of seven months' of historical data.
Similar to the concern the court had with Evans's use of data for companies under SIC code 3493, the court has problems with Garrett's use of his "sample of three" in his cost of capital calculation, particularly after cross examination revealed he knew nothing about the three SIC code 349 companies he used. It also finds his WACC is beneath the lower end of the spectrum for this debtor. Although partly due to the cost of capital he derived from the sample of three, his WACC of 11.55% resulted, in large part, from his 2.7 debt to equity ratio, a ratio the court believes is on the high end for the Debtor as demonstrated by the difficulty it had handling its high debt load when it filed and its inability to make its first set of plan payments. As the DCF formula provides, if the cost of capital were higher and the debt to equity ratio lower, a higher WACC would have resulted, thereby resulting in a lower valuation.
For all of these reasons, the court continues to find that it cannot accept the valuation Garrett proposes.
C. Jones's Rebuttal Testimony
Of the three experts the court heard from at trial, Jones provided the clearest and most informative testimony regarding the two income methods Garrett and Evans used. The court finds his explanation of adjusting the WACC upwards to reflect the impact such high leverage has on the cost of capital credible. It recognizes, of course, that the Debtor offered Jones's valuation as rebuttal, not evidence in chief. However, given that the valuation of approximately $4 million is very close to the midpoint of the two income method valuations Evans and Garrett proferred and given its closeness to the $4.2 million sale offer Olsen Technologies made, albeit unconsummated, the court concludes the use of the more accurate WACC of 17.41% and its application in accordance with the unchallenged DCF formula Garrett reflects what the court concludes is the Debtor's business enterprise worth: at most, $4 million.
CONCLUSION
Based on the above findings of fact and conclusions of law the court allows the admission of the ACR report and Garrett's testimony and finds the value of the Debtor's business enterprise is, at most, $4 million. It will conduct a status conference on February 1, 2002 at 10:00 a.m. to discuss all remaining issues regarding confirmation of the modified plan.