Matter of Unimet Corp.

74 B.R. 156, 1987 Bankr. LEXIS 781

• Court: United States Bankruptcy Court, N.D. Ohio
• Decided: February 25, 1987
• Docket: 19-30488
• Status: Published

Opinion

FINDINGS AND CONCLUSIONS RE: OBJECTION TO CLAIM NO. 441

JAMES H. WILLIAMS, Bankruptcy Judge.

The Chapter 11 debtor and debtor in possession, Unimet Corporation, fka The Union Metal Manufacturing Company (Uni-met), has objected to the $830,444.91 proof of claim, as amended, filed by the United States of America on behalf of the Internal Revenue Service (IRS or government). The IRS claim asserts liabilities for 1979 to 1983 corporate income taxes, 1984 and 1985 Federal Insurance Contribution Act (FICA) taxes, 1984 Federal Unemployment Tax Act (FUTA) taxes, interest on all unpaid taxes up to the petition date and penalties for failure to pay the taxes in a timely fashion. After several months of discovery and negotiations, a one week trial was scheduled on the matter. Immediately prior to trial, a proposed settlement on several of the issues in dispute was reached by the parties, and has since been approved by the Joint Committee on Taxation of the United States Congress. 1

At the trial the court heard testimony and received evidence on the remaining issues, which essentially involved the valuation of assets of three companies acquired by the debtor several years before it filed for relief before this court: Car Rack, Inc. (Car Rack), Munck Systems, Inc. (Munck) and a subsidiary of Munck, Morg Controls, Inc. (Morg) (Sometimes hereinafter the latter entities may be treated as one and referred to as Munck/Morg). 2 A chart, as prepared by the debtor, which indicates the basic valuation issues presented, is set forth in Appendix A. See also, Debtor’s Proposed Findings of Fact and Conclusions of Law with Respect to the Debtor’s Objections to Claim No. 441 at 4. At the conclusion of the trial, the court established a schedule for the filing of proposed Findings of Fact and Conclusions of Law. Both parties have submitted their proposals and after review of the testimony, evidence and pleadings, the court now enters its decision.

I.

FACTS

Unimet purchased all of the shares of Car Rack in 1979 and all of the shares of Munck/Morg in early 1981 for a net purchase price of $12,300,000.00 to form a subsidiary of the debtor called Intech Systems, Inc. (Intech). Pursuant to Section 334(b)(2) of the Internal Revenue Code the debtor treated these stock acquisitions as purchases of the assets of the companies. The debtor then allocated the purchase price of the stock to each company’s assets based upon the debtor’s determination of the fair market value of the assets as of the date they were liquidated into the acquiring entity. 3

On March 8, 1985, Unimet filed a voluntary petition for relief under Chapter 11 of Title 11 of the United States Code. The court set July 15, 1985 as the deadline for filing proofs of claim. On July 9, 1985 the IRS filed its proof of claim totaling $941,-439.74 for corporate income, withholding, FICA and FUTA tax liabilities. The debtor filed an objection to the IRS’ claim on January 29, 1986 and the government amended its claim on February 20, 1986 to reduce the claimed tax liability to $830,441.91. On August 22,1986, three days prior to trial, a statutory notice of deficiency was sent by the IRS to the debtor. The notice informed the debtor that a deficiency of $18,228.00 in tax liabilities existed for the years 1976, 1979, 1982 and 1983.

A

VALUATION

The debtor’s valuations of the assets in dispute are based upon appraisals conducted by independent appraisers near the time of the purchases. The debtor hired Manufacturers’ Appraisal Company (MAC), a recognized and experienced appraisal firm, to conduct an evaluation of certain of the real and personal property of Munck/Morg and Car Rack. George Sees, Executive Vice President of MAC and an employee with 18 years’ experience, testified as to his knowledge and opinion of the methodology and techniques utilized by MAC in conducting the appraisal. Mr. Sees based his testimony on a review of the final appraisal reports and interviews with MAC personnel.

The normal MAC procedure in evaluating real property begins with an inspection of the property by an individual with expertise in the real estate appraisal field. This individual is responsible for collecting all available and relevant data concerning the property, including information on topography, size and shape of building, type of construction and comparable sales of similar property in the area. The appraiser performs an analysis of the functional, economic and physical depreciation of each asset and, if appropriate, utilizes compara ble sales of similar property to assist in determining fair market value.

Three appraisal methods were used by MAC to arrive at a figure for fair market value for real property: cost, market data and income. The cost approach determines fair market value by estimating the cost of reproducing new the buildings and improvements and then reducing that value for depreciation. The land value is then added to the result to produce a total property value. The market data approach develops a value for property through a comparison of similar property sales. Adjustments are made to the comparable sales to reflect differences between the comparable and the subject property. The third method, the income approach, determines value by capitalizing the net future income that the property is capable of producing.

The MAC methodology used to appraise personal property is similar to that performed for real property. An MAC appraiser takes inventory of the equipment and machinery, noting its age, condition and marketability. The MAC research department then compares this data to prices gathered from machinery and equipment dealers and from its own price files to establish a figure for fair market value.

MAC defines fair market value in its appraisal reports essentially as follows:

Fair market value as part of a going concern is considered to be the price expressed in terms of money which a willing and informed buyer would pay for the property to a willing and informed seller, neither acting with undue haste, both exercising prudent judgment, and with the buyer contemplating continued use of the property....

See, Fair Market Value Report, Car Rack, Inc., by The Manufacturers’ Appraisal Company, Debtor’s Exhibit 3 at 1; The Manufacturers’ Appraisal Company Report, Munck Systems, Inc. and Morg Controls, Inc., Debtor’s Exhibit 2 at 1. Mr. Sees conceded that this definition, which indicates that a going concern value has been included, is confusing and misleading. Mr. Sees stated that the term “going concern” as used by MAC was a term of art to indicate that the assets were valued “in place.” Mr. Sees further explained that “in place” value represents a value which includes such items as the cost of shipping, dealer installation and wiring. Mr. Sees distinguished “in place” value from going concern value and stated that going concern value was the additive value for “an assemblage of items,” and was a “premium over and above installed costs.”

With respect to the accounts receivable, Lee J. DiCola, President and Chief Executive Officer of the debtor, a certified public accountant and a former accountant with the firm of Ernst & Whinney, testified that he engaged Ernst & Whinney to audit the accounts receivable of the companies and to make a recommendation as to their value. Mr. DiCola testified that such an audit should include a review of the accounts and an evaluation of the risk inherent in each receivable and a discount therefor.

The debtor hired Dr. Phillip V. DeSantis, an engineering consultant with experience and knowledge in the computer field, in December, 1980 to evaluate 40 basic programs of the Munck/Morg computer software system. Dr. DeSantis did an actual study of a portion of the system, evaluating computer code line by line to determine its length, complexity and quality, and arrived at an opinion of fair market value using both a future profit stream and a replacement cost analysis.

In contrast, the government witnesses and experts did not do an on-site evaluation of the debtor’s assets, but instead critiqued the appraisals performed for the debtor. The IRS did not hire independent appraisers to conduct its own evaluation of either the real property or personal property of Munck/Morg or Car Rack. Rather, the government offered the testimony of Greg Zivoder, a general engineer for the IRS, as to the value of the tangible and intangible assets of the debtor’s acquisitions. Mr. Zivoder’s testimony criticized the debtor’s evidence and allocated the purchase price to the assets in dispute using the balance sheets of the companies. Dr. George S. Sacerdote, the holder of a Ph.D degree in mathematics and an employee of the Ar thur D. Little Co., testified as to the value of the computer software. According Dr. Sacerdote, Dr. DeSantis’ report contained several flaws, but he used the same basic methodology to form his own view of the fair market value of the software.

1.

BUILDINGS

MAC performed appraisals of both Munck/Morg’s Hampton, Virginia facility and Car Rack’s Macedonia, Ohio plant on behalf of the debtor. The income approach was omitted in both the Munck/Morg and Car Rack appraisals as both buildings in question were properties which are usually owner occupied and not leased. The evaluations of the facilities were conducted on site by individuals with over 15 years of experience in real estate appraisals.

The MAC report lists the fair market value of Munck/Morg’s ten acre Hampton, Virginia facility at $1,064,000.00. The ten year old building is described as “[o]ne 34,188 sq. ft. 1.1 story average industrial plant erected 1970-1980,” consisting of “33% office area, 26% storage and 59% production.” (Sic) See, The Manufacturers' Appraisal Company Report, Munck Systems, Inc. and Morg Controls, Inc. Debtor’s Exhibit 2 at p. 11. The plant site is located on six acres of land with four adjoining acres to the north.

Using the cost approach, the MAC appraisal calculated the fair market value of the Hampton facility to be $1,064,000.00. MAC arrived at a total current plant replacement cost of $1,014,800.00 and reduced that figure by 18% for physical depreciation with no functional and economic depreciation. The land was given a total value of $232,000.00 consisting of $132,-000.00 for the plant site and $100,000.00 for the excess land. Under the market data approach, MAC utilized three comparable sales to arrive at a fair market value of $1,006,000.00. MAC’S comparables and resulting calculations are listed in Appendix B. Mr. Sees testified that the ultimate determination of $1,064,000.00 as a fair market value was due to the fact that the comparables were not sufficiently similar and that sound judgment dictated use of the cost approach.

The Car Rack Macedonia, Ohio facility contains a total land area of 20.6 acres. A one story, 46,401 square foot industrial building is located thereon. The building was constructed in the 1950’s with major additions built in 1971, 1973 and 1979, resulting in varied story heights and an irregularly shaped building. Approximately 58% of the building was erected after 1971.

The MAC cost approach indicates a total value for the Macedonia facility of $806,-000.00. MAC calculated the cost of the plant new to be $788,817.00 and deducted 27% from that figure for depreciation, leaving a value of $575,836.00. Site improvements added $30,000.00 to its value, while the land value was calculated to be $200,-000.00. Although the text of the MAC report indicates that a physical depreciation factor of 25% and a functional depreciation factor of 15% was appropriate, the actual calculations took into account only 22% physical depreciation and 5% functional depreciation. Mr. Sees indicated that the latter was the correct percentage after consultation with the appraiser. The market data approach was also relied upon, utilizing four comparables. Adjustments were made for time, location and physical characteristics as reflected in Appendix C yielding a $650,000.00 value for the building area, including the land, and a total of $800,000.00, including what is designated as excess land. This time MAC found the market data approach to be more reliable and ultimately valued the property using those figures. Its report suggests, however, that an appropriate allocation of values would be $600,000.00 for building and site improvements and $200,000.00 for land. Mr. DiCola testified that the debtor’s valuation of the Car Rack building at $612,-549.00 instead of the $600,000.00 recommended by MAC could be attributed to the fact that the appraisal was conducted in December, 1979 whereas the relevant valuation date for purposes of the Internal Revenue Code was April, 1980.

The government’s sole evidence as to the valuation of the facilities was the testimo ny of Mr. Zivoder. With respect to the Hampton facility, Mr. Zivoder testified that he had not been furnished a copy of the MAC appraisal at the time of his investigation, and accordingly based his evaluation on the balance sheet of the company. Mr. Zivoder stated that the value of the buildings generally will not increase over time since depreciation normally offsets inflation and he thus valued the building at $710,597.00 as reported on the Munck/Morg balance sheet. As to the Macedonia facility, Mr. Zivoder used a value of $570,396.00, again based upon the cost basis before depreciation reported in the balance sheet. Mr. Zivoder argued that the MAC Car Rack building appraisal was flawed as it did not take into account any economic depreciation. Mr. Zivoder also criticized both MAC appraisals as they improperly included, in his opinion, a going concern value in their fair market value definition.

2.

MACHINERY AND EQUIPMENT

The debtor valued the machinery and equipment of Munck/Morg at $468,600.00 and that of Car Rack at $832,700.00, based upon the MAC appraisals. The appraisal reports are grounded upon actual inventories and inspections of each item, along with price research. Mr. Sees testified that a figure higher than cost can be reached under certain circumstances and gave a specific example. On the other hand, the IRS witness, Mr. Zivodfer, indicated that the figures presented in the MAC report were unsubstantiated, did not take into account depreciation, and erroneously included a going concern value. Again, the IRS expert based his opinion of value on the cost basis before depreciation as listed on the balance sheets to arrive at figures of $244,-363.00 for Munck/Morg and $680,080.00 for Car Rack.

3.

LEASEHOLD IMPROVEMENTS

Munck/Morg’s MAC appraisal valued certain leasehold improvements such as partitions, shelving, air conditioning, electrical service and light fixtures at a total of $16,500.00. Mr. Sees explained on cross examination that MAC appraised the value of the assets in that particular location and not, as the IRS argued, the value of the installation of the assets. Mr. Zivoder testified that the values given by MAC were unsubstantiated, that the improvements were improperly valued as a going concern and that the leasehold improvements were not carried on the balance sheets. Accordingly, Mr. Zivoder indicated that a zero value was appropriate.

4.

ACCOUNTS RECEIVABLE

Mr. DiCola stated that he used the valuations reported to him by Ernst & Whinney in his allocation of the purchase price and that these figures should have been discounted for collectibility and other risks by Ernst & Whinney. Mr. DiCola verified that all receivables, except retentions, were collected within 45 days and that retentions were received within the first year. Mr. Zivoder testified that he discounted the re-tainage portion of the Munck/Morg receivables by 10% to compensate for risks and costs of collection and the time value of money. Mr. Zivoder indicated that he had no knowledge of the Ernst & Whinney audit and noted that the Car Rack receivables have been properly discounted by the debt- or.

5.

SOFTWARE

The major item in dispute is the valuation of the computer software system of Munck/Morg. Munck/Morg was in the business of developing Automated Storage and Retrieval Systems (ASRS) and the computer software necessary to run such systems. The debtor valued the software at $10,000,000.00, while the government alleges that its value, if any existed, was no more than $1,140,000.00. For tax purposes, the debtor allocated approximately $7,200,000.00 of the purchase price to the software. During its reorganization, the debtor sold the software for $50,000.00 to Ranger Automated Systems, Inc. (Ranger). Unimet had earlier sold Ranger the rights to market the software in South Africa and certain of its former colonies for $425,-000.00 plus a royalty.

Unimet decided to get into the automatic storage and retrieval business in 1980 due to what it perceived, to be a high growth potential in that industry. The debtor investigated nine companies in the ASRS business and decided to acquire Munck/Morg based upon growth reports. Mr. DiCola testified that while Munck/Morg had a low awareness factor in the industry, (only 9.3% of those surveyed had ever heard of Munck/Morg), of those who knew, 75% preferred the company as compared to a 60% preference factor for the industry leader. Once the debtor initiated negotiations for the Munck/Morg purchase, the debtor attempted to verify the value of Munck/Morg’s assets by independent means. The debtor hired Dr. De-Santis, the holder of a Ph.D degree in computer engineering from Case Western Reserve University, to study the computer software. Dr. DeSantis specialized in For-tran programming, the major language used in the Munck/Morg software, and also had significant experience in several other programming languages. Dr. De-Santis had previously performed services for the debtor on a free-lance basis and both parties had been satisfied with the arrangement.

The debtor requested Dr. DeSantis to evaluate the Munck/Morg library based on a review of 40 basic programs. Dr. DeSan-tis devoted over 200 hours to this effort. During his first visit to the Munck/Morg facilities, Dr. DeSantis realized that actual reading and counting of each line of software in the library would be an impossibility in the time allotted. Accordingly, he decided to physically count and review the code in one ASRS project, ARAMCO, and then extrapolate those results over five other projects which were readily available. Dr. DeSantis would then calculate his final evaluation of the entire library, which included a total of 20 different projects.

Dr. DeSantis determined the number of reusable lines in the entire library to be 205,000, calculated as follows: First, he omitted 14 projects from his review and concentrated solely upon the six projects most readily available. Second, he estimated the number of lines contained in the six projects utilizing his detailed analysis of the ARAMCO project. Third, he testified that he subtracted from the number of lines in the six projects a sum representing duplicates and comment lines. 4 Fourth, he reduced the total lines by 23,000 to account for functional depreciation. From these calculations, Dr. DeSantis arrived at his total of 205,000 reusable lines.

The next step in Dr. DeSantis’ cost evaluation was to compute a cost per line for the software. He testified that a Munck/Morg programmer cost approximately $25.00 per hour including overhead and profit. The $25.00 per hour figure was low, according to Dr. DeSantis, who indicated that the figure would be higher in other geographical areas, e.g., $40.00 per hour in Cleveland. He further testified that the average national productivity rate for a programmer creating code similar in complexity to Munck/Morg code was 100 lines per month or four lines per day. Using various commercial publications, the national productivity rate and the cost per hour of Munck/Morg programmers, Dr. DeSantis concluded that a cost of $50.00 per line was appropriate. By multiplying the 205,000 lines by the $50.00 per line cost, Dr. DeSan-tis concluded that the value of the software was approximately $10,000,000.00.

Dr. DeSantis also did what he admitted to be a “rough” calculation of a 10-year future profit stream for the software. Using unverified sales and productivity figures, Dr. DeSantis calculated the profit stream to be approximately $19,000,000.00 and indicated on direct testimony that a five year analysis would have yielded a figure of approximately $8,300,000.00. Dr. DeSantis admitted that he was not a finance expert, but stated that he used the profit stream figures as a cross check on his reproduction cost analysis. Mr. DiCola testified that while the debtor found Dr. DeSantis’ profit stream analysis corroborated his reproduction cost analysis, it did not rely on the profit stream analysis to fix its ultimate determination of value.

Finally, Dr. DeSantis found that in 1981 approximately 60% of the modules in a new project could be taken from the existing software library, while 40% would need to be created. His report indicated that the Munck/Morg staff totaled 17, of which 12 were programmers.

As earlier noted, the government hired Arthur D. Little, Inc. to undertake an independent examination of Dr. DeSantis’ work and offered the testimony of Dr. George Sacerdote of that company. Dr. Sacerdote assumed that Dr. DeSantis’ raw data was correct and concentrated solely on his economic analysis.

Dr. Sacerdote found two key errors in the DeSantis estimate for the reproduction cost of the reusable lines: (1) the assumption that nearly all lines of code constitute a useful asset; and (2) the assumption that computer costs equaled programmer costs. Dr. Sacerdote found three basic flaws in DeSantis’ profit stream analysis: (1) inclusion of ten years in the analysis when depreciable life of software is only five years; (2) failure to allow for the time value money; and (3) use of inflated profit margins due to failure to account for sales and administrative expenses.

Using Dr. DeSantis’ data, Dr. Sacerdote calculated that only 12% of the total lines in the six projects were ever used in more than one system. He found that the six projects contained a total of 299,036 lines (total lines studied), of which 227,065 or 76% were dissimilar (i.e., unique) lines (total dissimilar lines). The balance of 71,971 represented lines reused (reused lines), although only 27,329 of those were reused for the first time (reusable, lines). 5 He arrived at the 12% figure by dividing the number of reusable lines by the total number of dissimilar lines. Dr. Sacerdote also accounted for the possibility that a portion of the un-reused lines would eventually be reused. Accordingly, he determined that an estimate of 5% was appropriate and concluded that a total of 37,300 lines were reusable/potentially reusable. His calculations may be shown as follows:

$77,000.00 to $102,000.00, calculated thus: $27,000.00 to $37,000.00, the cost of financing a computer system for the seventeen programmers and analysts that Munck/Morg employed; $37,000.00 to $49,-000.00 for depreciation over five years and a factor of 20% for other supplies, power, etc., or $13,000.00 to $17,000.00. Based upon these computations, Sacerdote concluded that the total cost for reproducing the 37,300 lines would be between $850,-000.00 and $1,140,000.00 or, as indicated, $23.00-$31.00 per line.

Dr. Sacerdote also performed his own future profit stream analysis. Using Dr. DeSantis’ sales projections, current price earnings ratios of comparable companies and a discount of 15% to reflect the time value of capital, he concluded that a figure of $1,440,000.00 was an accurate estimate.

On cross examination, Dr. Sacerdote admitted that his evaluation assumed that the total library consisted of only the lines actually studied and not the 1,000,000 lines it actually contained. Dr. DeSantis admitted on cross examination that he had not subtracted the comment lines from his calculations as he had testified, but insisted that comment lines were valuable to a programmer. Dr. DeSantis also admitted that using his cost per line of $50.00, the cost of producing the average new program at Munck/Morg would be $1,000,000.00 because, according to his report, 40% or 20,- 000 lines of the average 50,000 line program needed to be newly created. Dr. DeSantis verified that his own calculations indicated that the debtor had received only a net revenue of $216,000.00 per program to cover the software cost.

Finally, Mr. DiCola testified that the software was sold to Ranger under distressed conditions. He indicated that Ranger was the only interested purchaser and that the low purchase price was a result of the lack of a going concern.

B.

MISCELLANEOUS DEDUCTIONS

Unimet contests the IRS’ denial of five deductions allegedly taken by the debtor on its 1980 and 1981 consolidated corporate income tax returns. 6 The debtor asserts that the first three of those disallowances arise from the government’s valuing unnamed buildings and software. The government characterizes the first two disputes as involving disallowance of amortization deductions taken in 1981 for a step-up in basis of the Munck facility and the Morg software. Similarly, denial of a claimed 1980 amortization deduction for a step-up in basis of the Car Rack Building is the focus of the third issue, according to the government. The parties agree that the final two disputes center upon the tax valuation of the Car Rack facility and equipment by the debtor. The debtor believes that it valued the building and equipment at $612,549.00 and $832,700.00, respectively. The debtor objects to the IRS’ assertions that the values actually used by the debtor were $765,596.00 and $1,314,-776.00.

Stuart Barnes, a 31-year veteran of the IRS, outlined the procedure by which the government determined that the five deductions were indeed taken. Mr. Barnes testified that the first three deductions arose when the debtor took a step-up in basis for the Munck/Morg software and the Munck and Car Rack buildings. Mr. Barnes tied together the lump sum figures reported on the debtor’s 1982 and 1980 tax returns, the corresponding Schedule M-l Reconciliation of Income, and the detail provided by the debtor’s comptroller, John Scheetz, to establish that the amortizations had been used. As to the remaining two issues, Barnes indicated that the IRS’ figures resulted from a review of Form 4562 Depreciation and the detail provided by the debtor. Barnes indicated that the figures asserted by the IRS had been computed by adding the book cost to the step-up in basis of the property.

Mr. DiCola testified that Unimet did not take any additional deductions in 1981 for its buildings or software, but did so in 1982. He further indicated that he knew of no additional buildings or software on which the claimed deductions could have been taken in 1981. Mr. DiCola noted that he believed the values utilized by the debt- or for the Car Rack building and the equipment to be $612,548.00 and $832,700.00, respectively, although he admitted he did not prepare or sign the returns.

C

THE TAX BENEFIT RULE — RECAPTURE

The debtor alleges that expenses incurred in producing and creating the Munck/Morg software qualified as research and experimental expenditures deductible under Section 174 of the Internal Revenue Code. The IRS asserts that the software expenses were deducted by the debtor under Section 162 of the Internal Revenue Code and are subject to recapture whether deducted pursuant to Section 162 or Section 174. See, Rev. Rui. 85-186, 1985-2 C.B. 84. The sole evidence present ed on this issue was the testimony of Dr. DeSantis who stated that the computer software was research or experimental in nature when originally developed.

II.

DISCUSSION

A.

BURDEN OF PROOF

Section 502(a) of the Bankruptcy Code, read in conjunction with Bankruptcy Rule Rule 3001(f), establishes that a properly executed and timely filed proof of claim constitutes prima facie evidence of the validity thereof. If the trustee objects to the claim, he must produce evidence of probative force equal to that of the allegations of the creditor’s proof of claim to rebut the presumption in favor of the creditor. Once the trustee succeeds in overcoming the prima facie effect given the claim, the burden rests upon the claimant to prove the validity and amount of his claim by a preponderance of the evidence. 3 Collier on Bankruptcy, para. 502.01[3] (15th ed. 1986); In re DeLorean Motor, 59 B.R. 329 (E.D.Mich.1986); In re Century Inns, Inc. 59 B.R. 507 (Bankr.S.D.Miss.1986). “The burden then rests on the court to determine whether or not the claimant has prevailed, with due regard to the probative value of the proof of claim, for, after all, whether or not a claim is to be allowed in the face of the contest remains a judicial determination.” 3 Collier on Bankruptcy, supra.

The government argues that the burden of proof is different with regard to tax liabilities. It asserts that once a statutory notice of deficiency or assessment has been issued, the burden of ultimate persuasion lies with the objector. The IRS cites United States v. Kontaratos, 36 B.R. 928 (D.Me.1984) and certain nonbankruptcy cases to support its position. See, Sinder v. United States, 655 F.2d 729 (6th Cir.1981), Helvering v. Taylor, 293 U.S. 507, 55 S.Ct. 287, 79 L.Ed. 623 (1935).

In Kontaratos, the district court upheld the bankruptcy court finding that since no assessment had been made, no presumption of correctness attached to the government’s claim. The district court agreed that case law normally supported a presumption in favor of the government and explained:

There are several reasons that courts have given to support this presumption. One reason has been that it is presumed that the IRS Commissioner who makes the assessment has done so in good faith and with sufficient evidence at his disposal to “satisfy a just and fair-minded person that such assessment ought to be made.” United States v. Rindskopf 105 U.S. [15 Otto] 418, 26 L.Ed. 1131 (1881). The presumption of correctness afforded an assessment by the Commissioner encourage (sic) taxpayers to keep accurate records and to avoid the necessity of the government to borrow money to cover the impact of unpaid taxes and to initiate litigation for their recovery. United States v. Rexach, 482 F.2d 10, 17 (1st Cir.1973).

Kontaratos at 930 n. 4. The court then concluded that the presumption of correctness is afforded to the government only if there has been a valid assessment.

It is clear that no assessment has been made by the government on the issues presented in this case, and is now barred from issuing any such assessment by Section 362(a)(6) of the Bankruptcy Code. However, the government argues that since a statutory notice of deficiency has been issued, the same presumption should exist.

This argument- is clearly without merit. First, the court finds it totally inequitable to allow the government to attempt to shift the burden of proof on the eve of the trial. Second, and more importantly, the court has grave doubts about the relevance of the alleged statutory notice of deficiency to the issues at hand. A careful review of the testimony indicates that Mr. DiCola received a statutory notice of deficiency but that it concerned tax liabilities totaling only $18,228.00 and only for the years 1976, 1979, 1982 and 1983. This, coupled with the fact that no notice of deficiency was ever submitted into evidence, leaves the court with the firm conviction that there has been insufficient evidence to demonstrate the existence of a statutory notice of deficiency for the tax liabilities in question. Accordingly, the court concludes that if the debtor can rebut the prima facie validity of the claim, then the government will carry the ultimate burden of persuasion.

The debtor, as previously noted, treated the stock acquisitions of the companies as a purchase of assets and allocated the purchase price according to its opinion of the fair market value of those assets as of the date they were liquidated into In-tech. See, 26 U.S.C. § 334(b)(2); 26 C.F.R. § 1.334-l(c)(4)(viii). The allocated price then became the new tax basis for the assets and any depreciation or amortization was taken with reference to that basis. Thus, if the court finds the value of an asset is lower, the amount' of the allowed depreciation and amortization will be less, and the debtor’s tax liability will be greater.

Fair market value is defined as “the price [on] which a willing buyer and willing seller agree, neither being under any compulsion to buy or sell, and both having a reasonable knowledge of the relevant facts.” 26 C.F.R. § 1.170A-1(c)(2); 26 C.F.R. § 20.2031-l(b), See also, VGS Corporation v. Commissioner, 68 T.C. 563, 589 (1977) citing Jack Daniel Distillery v. United States, 379 F.2d 569, 574, 180 Ct.Cl. 308 (1967). The government argues that the MAC appraisals erroneously include a going concern value in their fair market valuations as the definition of fair market value contained in the MAC reports embodies the language “fair market value as part of a going concern.”

“Going concern value is the additional element of value which attaches to property because of its existence as an. integral part of a going concern.” Black Industries, Inc., 48 T.C.M. (P-H) 243 (1979); VGS Corporation v. Commissioner, supra; Conestoga Transportation Company v. Commissioner, 17 T.C. 506 (1951). It is well settled that going concern value is nondepreciable. As the court in VGS Corporation v. Commissioner explained:

The going concern element of an operating business cannot be classified as an enhancement in market value of depreciable assets for purposes of depreciation. While the individual tangible assets may wear out and be replaced, going concern value does not wear out with the individual assets. And when a worn out tangible asset must be replaced the cost to the business of doing so is not augmented by the fact that the acquisition is to become part of a going concern. Thus it is not within the meaning of section 167 of the Code to allow taxpayers to recoup the cost they paid for the going concern value through an increased depreciation base on the tangible assets.

VGS Corporation v. Commissioner, at 592 citing United States v. Cornish, 348 F.2d 175 (9th Cir.1965). “[A]ny assemblage of assets into a functioning, ongoing business is capable of giving rise to going concern value.” Black Industries, supra at 252.

The court is convinced that going concern value is one of the assets Unimet purchased from Munck/Morg. Munck, and its subsidiary Morg, were viable companies when purchased by the debtor, with existing trained employees, business organizations and business relationships. The testimony of Mr. DiCola that the drastic decrease in value of the software was attributable in part to the absence of a going concern illustrates this point. However, the issue before the court is not whether going concern value was present but whether the MAC appraisals included an increment for that value in each and every item of property it appraised. Mr. Sees testified, and the government’s witness agreed, that it is very difficult if not impossible to attribute a going concern value to an individual piece of property. Theoretically, it is an impossibility. Mr. Sees testified that the term “going concern value” as employed by MAC was a term of art meaning “in place” value and included such items as installation costs. This term is distinguishable from going concern value as used by the government for tax purposes. The court is persuaded that the appraisals performed by MAC did not erroneously include a “going concern value” as the term is used for purposes of the Internal Revenue Code.

Having resolved an issue of general concern, the court now turns to the more specific disputes. The court finds that the MAC appraisals, for the most part, are to be afforded heavy weight. These studies were conducted on site by experienced and independent personnel using standard methodology during the time period in question. In contrast, the IRS utilized its own personnel, did no actual inspection or on site evaluation and performed its evaluation over five years after the relevant time period.

A review of the appraisal of the Hampton, Virginia facility establishes the value of the plant site at $1,064,000.00, allocated as follows: $832,000.00 for the building and $232,000.00 for the land. Both the cost and market data approaches used by MAC support this figure and satisfy the court that the sum is appropriate. However, the appraisal of the Car Rack Macedonia facility contains two flaws in its cost analysis. First, no economic depreciation was taken. Second, percentages used for functional depreciation are too low. The report and testimony indicates that the structure had varied story heights and would be unsuitable for other purchasers. Although the lower market data approach was the one actually relied upon by MAC, the court believes an additional 5% reduction for depreciation is necessary in the cost approach analysis and that the correct value under the cost analysis should be $766,400.00. The court believes a fair market value representing the average between the two methods is appropriate. Finally, the court finds insufficient evidence to substantiate the $12,549.00 increase in value asserted by the debtor. Therefore, the court finds the value of the Macedonia, Ohio facility to be $583,000.00 for the building and $200,000.00 for the land, for a total value of $783,000.00.

The court finds the appraisals performed by MAC to be sound and accordingly finds the fair market value of the machinery and equipment of Munck/Morg to be $468,-600.00 and of Car Rack to be $832,700.00.

The government argues that the value of leasehold improvements on the Munck/Morg property is zero due to the fact that these assets do not appear on the balance sheets. However, the court is persuaded that such improvements did in fact exist. MAC did an on-site evaluation of the property and found certain leasehold improvements to be present and to be of value. No actual inspection was performed by the government’s witness. Further, the court finds that Mr. Sees’ testimony establishes that MAC properly valued the improvements themselves, and not the value of the installation of such improvements. Accordingly, the court finds the fair market value of the leasehold improvements to be $16,500.00 as reported by MAC.

The debtor relies on the testimony of its Chief Executive Officer and President, Mr. DiCola, to establish the value of the accounts receivable. The court finds that the bare allegations of Mr. DiCola are without any substantiation and thus insufficient to rebut the prima facie validity of the IRS’ claim. No report or testimony from Ernst & Whinney was presented. Further, the court finds it unusual that while not discounting the Munck/Morg receivables, the debtor felt it appropriate to discount those of Car Rack. The court believes that the accounts receivable of Munck/Morg should be discounted for the time value of money and the risks and costs of collection. The court finds a re duction of 10% of the retainages, as suggested by the government, to be fair and finds the fair market value of the accounts receivable to be $3,163,719.00.

The most critical issue before the court is the valuation of the computer software of Munck/Morg. The parties, using similar methodology have arrived at figures in excess of $8,800,000.00 apart. Both parties agree that an appropriate methodology to calculate the value of the software is to compute the cost of reproducing the reusable lines of computer code. While both parties also analyzed the value of the software using a profit stream analysis, the evidence presented indicates that the cost reproduction analysis is the preferred method and the one primarily relied upon by both parties.

The court has reviewed both evaluations before it and concludes that each of the reports contains major defects. Dr. De-Santis based his entire analysis on the assumption that 20% of the library was reusable. There was absolutely no factual or evidentiary basis for this assumption. In fact, his very own data from the six projects reviewed reveals that only 12% of the lines were actually reused.

At first blush, Dr. DeSantis’ cost reproduction analysis seems sound. However, a careful examination and check of his calculations leads to a contrary conclusion. As noted, Dr. DeSantis based his entire analysis on the unsubstantiated 20% estimate of reusable lines. Second, he computed the cost per line to be $50.00 which he arrived at by using four as the average production of lines per day per programmer and an hourly programmer cost of $25.00, including overhead and profit. Again, Dr. De-Santis’ own data and calculations contradict his figures. The average project size is 50,000 lines. His report and testimony indicate that approximately 40% of the modules will need to be created new, while 60% will be reused from prior projects. Therefore, according to Dr. DeSantis, 20,-000 new lines must be produced in a project. At a cost of $50.00 per line, a project would cost approximately $1,000,-000.00, yet the average sales price for a project was only approximately $480,-000.00. See, Review of Computer Software, supra at 13. Of the $480,000.00, Dr. DeSantis himself allocated only $216,000.00 to cover the software cost! Id. at p. 15. Further, an analysis of five of the six projects actually reviewed 7 reveals that the maximum cost per new line in any of the projects was $31.50. See Appendix D. 8 Finally, while Dr. DeSantis indicates the average programmer can create only four new lines of code per day in his cost reproduction analysis, in his sales analysis he indicates that only 6,000 Munck/Morg man hours are needed to complete a job with a 50/50 split between old and new code. Again, using a 50,000 line project as typical, this calculates to be over four lines of code per hour, 32 lines per day or 704 lines per month and is seven times DeSantis’ national average. 9

Dr. DeSantis’ profit stream analysis is also subject to question. He did not take into consideration the time value of money and erroneously used a ten-year period. In addition, all of the figures used by Dr. DeSantis were unverified and unsubstantiated.

On the other hand, the analysis completed by Dr. Sacerdote, a mathematics expert, was clear, concise, and supported by the facts. 10 Dr. Sacerdote, using Dr. DeSantis’ raw data, reviewed each line of code in the six projects to arrive at the number of lines actually reused in more than one project. Dr. Sacerdote based his cost per line on trade publications, the Munck/Morg programmers’ hourly rate and the overhead costs in connection with generating the software. Unfortunately, Dr. Sacerdote did not realize the six projects reviewed in Dr. DeSantis’ report did not constitute the entire Munck/Morg software library and he did not extrapolate his results over the remainder of the library. In addition, Dr. Sacerdote did not take into consideration any functional depreciation. 11 Accordingly, the court will undertake its own analysis, basing its methodology upon that presented by Dr. Sacerdote.

The court finds the value of the Munck/Morg software to be $8,364,200.00, calculated as follows:

Total lines studied 299,086

Total dissimilar lines (76%) 227,065

Reusable lines (12%) 27,329

Balance of total dissimilar lines

(227,065 - 27,329) 199,736

Potentially reusable (5%) 9,987

Total reusable/potentially reusable lines 37,316

(Rounded) 37,300

See, A Valuation of the Software Assets of Munck Systems, Inc., by Arthur D. Little, Inc., Creditor’s Exhibit 10 at p. 10.

Dr. Sacerdote also disputed Dr. DeSan-tis’ per-line cost of $50.00. He arrived instead at a cost per line of $23.00-$31.00 by estimating the labor, computer and supply costs involved in producing the 37,300 lines of code. His labor factor is calculated as follows: a “reasonable software engineer” in the witness’ opinion, could design, produce, test and document an average of 150 to 200 lines of code per month; at that rate, 187 to 249 man/months or 15.5 to 20.7 man/years would be required to create the 37,300 lines of code; at $25.00 per hour, the Munck/Morg rate, the typical programmer, working 2000 hours per year would be paid at an annual rate of $50,000.00; labor costs then, in generating the 37,300 lines of reused/potentially reusable code would range from $780,000.00 to $1,040,000.00. The government’s expert then analyzed the computer and supply costs necessary to produce the code. He arrived at an estimate of

Number of lines:

Library 1,000,000

Total dissimilar

lines (76%) 760,000

Reusable lines (12%) 91,200

Less functional depreciation (10%) 12 9,120

Balance of reusable lines 82,080

Balance of total dissimilar lines (760,000 - 91,200) 668,800

Potentially reusable lines (5%) 33,400

less functional depreciation (10%) 3,340

Balance of potentially reusable lines 30,060

Total reusable/potentially reusable lines 112,140

Cost per line 13 $ 30.00

Value of reusable/ potentially reusable lines $ 3,364,200.00

C.

The debtor objects to the government’s disallowance of five deductions allegedly taken by Unimet in 1980 and 1981. The first two issues involve the amortization deductions for the Munck building and Morg software listed in the Revenue Agent’s Report as being taken by the debt- or in 1981. The testimony and evidence presented by the IRS conclusively demonstrates that these deductions were in fact taken, but as part of the 1982 consolidated return. 14 The court also finds that an amortization deduction on the Car Rack building due to a step-up in basis has been established by the 1980 return, supporting detail and the Revenue Agent’s Report.

With respect to the final two disputes, insufficient testimony or evidence was presented to demonstrate what basis was used by the debtor for tax purposes on the return itself. However, consistent with its other findings, the court is convinced that the debtor did take a step-up in basis of the Car Rack building and equipment as illustrated in Exhibit C-37, the Revenue Agent’s Report and the 1982 tax returns. To the extent this finding impacts on these two issues, the court so finds.

D.

Both the debtor and the IRS address the applicability of the tax benefit rule to expenses incurred by the debtor in developing the software. The government argues that any cost incurred by the debtor in connection with the creation or development of the software does not qualify as research and experimental cost under Section 174 of the Internal Revenue Code and should have been deducted pursuant to Section 162 of the Code. In either event, the government asserts that any previously taken deduction will result in a recapture tax after the sale.

The record is totally devoid of any reference to an amount of claimed tax liability, the amount of deductions taken, and the years in dispute with respect to this issue. Without any demonstration by the government as to the nature and amount of its claim, the court finds that the government has not established its prima facie case and accordingly the debtor’s objections must be sustained.

E.

PRE-PETITION INTEREST

The final issue before the court is whether the pre-petition interest accrued on the tax liabilities is to be given priority status along with the underlying tax claim under Section 507(a)(7) of the Bankruptcy Code. There is a split of authority on the question. The clear majority of courts hold that interest is entitled to the same priority as the tax if it is compensatory in nature. See, In re Hirsch-Franklin Enterprises, Inc., 63 B.R. 864 (Bankr.M.D.Ga.1986); In re Reich, 66 B.R. 554 (Bankr.D.Colo.1986); In re Keller and Katkowsky, P. C., 55 B.R. 155 (Bankr.E.D.Mich.1985); In re Palmer, 53 B.R. 545 (Bankr.N.D.Tex.1985), aff’d Bankr.L.Rep. (CCH) para. 71,273 (N.D.Tex.1986); In re Treister, 52 B.R. 735 (Bankr.S.D.N.Y.1985); In re Hernando Appliances, Inc., 41 B.R. 24 (Bankr.N.D.Miss.1983); In re New England Carpet Company, 26 B.R. 934 (Bankr.D.Vt.1983). A minority holds to the contrary. See, In re Razorback Ready-Mix Concrete Company, 45 B.R. 917 (Bankr.E.D.Ark.1983); In re Ayala, 35 B.R. 651 (Bankr.D.Utah 1983).

Both sides of the controversy rely upon the legislative history of the Bankruptcy Code to support their views. At one time the Senate version of Section 507 expressly provided that pre-petition interest was to be afforded the same status as the underlying tax claim. See, Senate Report No. 989, 95th Cong.2d Sess. 20, U.S.Code Cong, and Admin. News 5787, 5806 (1978). However, that provision was deleted. The majority of courts reason that the deletion was because the language was unnecessary in light of the fact that a “claim” includes the right to payment of interest under Section 101(4). The minority justifies its position by stressing the obvious fact that the provision was expressly omitted..

After a review of the relevant case law and legislative history, the court concludes that pre-petition interest on tax claims is entitled to priority if it fits within the scope of Section 507(a)(7)(G) which gives priority to:

a penalty related to a claim of a kind specified in this paragraph and in compensation for actual pecuniary loss.

Three requirements must thus be met to qualify for priority status: (1) the claim must be a penalty; (2) the claim must be compensation for pecuniary loss; and (3) the claim must be related to a tax allowed under Section 507(a)(7).

The court finds that the interest assessed is a penalty, See, In re Palmer, supra at 548-549, compensatory in nature, See, In re Hirsch-Franklin Enterprises, Inc., supra at 873 and related to a claim set forth in Section 507(a)(7). Accordingly, the pre-petition interest claimed by the government is entitled to the same priority status as the tax claim.

III

CONCLUSION

In summary, the court fixes the values of the various items of property in dispute to be as follows:

Munck/Morg Value

Building $ 832,000

Machinery & equipment 468,600

Leasehold improvements 16,500

Accounts receivable 3,163,000

Software 3,364,200

Car Rack

Building 583,000

Machinery & equipment 832,700

The court further concludes that the debtor did not take an amortization deduction in 1981 on the Munck building and Morg software but did take an amortization deduction on the Car Rack building in 1980. The court further finds a step-up in basis on the Car Rack building and equipment was taken by the debtor in amounts as reflected in Exhibit C-37. No evidence was introduced to establish the government’s prima facie case as to software expense deductions and the resultant recapture tax liability. Finally, the court finds that interest assessed by the government on the tax claims is entitled to the same priority as the tax under Section 507(a)(7)(G) of the Bankruptcy Code.

An order consistent herewith shall issue.

APPENDIX A

VALUATIONS IN DISPUTE

Munck/Morg Debtor IRS

Building $ 832,000 $ 710,597

Leasehold Improvements 16,500 0

Machinery and Equipment 468,600 244,363

Accounts Receivable 3,326,062 3,163,719

Software 7,150,462 0

Car Rack Debtor IRS

Building $ 612,549 570,396

Equipment 832,700 680,080

APPENDIX B

COMPARABLE SALES: MUNCK/MORG LANDS AND BUILDINGS

Sale #1 Sale #2 Sale #3

Location 507 Copeland Drive 401 Copeland Drive 413 Industry Drive

Land Area (S.F.) 25,490 37,873 22,800

S.F. Bldg. Gross Area 14,174 6,600 4,600

Age and Condition 8 years 4 years 8 years

Land to Bldg. Ratio 1.8:1 5.74:1 4.96:1

Date 1979 1978 1979

Price $160,000 $125,000 $114,000

Bldg. S.F. Value with Land 11.29 18.94 24.78

Overall Adjustment to Subject + 15 +8 +3

Indicated Subject Value $26.29 $26.94 $27.78

The above sales indicate a given range in value. Considering comparability to the subject, it is our opinion that the concluded value of the subject is:

34,188 S.F. Building Gross Floor Area @ $26.50/S.F.

with Land = $ 905,982

Plus Excess Land 100,000

$1,005,982

(Rounded) 1,006,000

APPENDIX C

COMPARABLE SALES: CAR RACK LAND AND BUILDING

Location 585 Highland Rd. 8531 Freeway Dr. 556 Highland Rd. 9150 Valley View Rd.

Land Area (Acres) 2 1.41 1.96 6.39

Building Area (S.F.) 23,740 12,000 13,000 48,280

Land Coverage 27% 20% 16% 17%

Date 7/78 4/78 1/78 10/77

Price $450,000 $180,000 $275,000 $800,000

Bldg. S.F. Value With Land $ 18.96 $ 15.00 $ 20.22 $ 16.57

Subject property:

46,401 S.F. at $14.00/S.F. $649,614

Excess Land 150,000

799.614

(Rounded) $800,000

APPENDIX D

Project Total Contract Price Number of New Lines Cost per ■ New Line

Burroughs $ 287,217.00 44,727 $ 6.42

Aramco $1,045,416.00 48,088 $21.74

Navy- $ 392,000.00 48,333 $ 8.11

Xerox $ 403,500.00 . 31,860 $12.67

SLCF $ 600,262.00 19,328 $31.06

See, Review of Computer Software Munck System, Inc. Morg Controls, by Dr. Phillip V. DeSantis, Debtor’s Exhibit 1 at p. 13.

See, A Valuation of the Software Assets of Munck Systems, Inc., by Arthur D. Little, Inc., Creditor’s Exhibit C-10 at p. 11.

1 . The debtor alleges in its post-trial memorandum that on or about September 23, 1986 the Joint Committee on Taxation of the United States Congress gave final approval to the proposed compromise. See, Debtor’s Findings of Fact and Conclusions of Law with Respect to the Debtor’s Objections to Claim No. 441 at 2. The compromise, as of this date, has not been approved by this court.

2 . At the final pre-trial conference, counsel for the IRS indicated that the government would be seeking leave to amend its proof of claim to include certain excise tax liabilities. At the conclusion of the trial, the court heard oral argument and received evidence on the government’s motion and subsequently denied that request. See, In re Unimet Corporation, No. 685-00240 (Bankr.N.D.Ohio Nov. 17, 1986). That decision is currently on appeal to the United States District Court for the Northern District of Ohio.

3 . Car Rack was liquidated into Intech in April of 1980 while Munck/Morg was liquidated in March, 1981.

4 . Neither Dr. DeSantis’ report nor his testimony indicates a figure for the total number of lines in the six projects prior to his reduction for duplicates and comment lines. However, Dr. Sacerdote’s analysis reveals this figure to be 299,036. According to Dr. DeSantis' report. elimination of comment lines and duplicates resulted in a total of 228,696 lines of code. See, Review of Computer Software Munck System Inc., Morg Controls, by Dr. Phillip V. DeSantis, Debtor’s Exhibit 1 at C-2. (Review of Computer Software)

5 . An example helps clarify this point. Program A2A3 contains 249 lines and is found in each of the six projects studied. See, Review of Computer Software at C-3. Dr. Sacerdote’s analysis takes into consideration the fact that the figure for total lines studied includes program A2A3 six times, while the computation for reused lines accounts for program A2A3 five times. By calculating the number of "dissimilar" lines and lines reused for the first time, Dr. Sacerdote eliminates this problem.

6 . The court is unclear from the evidence presented whether the deductions were disallowed only for the years specifically addressed, 1980 and 1981, or if other tax years are at issue. The court realizes that the IRS’ testimony and evidence was constructed at the last minute due to its lack of knowledge that the debtor was still disputing the denial of the deductions. Although the court feels bound to limit its analysis only to the years directly addressed by the parties and will not independently review each tax year in dispute, it notes that its analysis may impact on other tax years.

7 . No total sales cost was given in Dr. DeSantis’ report for the RRAD project. In addition, the Collins-Melbourne and Avon projects were not a part of Dr. DeSantis' study group.

8 . The court notes that its calculations yield the highest possible cost per line and uses a zero profit margin.

9 . The debtor alleges that these differences can be attributed to the fact that Dr. DeSantis’ reproduction cost analysis utilizes a national productivity rate whereas his future profit stream analysis bases its findings on the Munck/Morg productivity rate. As indicated, the Munck/Morg rate is seven times that of Dr. DeSantis' national rate. The court finds Dr. DeSantis’ national productivity rate to be unrealistic based upon all the evidence presented.

10 . The court has independently spot checked Dr. Sacerdote’s calculations, and in some instances completely verified his numbers. The court’s own computations reveal an approximate 2% margin of error as compared with Dr. Sacerdote’s.

11 . Testimony indicates that approximately 23,-000 lines or 10% of the 227,000 total lines were functionally obsolete. While the debtor had accounted for this loss in its calculations, the government did not. For example, program FEXEC2 is listed by Dr. DeSantis as an obsolete program but is included in Appendix C of his report from which the government based its calculation of reusable lines. See, A Review of Computer Software, supra at C-ll, E-l. Accordingly, the court believes a 10% reduction is appropriate since a portion of the reusable lines are obsolete.

12 . Id.

13 . The $30.00 cost per line is based upon a 150 line per month average at a rate of $25.00 per hour for programmers, including overhead. This figure is supported by the Salt Lake City project which had a 55% to 45% split between old and new lines and a maximum cost of $31.06 per line. See, Appendix D.

14 . See n. 6.