Burlington Northern Railroad v. Bair

815 F. Supp. 1223, 1993 U.S. Dist. LEXIS 3073, 1993 WL 65678

Opinion

RULING AND ORDER

STUART, Senior District Judge.

The Burlington Northern Railroad Company (BN or the railroad) brings this action under section 306 of the Railroad Revitalization and Regulatory Reform Act of 1976 (§ 306 or 4-R Act), codified at 49 U.S.C. § 11503, against Gerald Bair, the Iowa Director of Revenue and Finance (Director), seeking relief from alleged discriminatory real property taxes for the 1989 assessment year. Section 306 prohibits the imposition of any discriminatory tax on railroads or rail property, confers federal court jurisdiction and creates an express federal injunctive remedy to enforce such prohibition. 1

These same parties were involved in similar litigation in this court for tax years 1981 and 1982. 2 After remand from the Eighth Circuit Court of Appeals, 766 F.2d 1222 (8th Cir.1985), this court, in an effort to determine the fair market value of BN’s real property in Iowa, stated several principles to guide the parties in making necessary computations. The cases were settled without further litigation. Both parties claim to rely in this litigation upon those principles set forth by the court.

BN claims that the taxes assessed against it by the State of Iowa for the tax year 1989 violate section 306 in four respects:

A. The Director overvalued BN in violation of section 306(l)(a);

B. The Director undervalued comparative commercial and industrial property in violation of section 306(l)(a);

C. The Director improperly apportioned BN’s value between real and personal property resulting in BN being assessed on tangible personal property in violation of section 306(l)(d);

D. The Director improperly taxed BN’s intangibles compared with other commercial and industrial intangibles in violation of section 306(l)(d).

The Director denies that BN was discriminated against in any way by the real property tax imposed on BN by the State of Iowa.

Several factors combined make the resolution of these claims complex and difficult. Although the parties agree that BN should be valued on a business enterprise basis as a unit, and they agree on the portion of that value assignable to the State of Iowa, the Director valued BN’s rail transportation property at $4.9 billion while BN claims its true market value did not exceed $2.5 billion. The court must arrive at its own true market value. Because Iowa does not tax most of the personal property in the state, it is necessary to determine what portion of BN’s true market value assigned to Iowa is personalty. It must also be determined whether intangible personal property is actually taxed in Iowa, and if not, what can be considered intangible personalty and its value. The court must also determine what value the intangibles contribute to the true market value of BN’s Iowa property. According to the Eighth Circuit Court of Appeals, it is the responsibility of this court to weigh the evidence and make its own factual findings regarding valuations. Bair II, 766 F.2d at 1225-26. A formidable task.

Because of the number of issues and the length and complexity of this opinion, the court believes that it would be helpful if an outline of the opinion is included. Therefore, such outline follows.

I.

BURDEN OF PROOF

II.

OVERVALUATION OF BURLINGTON

A. “Method” or “Application”

B. Discounted Cash Flow

C. Income Capitalization Indicator

1. Unrealistic Income Stream

(a)

Single year NROI

(b)

Accounting change

(c)

Add-back deferred taxes

(d)

Leased equipment adjustment

2. Appropriate Capitalization Rate

(a)

Return for debt holders

(b)

Return for shareholders

(c)

Proportion of debt capital to equity capital

(d)

Computation of discount or capitalization rate

3. Computation of Value Using Income Capitalization Indicator

D. Stock and Debt Indicator

E. Computation of True Market Value

III.

TANGIBLE PERSONAL PROPERTY BREAKOUT

IV.

INTANGIBLE PERSONAL PROPERTY DEDUCTION

A Taxation of Commercial and Industrial Intangibles

B. Value of Intangible Assets

V.

EQUALIZATION

A Assessment Jurisdiction

B. Commercial and Industrial Assessment Ratio

VI.

SUMMARY

I.

BURDEN OF PROOF

The burden is on BN, as the party attacking the accuracy of the Director’s values, to show by a preponderance of the evidence what the accurate values are. See Michigan Wis. Pipe Line Co. v. Iowa State Bd. of Tax Review, 368 N.W.2d 187 (Iowa 1985) (burden of proof does not shift from taxpayer). In evaluating the evidence, the district court should give due deference to the Director’s expertise in valuation. Because there is no intent element in section 306, “Burlington Northern need only prove the accurate values, not purposeful undervaluation or overvaluation.” Bair II, 766 F.2d at 1226.

II.

OVER VALUATION OF BURLINGTON NORTHERN

A. “Method” or “Application”

In Iowa, as in most states, railroads’ operating property is valued for property tax purposes as a unit. Under the unit concept of value, the fair market value of railroad operating property is measured by appraising the whole property as a going concern because it derives its true value not from the sum of its parts, but from the'integrated use of the parts to carry traffic, render public service, and from the investors standpoint, generate income.

In determining the true market value of BN as a unit, the Director used the stock and debt and the capitalized income approaches. These traditional approaches are widely accepted and used by appraisers, and they were used by BN in the prior litigation. BN now claims, however, that these methods are defective and that BN’s true market value should be determined by using only the discounted cash flow (DCF) method.

The first issue before the court is:

[W]hether a railroad may, in an action under [§ 306] challenge in the district corrt the appropriateness of the account-in methods by -which the State determined the railroad’s value, or is instead restricted to challenging the factual determinations to which the States preferred accounting methods were applied.

Burlington N. R.R. v. Oklahoma Tax Comm’n, 481 U.S. 454, 463 n. 5, 107 S.Ct. 1855, 1861 n. 5, 95 L.Ed.2d 404 (1987).

The Supreme Court expressed no view on that issue. Id. The issue was later present ed to the Eighth Circuit Court of Appeals in Burlington N. R.R. v. James, 911 F.2d 1297 (8th Cir.1990), but the court avoided the issue by assuming it had no jurisdiction under section 306 to hear a challenge to valuation methodology. The court held:

The stipulation between the parties at least arguably framed the issue as one involving application of an accepted methodology. Certainly BN’s claim to that effect is sufficient to survive the jurisdictional hurdle.

Id. at 1301.

The court also stated:

We recognize that our holding is not trouble-free. The parties agree that in cases like this the line between “methodology” and “application” is so thin as to be almost ephemeral. The district court obviously believed that, despite the parties’ characterization of the issue, it would be required to weigh competing methodologies were it to determine the merits. That belief may or may not prove to be true. BN, however, unquestionably has presented a color-able allegation that the methodology used by Minnesota has resulted in a discriminatory tax. That allegation alone is sufficient to establish subject matter jurisdiction.

Id. This issue thus remains one of first impression in the Eighth Circuit.

In James, Judge Rosenbaum, considered BN’s position as seeking “to have [the trial] court compel the State of Minnesota to use [BN’s] alternative valuation method in lieu of the state’s accounting method.” Burlington N. R.R. v. James, 725 F.Supp. 1058, 1063 (D.Minn.1989), rev’d, on other grounds, 911 F.2d 1297 (8th Cir.1990). He found “subject matter jurisdiction to be limited to a determination of whether there is a rational basis and non-discriminatory purpose underlying the State of Minnesota’s valuation methodology.” Id., 1063-64.

I, like Judge Rosenbaum, find Judge Jenkins’ analysis of that issue in Union Pac. R.R. v. State Tax Comm’n, 716 F.Supp. 543, 553 (D.Utah 1988) persuasive.

Congress did not intend ‘for the federal courts to become involved in establishing ... procedures for the states in valuing railroads.’ [Burlington N. R.R. v. Lennen, 573 F.Supp. 1155, 1164 (D.Kan.1982) ].... It is one thing to say that the 4-R Act allows federal courts to review railroads’ claims of overvaluation. It is quite another to say that the Act dictates a state’s choice of valuation methods. The 4-R Act may provide relief if a state overvalues a railroad by misapplying its chosen methodology or by using a methodology that has no rational basis or is chosen for the purpose of overvaluing railroads.

In Chesapeake W. Ry. v. Forst, 938 F.2d 528, 533 (4th Cir.1991), cert. denied, — U.S. -, 112 S.Ct. 1577, 118 L.Ed.2d 220 (1992), the railway brought an action under section 306 challenging the State of Virginia’s “over the fence” method of valuing railroad property and arguing for use of the unit method. The Fourth Circuit stated:

We conclude that § 306 does not provide a basis for railroads to challenge a state’s preferred methodology for ascertaining the true market value of railroad property. The statute simply does not authorize the judiciary to enter into the difficult task of second-guessing a state’s policy assessment of how -best to value property. In light of Congress’ expressed preference against federal interference in state taxation policy, we believe that § 306 should be read narrowly, authorizing challenges to a state’s calculation of fair market value, but not to a state’s method for calculating such value. If Congress determines that one valuation methodology produces more accurate values than another, it may require states to utilize that accounting method. Section 306, however, does not authorize the courts to perform that task.

In § 306, Congress has insured that railroad and non-railroad property will be treated similarly under whatever valuation methods have been selected by a given state. If the Congress further desires to standardize the valuation side of the-taxation process, it could set out the methodologies it thinks are fairest to railroads. The courts, however, cannot make that judgment without completely displacing the fact-finding and policymaking powers of state government. The 4-R Act does not authorize the judiciary to undertake the difficult, and arguably impossible, task of determining which specific assessment methodology produces true market value of railroad property. Consequently, we affirm the order of the district court granting summary judgment.

Id.; see also Richmond, Fredericksburg, & Potomac R.R. v. Forst, 797 F.Supp. 494 (E.D.Va.1992) and Burlington N. R.R. v. Department of Revenue, Nos. C89-518(T)WD, C90-5417(T)WD, 1992 WL 201947 (W.D.Wash. Aug. 13, 1992).

The court has found no cases indicating that there is jurisdiction in this court to order taxing authorities to use a specific accounting method to determine values. 3

BN relies on the Eighth Circuit’s statement in James, 911 F.2d at 1301, that the difference between methodology and application “is so thin to be almost ephemeral,” to argue that the stock and debt, capitalized income, and discounted cash flow approaches are “applications”, not “methods”. The court does not agree. All three methods of valuing property as a going business are widely accepted. Both BN and the Director used stock and debt and capitalized income in arriving at their true market values of BN’s railroad property in Bair III. Now BN argues that the DCF approach is the only proper approach to use in arriving at the true market value of the railroad property.

The appraisals are theoretical and based on assumptions and estimates that are at best, educated guesses. Experts agree that comparable sales provide the best way to appraise property. But if comparable sales are not available, as is the case with railroads, appraisers do the best they can with the information available to them.

Each method may represent what some buyers and sellers actually do. All'the methods may be equally rational given their underlying assumptions. And they are all irrational if pressed to ex-tremes____ Each method or theory depends on certain assumptions that cannot ultimately be proved or disproved by reason alone nor replicated in experience. Thus, this court cannot say that any one method is necessarily more rational than any other. Nor can the court say that one method alone arrives at the railroad’s ‘true market value.’ Rather, the evidence suggests that the. term true market value is a judgment not subject to mathematical precision that is based on a wide variety of factors and is at best an approximation.

Union Pac. R.R. v. State Tax Comm’n, 716 F.Supp. at 555.

The line between method and application may at times be “ephemeral”. Nevertheless, I believe that a bright line can be drawn between formulas that are used in the various methods of appraising and the specific numbers, percentages, types of property and any other matter that the individual appraisers use to make the formula applicable to the specific appraisal in question.

This court holds that stock and debt, capitalized income, and DCF are methods of arriving at true market value of BN’s unit value, and that the court does not have jurisdiction to order the State of Iowa to use any specific method of appraisal. In order for BN to prevail on its claim that only the DCF method Of appraisal is the proper one, it must establish by a preponderance of the evidence that there is no rational basis for the State of Iowa to use the stock and debt and capitalized income approaches, or that these approaches were used to discriminate against BN. Burlington N. R.R. v. James, 725 F.Supp. 1058, 1063 (D.Minn.1989), rev’d on other grounds, 911 F.2d 1297 (8th Cir.1990). Here, the Director used the appraisal methods both parties used in the prior action that were approved by the court.

The court holds that BN has failed to carry its burden of proving that there is no rational basis for the appraisal methods used by the Director to arrive at the true market value of BN’s real property- in Iowa. There is no claim that the methods were used to discriminate against the BN.

B. Discounted Cash Flow

The Eighth Circuit Court of Appeals requires the trial court to weigh the evidence and make its own factual findings regarding valuations. Bair II, 766 F.2d at 1225-26. The question then arises: Whether the trial court can consider the evidence of true market value of BN’s operating property as determined by the DCF method. The court has determined that it cannot. The court must arrive at its factual findings as to values using the capitalized income and stock and debt methods used by' the Director. Otherwise, the court would be, if .effect, ordering the Director to use DCF method to arrive at true market value. 4 The court, therefore, will not base its decision on the relative merits of the methods of appraisal used here. Rather, the court will limit its decision to BN’s claims that the Director did not properly apply the appraisal methods he used in arriving at true market value of BN’s operating property.

C. Income Capitalization Indicator

Under the income approach, the Director capitalized the 1988 calendar year net railway operating income (NROI) using a discount rate determined in the Director’s normal manner. BN claims the Director used an unrealistic income stream as NROI and an improper capitalization rate.

1. Unrealistic Income Stream

BN claims the income stream was unrealistic because the Director: (a) used a single year’s income; (b) failed to allow for the change in accounting, practices; (c) added back deferred taxes; and (d) added back separately valued operating leases.

(a)

Single years NROI

The Director used as NROI the income figures for the latest year to forecast BN’s income for the next year. BN claims the use of only one year’s income “leads to problems of variability in the assessment from year to year arising from transient economic results.” The same argument was made in Bair III. The court there held that it was not error to use the latest year’s income saying:

Although the Department’s use of current earnings may cause the projected income to fluctuate more from year to year, there was no showing that over the years plaintiff would be prejudiced. Very few railroad appraisers use the five year income average. The evidence has shown that the operating income of Burlington Northern has risen substantially over the last several years-and its president expects the trend to continue. This evidence, coupled with the general long term inflationary trend, persuades the Court that the use of the current year’s operating income to forecast the next year’s income is more appropriate. Use of the five year average would result in a consistent under estimation of plaintiffs earnings for the next year, and consequently undervalue plaintiffs transportation property. If the trend should reverse and the operating income decrease, the plaintiff would have the immediate advantage of the effect of that decrease on the next years evaluation.

Bair III, 648 F.Supp. at 95. No evidence has been produced or arguments made that would cause the court to change its position.

However, BN has produced evidence showing that the Director has not reduced the assessed value in the same percentage as the decreases in NROI. If the Director continues to use the latest year’s income, the assessed value, less changes otherwise accounted for, should change in proportion to the change in income.

The Director did not err in using the income from the latest year as NROI. The figure he used and the one the court will use is $432,280,000. No claim is made for an adjustment based on this evidence.

(b)

Accounting change

BN claims that the change from RRB accounting (retirement, replacement or betterment) to depreciation accounting makes “book income” less suitable as a starting point. The court does not agree. Although the change may shift capital expenditures to different years because certain maintenance expenses will now be depreciated instead of deducted in the year they were made, BN will still recover all of its costs.

(O

Add-back of deferred income taxes

The Director followed this court’s ruling in Bair III by adding deferred income taxes back into the income stream. BN claims this is error because: “First, it is premised on the doubtful assumption that such taxes are not merely deferred, but will actually never be paid. Second, by making such an adjustment, the Director intrinsically mismatches the income stream to capitalization rates derived by market measures relative to book income. In effect, he compares apples to oranges. Third, and perhaps most importantly, such a practice is not commonly accepted as proper by appraisal experts.”

The court discussed this issue in Bair III stating:

The Court is of the opinion that the assumption that deferred taxes will continue to grow and actually will never be paid back is supported by the studies and logic. If deferred taxes are not added into the income stream, the plaintiff would have an increased flow of capital that would not be capitalized. On the other hand, if investments would be less than the depreciation, a negative deferred tax situation would develop where more taxes were actually being paid than would have been paid under the straight-line method, plaintiffs income would be reduced by that amount and the property value lowered accordingly. As it appears that a consistent underestimation of operating income will probably develop if it is not determined by taxes actually paid, and as there would be no prejudice over the long run to plaintiff if deferred taxes are not. treated as an expense in arriving at income, the Court believes that the plaintiff has failed to establish by a preponderance of the evidence that the defendant erred in computing operating income by using taxes actually paid and not treating deferred taxes as an expense.

Bair III, 648 F.Supp. at 96.

I have not changed my opinion. The Director was correct in adding $52,375,000 to the NROI.

'(d)

Leased equipment adjustment

BN leases and has exclusive use of certain operating personal property such as locomotives, railcars and computers. Because commercial and industrial personal property is exempt from taxation under Iowa law, the value of leased operating property is not part of BN’s value to be taxed. The Director separately values BN’s operating leases and adds that value to its unit value estimations.

BN claims this approach “double counts the value of this leased equipment because the value of such property is already subsumed within BN’s operating income (which was capitalized by the Director) and is captured in BN’s stock, price which the Director uses as a surrogate for asset value in its stock and debt indicator.”

The Director argues that “under Iowa law the court does not look to ownership alone as the controlling factor in whether property is taxed to a railroad. Rather the court determines whether the railroad is using the property exclusively for railroad purposes. Where property is used exclusively for railroad purposes, regardless of the ownership of the property, the court will uphold the property being separately taxed to the railroad.”

The Director’s argument misses the point. Leased personal property cannot be taxed under the holding in Bair II and Bair III. The question is whether the lease add-back is required to arrive at the true market value of BN as a unit. The income generated by non-capitalized operating leases is reflected in the income stream in income approach and in the stock price in the stock and debt approach. Adding the values back double counts the operating lease value. Burlington N. R.R. v. Department of Revenue, Nos. C89518(T)WD, C90-5417(T)WD, 1992 WL 201947 (W.D.Wash. Aug. 13, 1992); Ring & Boykin, Valuation of Real Estate, p. 434. The Director erred in adding operating lease values to its unit value estimations.

2. Appropriate Capitalization or Discount Rate

After the appropriate cash flows or income streams are identified, the appropriate rate to discount cash flows or capitalize income must be determined. This element is the most sensitive factor in the valuation of the railroad as a business enterprise in the income approach. The parties arrived at materially different rates using the same formula. 5

The formula is based on the weighted average cost of capital for the railroad on January 1, 1989. It takes into consideration the rates of returns on their investments debt holders and shareholders would demand before investing in BN’s debt instruments or common stock. After those rates are determined separately, they are multiplied by the percentages of debt capital and equity capital that make up BN’s capital structure.

The following matters must be determined: (a) the rate of return demanded by debt holders; (b) the rate of return demanded by shareholders; (c) the appropriate percentage of each that is applicable to BN; and (d) the calculation to arrive at the capitalization or discount rate.

(a)

Return for debt holders

The Director used the rate of 9.875% for the railroad’s “Funded Debt Unmatured” and 9.93% for equipment obligations. BN estimated the cost of debt to be 10.07% before tax and 6.24% after tax based on the average of ratings for Moody’s Aa and Baa railroad bonds for January 1,1989. There is no indication that the Director considered an after tax rate. However, the Director’s expert did reduce the debt yield by the applicable tax rate. The court concludes that the rate of 6.24% is the most appropriate rate for debt financing and should be used in the formula.

(b)

Return for shareholders

The cost of equity financing is defined as the minimum rate of return that a publicly traded company must earn on its equity financed projects in order to leave its stock price unchanged. Both parties used the capital asset pricing model (CAPM) 6 to calculate the rate each deems appropriate for the cost of equity capital.

Under that formula the cost of equity capital is computed by adding the difference between the market risk premium rate appropriate for BN’s equity financing and the risk-free rate of return (usually the rate for short term government securities) to the risk-free rate of return. The greater the risk, the larger the market risk premium rate.

The Director used a risk-free rate of return of 6.9%. BN used a risk-free rate of return of 8.49% based on the rate of interest on one year government bonds on January 1, 1989. In its CAPM model the Director’s expert used a risk-free rate of 8.9 to 9.0%. The court holds that a risk-free rate of return of 8.49% is the most appropriate.

The Director determined that the risk premium was 5.7% above the risk-free premium of 6.9%. The railroad’s expert determined that the risk premium was 8.3% above the risk-free premium of 8.49%. The Director’s expert determined that the risk premium was four to five percent above the risk-free premium of 8.9 to 9.0%.

The Director did not offer an explanation for the figures he used. BN used the research conducted by Ibbotson Associates of Chicago during the period from 1926 through 1987. During that period, common stock investments in Standard & Poors 500 index provided an 8.3% average annual rate of return over that of one year government bonds. The Director’s expert expressed difficulty with the sixty year historical average testifying:

The risk premium, in essence, has to be looked at in terms of where it is today, what’s the expectation in the market today versus the historical average. I believe the historical average is inappropriate. It’s much too high relative to current risk relationships. And I believe looking at a fair amount of data — and that would be different analysts’ reports, various ways of measuring historically the risk premium— that the number is in the 4 to 5 percent range.

BN’s expert responded that he used the long-run average because the seventies and early eighties had such extraordinarily high interest rates that the numbers for that period do not give reasonable estimates of the return equity investors expect compared to treasuries. The court believes that the sixty year period is too long but agrees that the seventies and early eighties were unsettled.

The court is of the opinion that equity investors would require more than a return of five percent above treasuries to invest in BN common stock. After the restructuring BN was highly leveraged with about sixty percent of its capital structure depending on debt instruments which means that a large portion of the cash flow would be paid out in interest and would not be available to the shareholders. A rate of eight percent above risk-free interest or 16.49% is a reasonable market risk premium for 1989.

The court computes the cost of equity capital as follows:

RISK FREE RATE + BETA (Market Risk OF RETURN Premium - Risk Free Rate) = Cost of of Return Equity Capital

8.49 + 1.2 (16.49 - 8.49)

8.49 + 9.6 = 18.09

Rounded to 18%

(0

Proportion of debt capital to equity capital

In 1989 Burlington Northern Industries (BNI), BN’s parent company, underwent a major restructuring. As a result, the natural resources and industrial development properties of BNI were spun off to form Burlington Resources. BNI is now involved solely in railroad transportation. BN is its sole principal asset. After the spin-off of the natural resource assets of BNI and the assumption of BNI’s debt, BN was highly leveraged with debt representing at least sixty percent of its capital structure. The Director used the actual percentage of debt and equity capital to compute BN’s weighted average cost of capital. Expert witnesses for both sides used the average proportion of debt and equity capital typical for the railroad industry. Their figures varied from thirty to forty percent for debt and seventy to sixty percent for equity. Because of the difference in the rate of return required by holders of debt and those investing in stock, the percentage figures are an important component of the formula.

At trial the court questioned the use of industry-wide percentages of debt and equity capital to compute BN’s average cost of capital when BN’s capital structure has a much higher percentage of debt financing. It appears obvious that the capitalization rate to be applied to BN would be affected by BN’s specific capital structure. As the answers given at trial did not satisfy the court, further briefing was requested.

The Director’s additional brief was not helpful to the court. BN’s brief referred to the testimony in the record on this point. The experts for both parties used their version of industry figures. The court is not in a position to quarrel with the experts. The court, therefore, holds that the proper percentages of capital to be used in arriving at BN’s weighted average cost of capital is forty percent debt and sixty percent equity. Use of the actual percentage of debt and equity capital would tend to lower the capitalization rate when the risk to the equity holders would be greater.

(d)

Computation of discount or capitalization rate

Cost of Equity capital x proportion of equity capital to total capital

18 x 60% 10.8

Cost of Debt capital x proportion of debt capital to total capital

6.24 x 40% 2.496

Total 13.296

Rounded to 13.3%

Under the income approach, the capitaliza- railroad system as a unit is 13.3%. tion rate used determine the value of BN

3. Computation of Value Using Income Capitalization Indicator

NROI (II C 1(a)) $ 432,280,000

Non-operating taxes (not an issue) 13,853,125

Deferred Income Tax (II C 1(c)) 52,375,000

Total income stream 498,508,125

Capitalization rate _13.3%

BN’s true market value under the income capitalization indicator $3,748,181,390

D. Stock and Debt Indicator

The stock and debt method of valuing a business enterprise is a substitute for the market approach when no comparable sales are available. The stock and debt indicator is frequently used in the appraisal of large interstate businesses. This method is based on the accounting premise that assets equal liabilities plus equity. Thus, by valuing the liability side of the balance sheet one can infer that the assets are of equal value.

In the earlier litigation both parties used the stock and debt indicator. The court approved of weighting the income indicator and the stock and debt indicator equally. The Director, with the exception of adding back net operating leases, followed the procedure approved in Bair III for the 1989 assessment year. BN’s experts now do not consider the stock and debt approach reliable and, as previously stated, rely only on the DCF methodology. While there are several deficiencies in the stock and debt indicator, 7 they are not of recent recognition. The court believes the stock and debt indicator is as reliable as income indicators that involve estimates, forecasts, projections and subjective analyses, all involving human judgment. A good example is the wide disparity between the values arrived at by the experts for the two parties when they used the DCF indicator. 8

The stock and debt indicator’s great strength is that it is the only way in which an appraiser can give objective consideration to the prospective earnings of a corporation. If you accept the basic assumption that assets equal liabilities plus equity, you use specific figúres appearing in the company’s records and stock market reports of publicly traded companies. In the prior litigation the process was much more theoretical, subjective and complex. At that time BN’s holding company, BNI, also owned several other asset holding and income producing entities. As BNI held 100% of BN’s stock, BN’s stock was not publicly traded. To use the stock and debt indicator, it was necessary to determine how much BN’s stock contributed to the value of BNI’s stock.

In Bair III, 648 F.Supp. at 97-99, the court approved of a complex procedure designed to determine that part of the value of BNI’s stock attributable to the railroad. This procedure involved the deficiencies in the income indicator as well as those inherent in the stock and debt method. In Bair III both parties produced evidence to support their values using this technique. As a matter of fact, BN gave more weight to the stock and debt method than the Director.

None of these problems are present in current litigation because all of the other entities have been spun off and, consequently, BN is the only source of operating income remaining in BNI. Therefore the market price of BNI’s stock is primarily attributable to BN’s operations. The court has difficulty understanding BN’s vehement rejection of the stock and debt indicator.

With two exceptions, the Director arrived at the true market value of the operating portion of BN’s stock and debt by taking 81.5% of the market value of the sources of capital he listed. He computed the non-operating value to be 18.5% of the total. However, the Director did not reduce the market value of the common stock to 81.5% to arrive at a market value of the operating portion. His only explanation for not doing so was that the market value of the stock was so much lower than the value he computed using an equity rate of 13.33% that the market value “was not indicative of the value of the common stock” for BN, but that he “could not put a value higher than the market value or shares times the stock price.” The court cannot accept this unclear reasoning as a reason for not reducing the value of the common stock to 81.5%. The Director conceded that “the value of common stock would be operating and non-operating.”

There is no justification for treating the common stock differently from the other sources of capital. Although the stock was on the market at a “when issued” price, stock was sold at the reported prices. More likely the difference in valuations was an indication that the Director’s equity rate was not accurate, as the court has decided. The market value of the operating portion of BN’s common stock should be reduced from $1,752,-621,470 (100%) to $1,428,386,498 (81.5%), a reduction of $324,234,972.

The Director also left the operating leases at 100% of their value, or $623,263,805. However, as the value of operating leases are being totally removed because their value is picked up in the .market price for the common stock as previously determined (see section II Cl(d)), the court need not decide whether it should have been reduced to 81.5% Together these reductions total $947,-498,777. Therefore, the Director’s stock and debt indicator should be reduced to $3,957,-568,377. 9

E. Computation of True Market Value

In the court’s opinion, the valuation of a railroad by discounting future cash flow or capitalizing book income, while well intentioned, is highly theoretical and requires estimates, projections and assumptions that call for subjective judgment. The method has many weaknesses. As stated before, many things beside asset value influence the decisions of persons to lend money to or invest in railroad businesses. At least this method deals in real figures rather than projections, For that reason, I believe that the stock and debt and capitalized income approaches are equally good indicators and the court will give them equal weight in arriving at its own market value of BN as a unit.

Computation

Market value income indicator 3,748,181,390

Market value stock & debt indicator 3,957,568,777

Total' 7,705,750,167

divided by 2

True market value of BN 3,852,875,084 10

III.

TANGIBLE PERSONAL PROPERTY BREAKOUT

In Bair I an effort was made to determine the portion of BN’s property in Iowa that was personal property. The Director’s position at that time was that BN’s portion of its unit value allocated to Iowa must be considered as inseparable from its real property. Therefore only BN presented evidence as to what percent of BN’s Iowa property was personalty. BN’s witnesses used BN’s annual reports to the Interstate Commerce Commission (ICC) and segregated the accounts as real property or personal property. Dr. Schoenwold testified that his analysis was used to segregate railroad realty from personalty. Because the court recognized that it would be impossible to arrive at the exact percentage of the railroad’s unit that constituted personalty, it used a common sense approach and approximated the split to be 50-50.

The Director, using the method presented by BN in the prior litigation, determined that for 1989 only 39.6537% of BN’s Iowa unit value was personalty.

A BN employee offered oral testimony that the mix between real and personal property in BN’s total unit has not changed and argued that the Director’s use of the previously approved formula results in a percentage less than fifty percent because of changes in accounting procedures since the previous litigation. 11 Although it is doubtful that there has been an actual twenty percent change in the percentage of BN’s property that is personalty (50% down to 40%) in this short period of time, it is essential that some formula be presented that can be used in the future. Oral opinion testimony of a knowledgeable BN employee will not suffice.

BN’s rebuttal witness, Dr. John Davis, whose primary occupation is valuing public utilities' and railroads for ad valorem tax purposes, testified concerning the Director’s method of computing personal property on page 6 of plaintiffs Exhibit 61. Dr. Davis opined that the net book cost basis used by the Director is not a good method, but it is the best available and most frequently used method. 12

Although it is probably true that the change in percentage is due primarily to a change in bookkeeping, one cannot expect the Director to use any method better than the best available. That does not mean that this percentage of personal property is wrong — perhaps it was the fifty percent figure that was inaccurate. In any event, the court approves of the Director’s use of the net book cost basis, the best available methodology. He was correct in using 39.6537% as personal property.

BN also contends that ballast ($417,573 million), ties ($627,324 million) and track ($1,454,830 million) in place and part of the railroad system are in fact personal property and should be deducted from the unit value of the operating assets of the railroad. The court believes these items are properly classified as improvements to the land and are taxable as real estate. However, the court need not rely on this position because ballast, ties and rails owned by commercial and industrial taxpayers in Iowa are taxed as real estate. BN is not being subjected to discriminatory taxation by including these items as real property.

ÍV.

INTANGIBLE PERSONAL PROPERTY DEDUCTION

When a railroad is valued as a unit on a business enterprise or going concern basis, that valuation necessarily includes intangible assets. If a state exempts most personal property from ad valorem taxation, that exemption would include identifiable and quantillable intangibles. In this case BN has identified and valued three kinds of intangible personal property: computer software, assembled work force, and long term coal hauling contracts.

BN claims that section 306 prohibits Iowa from including these intangibles in its valuation for tax purposes because Iowa does not tax intangible personal property of commercial and industrial taxpayers. The Director argues that going concern intangibles are assessed in Iowa as real property because local assessors capture intangibles in the assessment of real estate. Therefore, BN has no claim under section 306(l)(d) “even if methodologies for centrally assessed properties differ from commercial and industrial methodologies.” The Director further argues that “the exclusion of financial intangibles from the [railroads’] unit value and the removal of personal property for railroads and the adjustments of the cost value for commercial and industrial real estate to account for sales price (by the county assessors) and sale ratios (by the Director) leave both parties similarly situated.”

Assuming for the purposes of argument that such approach would leave the parties similarly situated, which is questionable, the evidence does not show that adjustments and sales ratios pick up the value of commercial and industrial intangibles. Computer software, assembled work force and coal contracts have some value that remains in BN’s unit value and make some contribution toward the cash flow. The Director’s evidence that there are adjustments to the value of commercial and industrial real estate to account for the sales price to include such intangibles is not persuasive. The court holds that Iowa does not tax identifiable intangibles as real estate.

A. Taxation of Commercial and Industrial Intangibles

Section 441.21(2), Code of Iowa, provides that “the special value or use value of the property to its present owner, and the good will or value of a business which uses the property as distinguished from the value of the property as property” shall not be considered in determining market value.

The railroad has been valued as a business and the intangible personal properties involved here are components of that value. It appears that the foregoing statute and Iowa case law, Heritage Cablevision v. Board of Review, 457 N.W.2d 594, 598 (Iowa 1990), prohibit the assessor from considering such intangibles when appraising commercial and industrial property. Thus BN is being discriminated against unless commercial and industrial intangibles are being taxed in violation of the statute.

In Heritage, the Iowa Supreme Court approved of the following statement from the district court.

[W]hile the market approach method utilized by Kocer may have validity if he were attempting to determine the market value of the system, it does not follow that the method accurately determines the value of taxable assets. That is, the court concludes that the total fair market value of the system necessarily includes nontaxable assets such as a franchise to operate, an established customer base, experienced personnel in place, goodwill, and other intangibles.

Id. at 598.

The supreme court then said:

We believe the district court’s conclusion concerning these comparable sales was valid. The evidence suggests that a cable television system, as a going business enterprise, can generate tremendous income for the owners. We believe that the valuations which Kocer derived from these sales failed to exclude the substantial value which the buyers were receiving from the business enterprise with which the taxable assets were associated. Kocer attempted to determine that portion of the total price paid for other cable television systems as going concerns which was attributable to the taxable assets. He did so by taking the franchise fee payable to the governmental body, capitalizing that sum and subtracting the latter figure from the total sale price. We do not believe this adjustment produced a figure which is in any way representative of the fair and reasonable exchange between a willing buyer and willing seller of only the taxable assets.

Id. at 599.

The Director relies heavily on Maytag Co. v. Partridge, 210 N.W.2d 584, 590 (Iowa 1973). At that time section 428.22 of the Iowa Code provided that machinery used in manufacturing establishments shall for the purpose of taxation be regarded as real estate. The issues in Maytag were quite different. Maytag argued that machinery used in its operation should be valued and taxed at the market price of used machinery for sale. The Iowa Supreme Court rejected this claim and approved of the assessor recognizing the effect of the machinery’s use on the value of the property itself. Maytag' acknowledged that intangibles were not to be taxed. Id. at 590.

In Heritage the court said:

The application of the “going concern” principle in Maytag Co. was in the context of rejecting the property owner’s claim that its machinery should be valued exclusively by a market data analysis of the used machinery market. We approved the assessor’s depreciated cost appraisal in that case under the “other factors” approach. We did not suggest in Maytag Co. that in trying to value taxable assets under a comparable sales analysis the statutory directives against considering “special value or use value of the property to its present owner, and the goodwill or value of a business which uses the property as dis tínguished from the value of the property as property” may be disregarded.

Heritage, 457 N.W.2d at 599.

The Director introduced testimony from local assessors and state officials attempting to show that the Declarations of Value included intangibles in the purchase price of real estate. The forms, if properly completed, would not include intangibles because they ask the price for real property only. If an intangible is listed, its cost is deducted from the total price. Intangibles may inadvertently be included in the gross price if not identified with a price. However, that in the court’s opinion does not change Iowa law that intangibles are not to be taxed as real estate.

The court rejects the Director’s argument that intangibles are in fact taxed as. part of the real estate in Iowa. The Iowa tax system does not tax intangible property as such.

B. Value of Intangible Assets

BN’s experts identified intangible assets that contributed to BN’s value as a business enterprise. The cost approach was used to value BN’s computer software ($176,-000,000) and assembled work force ($233,-000,000). The income approach was used to value the coal contracts ($263,000,000). The experts deducted these determined values dollar for dollar from their computed $2.5 billion for the system as a whole. It is acceptable to use the cost approach to arrive at values for the computer software and the work force in place. The income approach is an appropriate way to value the coal contracts. But the dollar for dollar deductions from the unit value of the railroad is not the proper method to arrive at the value of BN’s real property. Use of this procedure would mean that the value of BN’s real estate would be whatever value is left after the computed values of tangible and intangible personal property are deducted from the unit value. For example, combining all of . the claims advanced by BN in this litigation you come tip with this very questionable result.

Unit value of the railroad $2,500,000,000

Less 50% as personal property . 1,250,000,000

1,250,000,000

Less value of intangibles 672.000.000

Value of real estate 578.000.000

This value for BN’s operating real property is ridiculous. It operates one of the largest railroad systems in the United States. BN’s system consists of 23,500 miles of operating track connecting the Midwest, the Pacific Northwest and the Gulf Coast. Under BN’s approach the value of all its operating real property would be considerably less than the value of the identified intangibles, although the book value of the railroad real property is $3,500,000,000. 13

The value of BN’s real estate cannot be determined by accepting the. remainder of the unit value of BN after the values of these intangibles is computed and deducted. If separate values had been determined for the real property as well as tangible and intangible personal properties and all three added together and the result proportioned, it would have made some sense. But to say the value of the real estate is just what is left, is unacceptable. There is no proof of the relationship between these values and the market value of the railroad whether the income or stock and debt indicators are used. There is no proof of the contribution these intangibles made to the cash flow or the price of the stock.’ The court cannot accept the proposition that every'dollar of their value went into the cash flow or stock price.

The court could therefore hold that BN has failed to establish by a preponderance of the evidence the contribution that these intangibles made toward the cash flow or the price of the stock of BN and deny it any relief from the state’s taxation of these identified intangibles. However, these intangibles do have some value that should be recognized and the court believes there is evidence in the record by which their contribu? tion to the cash flow and stock price can be considered, although there is no expert testimony supporting this analysis.

Why shouldn’t intangible personal property be broken out of the unit value the same way tangible personal property is? Book values were used to determine the.proportion of unit value represented by tangible personal property. BN’s expert conceded that this is not a good method, but it is the best available method of taking particular assets out of the unit value. See footnote 12. The court will use the book values of the BN railroad and add the computed values of the intangibles, which have no book value, determine the percentage applicable to the intangibles and allow that percent as a deduction from true market value. 14 The net book values are taken from page 7 of Exhibit 61. The proportion attributable to intangibles is ten percent and the allowable deduction is $672,000,000. 15

IV.

EQUALIZATION

The 4-R Act prohibits assessing “rail transportation property at a value that has a higher ratio to the true market value of the rail transportation property than the ratio that the assessed value of other commercial and industrial property in the same assessment jurisdiction has to the true market value of the other commercial and industrial property.” 49 U.S.C. § 11503(b)(1).

The court has attempted in the foregoing divisions to arrive at the true market value of the railroad as an operating unit and ■ the 100% value of its real property in Iowa subject to Iowa property tax. Iowa, by statute, also assesses commercial and industrial property at 100% of true market value. Iowa and five other states permit an assessment variance from true market value of five percent (95% to 105%) which is the strictest tolerance among the states and matches the five percent tolerance permitted by section 306.

BN claims that commercial and industrial property in Iowa is valued at less than ninety-five percent of its true market value. The Director claims that Iowa commercial and industrial property is valued well within the five percent variation permitted.

A. Assessment Jurisdiction

The 4-R Act defines an assessment jurisdiction as “a geographical area in a state used in determining the assessed value of property for ad valorem taxation.” 49 U.S.C. § 11503(a)(2). The question at issue here is whether the appropriate assessment jurisdiction for purposes of equalization is the State of Iowa as a whole or each individual county in which the railroad owns real property.

The Director’s equalization process operates on a county by county basis. The Director claims that discrimination against BN is correctly measured by the assessment ratios in the counties where BN has property weighted by the amount of BN property in the county. If this measure is proper, the appraisals are within the five percent tolerance. BN contends that the average level of assessment for commercial and industrial property in the state as a whole should be used for equalization purposes under section 306(l)(a) even though BN has real property in only twenty-eight of Iowa’s ninety-nine counties.

BN’s main line runs east and west across rural southern Iowa. Branch lines run into Polk and Dubuque counties. The distribution of BN’s value among the counties is in sharp contrast with the distribution of commercial and industrial values. The ten counties with the highest commercial property valuations account for more than sixty percent of the statewide commercial value but contain less than eight percent of BN’s Iowa value. Counties which have no BN value contain about fifty-five percent of Iowa’s commercial value.

The Director argues that “a statewide weighted assessment ratio is an inaccurate measure of alleged tax discrimination against [BN]. The statewide ratio will be determined primarily by the results of the ratios from the larger counties and the counties which have no [BN] value. These ratios will not affect [BN’s] ad valorem taxes because [BN] has little or no property in these counties and hence has little or no tax at stake.”

Section 306 is ambiguous. It could be interpreted to permit the use of the commercial and industrial properties in only those counties where BN has real estate. However, all courts addressing this issue have concluded that where railroads are centrally as-, sessed by the state, the assessment jurisdiction within the meaning of section 306 is the entire state. Arizona v. Atchison, Topeka & Santa Fe R.R., 656 F.2d 398, 405 (9th Cir.1981); Southern Ry. v. State Bd. of Equalization, 712 F.Supp. 1557, 1564-65 (N.D.Ga.1988); Atchison, Topeka & Santa Fe Ry. v. Lennen, 552 F.Supp. 1031, 1035 n. 7 (D.Kan.1982), aff 'd in part, rev’d in part, 732 F.2d 1495 (10th Cir.1984).

Although it is true that the statewide ratios of assessments to true market value will not directly affect BN’s millage rate, which is set locally, the amount of taxes paid is also dependent upon the value placed on BN’s property, which is centrally appraised. Iowa makes a strong effort to equalize the locally made assessments so that the appraisals in all counties approach uniformity. This is important because some state aid is distributed on the basis of property valuations. The assessment ratios should be about the same throughout the state. A railroad should not be benefitted or penalized if a local assessment, inadvertently, is not equalized. A comparison of the assessment ratio for commercial and industrial property throughout the state with BN’s true market value would tend to be more reliable because of the greater number of parcels involved. In addition, all the railroads that run through Iowa would be subject to the same assessment ratio for commercial and industrial property, thus eliminating any competitive advantage or disadvantage which might result from possible uneven assessments in different counties.

The. court, therefore, holds that the correct assessment jurisdiction for the comparison of the ratio of assessments to true market value of commercial and industrial property is the State of Iowa. The court will next consider whether the statewide assessment ratio for commercial and industrial property in Iowa is within the 95% to 105% range of true market value.

B. Commercial/Industrial Assessment Ratio

After hearing the expert witnesses testify, reading and re-reading their testimony,.and closely examining their sales ratio studies, the court finds that BN has failed to establish by a preponderance of the evidence that the non-railroad commercial and industrial property in the State of Iowa is appraised at less than ninety-five percent of its true market value. In fact, the court is persuaded that the commercial and industrial property in Iowa is appraised within five percent of its true market value.

Experts from both sides utilized the exact same sales ratio data accumulated by the Director from Declarations of Value that Iowa law requires to be filed before a real estate deed may be filed of record. The Director, following his regularly established procedure designed to screen out abnormal or non-arms-length transactions, furnished data on 796 “normal” deed sales of commercial property and thirty-three sales of industrial property.

Dr. Richards, BN’s expert, combined commercial and industrial sales during 1989 and worked from a total of 829 sales. The Director’s experts, Gloudemans and Dornfest, used commercial sales only. The omission of the industrial sales, which the Director’s experts believed should be treated separately, did not hurt BN because the assessment ratio for industrial properties was higher than for commercial.

Dr. Richards eliminated from his ratio study sixty-two sales in which the assessment ratio exceeded 200%, deciding “that these were most likely not truly arms length transactions and, therefore, should have been deleted by the State of Iowa.” Dr. Richards did not delete any sales at the low end of the assessment ratios. He eliminated sales where the appraisals were greater than double the sales price, but left in thirty-three sales where the appraisal was less then half of the sales price. This- was contrary to International Association of Assessing Officers (IAAO) standard 6.6 which limits trimming to five percent and requires equal trimming from the extreme ratios on both sides of the median. Gloudemans and Dornfest, the Director’s experts, eliminated only four “outlier” sales; two because the assessment ratio was fourteen times the sale price when the next highest ratios were 7.5, and two because the values- were so much greater than the rest of the sales that they would receive an inordinate amount of weight in the computations. The court believes that the elimination of these four outlier sales was more appropriate than the blanket elimination of all sales of Iowa commercial and industrial real estate where the ratios exceeded two times or 200%. . The Director had screened these sales and eliminated sales he considered abnormal.' He determined the sixty-two sales of commercial and industrial real estate were not abnormal.

Dr. Richards stratified the counties according to the number of commercial industrial properties they contained. There were eighty-three counties in Strata 1 containing less than 1000 parcels; seven counties in Strata 2 containing 1000 to 2000 parcels; seven counties in Strata 3 containing 2000 to 3500 parcels; and two counties in Strata 4 containing over 3500 parcels. Although stratification can add efficiency to a sales ratio study, the stratification by number of commercial and industrial properties in the county has little, if any, relationship to assessment ratios. Stratification by number of sales or value of Sales would have been more meaningful because there was evidence that samples’ accuracy increases as the number of sales increases, and that there is a tendency to over appraise properties of lesser value.

Dr. Richards used the weighted average mean to arrive at his level of assessment of commercial and industrial property. Excluding public utilities, he arrived at an assessment ratio of 89.34% of the true market value of commercial and industrial property. When he included non-rail public utilities the ratio rose to 90.07%. This testimony supported BN’s claim that Iowa commercial and industrial property was assessed at less than ninety-five percent of their true market value.

Gloudemans, the Director’s expert, made several different calculations using both the median and weighted mean. Gloudemans’ calculations all exceeded ninety-five percent of true market value except for the calculation on the weighted mean of 92.82% of the market value when the four “outliers” were left in.

The use of median or weighted mean each have advantages and disadvantages. In this particular situation the court felt the use of either or both was acceptable.

The prior discussion deals with the important differences between the experts’ studies. Further analysis of their opinions would not be helpful. The experts were well qualified and presented sound evidence to support their positions. The court has expressed the problems it had with Dr. Richards’ analysis. BN’s evidence relating to assessment ratio is, at best, no more persuasive than the Director’s.

As stated earlier, the court finds that under this record, giving due regard for the Director’s expertise, the Director has assessed commercial and industrial property at least ninety-five percent of its market value. In any event BN has failed to sustain its burden to prove otherwise.

VI

SUMMARY

In summary, I find and hold that:

1. Capitalization of income and stock and debt approaches to valuation are “methods” and not the “applications” of methods. The court cannot require the Director to use any particular method to value railroads as business enterprises.

2. The court will not use the DCF method to value BN as that would, in effect, require the Director to use that method. The DCF method is no more reliable than other accepted methods of appraisal.

3. The Director did not err in using the income from the latest year as NROI.

4. The change in the method of accounting did not make reliance on book income less suitable.

5. The Director acted properly in adding deferred income taxes to the NROI.

6. The Director erred in adding operating leased equipment to the NROI because their value is captured in the cash flow and stock price.

7. The rate of return to debtholders to be used in computing the appropriate capitalization rate is 6.24%.

8. The rate of return to shareholders to be used in computing the appropriate capitalization rate is eighteen percent.

9. The proportion of debt capital to equity capital to be used in computing the appropriate capitalization rate is forty percent debt and sixty percent equity.

10. The correct capitalization rate is 13.-3%.

11. The true market value of the BN as a business enterprize using the income capitalization indicator is $3,748,181,390.

12. The true market value of the BN as a business enterprize using the stock and debt method indicator is $3,957,568,377.

13. The income capitalization and stock and debt indicators are entitled to equal weight, and the true market value of BN as a business enterprise is $3,852,875,084.

14. The court approves of the Director’s use of net book cost basis to determine the tangible personal property breakout.

15. Ballast, ties and rails in place as part of the system are taxable as real property.

16. Computer software, assembled work force and long term coal contracts are intangible assets that contribute to the value of BN.

17. Iowa does not tax intangible property as part of commercial and industrial real estate. Therefore, whatever value the railroad’s intangibles contribute to BN’s value as a business enterprise should not be taxed.

18. The appraised value of the intangibles cannot be deducted dollar for dollar from the true market value of BN as a business enterprise.

19. Use of the net book cost basis to determine the proportion of BN’s true market value attributable to these intangible assets results in ten percent of their appraised value or $67,200,000.

20. The correct assessment jurisdiction for equalization purposes is the whole State of Iowa.

21. The Director has assessed commercial and industrial property at more than ninety-five percent of its value. In any event BN has failed to prove otherwise.

The court has, with trepidation, made the foregoing calculations. However, the remaining calculations will be left to the expertise of the parties using the court’s findings and holdings.

I cannot conclude this opinion without offering some gratuitous advice to the railroads, the states and the federal government. The 4-R Act was enacted to protect the railroads from discriminatory taxation by the states, a valid concern. However, the statute, as it now stands, has spawned considerable and very expensive litigation. The issues of national significance should be resolved on a national level. Different states should not have different unit values for the same interstate railroad. The percentage of that value consisting of tangible and intangible personal property for the specific railroad should be the same throughout the country. The statute should be amended to give the Interstate Commerce Commission, or other appropriate board or commission the authority and duty to value all interstate railroads and determine the percentages of that value attributable to tangible and intangible personal property. The states and the railroad should have input during the evaluation procedure and the right to appeal. But these issues should be resolved in one lawsuit between the interested states and the railroad. Those issues relating to taxation in a specific state, like equalization, could still be tried in federal court in the state involved.

The changes suggested should save the railroads and the states considerable sums of money, treat all states and railroads equally and reduce the burden on federal courts.

The court directs BN to take the responsibility for the preparation of Order for Judgment in accordance with this Ruling and Order. It shall be submitted to the Director for his approval and then submitted to the court. If there is disagreement over the contents of such an order, the matter will be resolved by the court.

IT IS SO ORDERED.

1 . The Act, 49 U.S.C. § 11503, provides in part:

(a) In this section—

(1) "assessment” means valuation for a property tax levied by a taxing district.

(2) "assessment jurisdiction" means a geographical area in a State used in determining the assessed value of property for ad valorem taxation.

(3) "rail transportation property” means property, as defined by the Interstate Commerce Commission, owned or used by a rail carrier providing transportation subject to the jurisdiction of the Commission under subchapter I of chapter 105 of this title.

(4) "commercial and industrial property” means property, other than transportation property and land used primarily for agricultural purposes or timber growing, devoted to a commercial or industrial use and subject to a property tax levy.

(b) The following acts unreasonably burden and discriminate against interstate commerce, and a State, subdivision of a State, or authority acting for a State or subdivision of a State may not do any of them:

(1)assess rail transportation property at a value that has a higher ratio to the true market value of the rail transportation property than the ratio that the assessed value of other commercial and industrial property in the same assessment jurisdiction has to the true market value of the other commercial and industrial property.

(2) levy or collect a tax on an assessment that may not be made under clause (1) of this subsection.

(3) levy or collect an ad valorem property tax on rail transportation property at a tax rate that exceeds the tax rate applicable to commercial and industrial property in the same assessment jurisdiction.

(4) impose another tax that discriminates against a rail carrier providing transportation subject to the jurisdiction of the Commission under subchapter I of chapter 105 of this title.

(c)Notwithstanding section 1341 of title 28 and without regard to the amount in controversy or citizenship of the parties, a district court of the United States has jurisdiction, concurrent with other jurisdiction of courts of the United States and the States, to prevent a violation of subsection (b) of this section. Relief may be granted under this subsection only if the ratio of assessed value to true market value of rail transportation property exceeds by at least 5 percent, the ratio of assessed value to true market value of other commercial and industrial property in the same assessment jurisdiction. The burden of proof in determining assessed value and true market value is governed by State law.

2 . Burlington N. R.R. v. Bair, 584 F.Supp. 1229 (S.D.Iowa 1984) (hereafter Bair I), aff'd in part, 766 F.2d 1222 (8th Cir.1985) (hereafter Bair II), on remand, 648 F.Supp. 91 (S.D.Iowa 1986) (hereafter Bair III).

3 . The cases cited by Chesapeake in its Petition for Certiorari on this point at p. 13 do not hold that the federal courts can require the states to use a specific methodology in valuing railroad property. CSX Transp., Inc. v. Forst, 777 F.Supp. 435, 442-43 (E.D.Va.1991), is to the contrary. The arguments made by BN here were included in Chesapeake's petition and certiorari was denied, — U.S.-, 112 S.Ct. 1577, 118 L.Ed.2d 220 (1992).

4 . There are deficiencies in all methods of appraisal, including DCF. DCF is a theoretically sound method of appraisal, but involves more assumption and estimates than other methods. It is difficult to make long range forecasts. DCF requires much research and is complex. It is doubtful DCF would be practicable for the state to employ to value all railroads in its jurisdiction every two years.

5 . Rcc = d(Rd) + e(Re)

where: Rcc = Weighted average cost of capital

d = Proportion of debt capital to total capital

e = Proportion of equity capital to total capital

Rd = After-tax cost of debt

Re = Cost of equity

6 . Re = Rf + B(Rm - Rf)

where: Re = Cost of equity capital

Rf = Risk-free rate of return

Rm = Market rate of return

(Rm - Rf) = Market Risk premium

B = Beta (the volatility of the subject’s equity compared to a market index such as the S & P 500)

7 . Stock prices are affected by influences having nothing to do with the value of a company’s underlying assets. Speculation on market trends, varying motivations, business booms and busts, tax treatment and other considerations tend to give an unrealistic picture of the value of a company measured by the stock and debt indicator.

8 . Coopers and Lybrand arrived at a true market value for BN as a unit of $2.5 billion. Michael W. Goodwin and Associates concluded there was only a 10% chance that the value was less then $4.8 billion and a 20% chance that the value would exceed $6.4 billion. The 2.5 billion figure is not realistic because the debt portion of the capital structure was about the same figure and common stock was worth $2.6 billion on the market.

9 . BN’s value under stock and debt approach as shown on plaintiff’s Exhibit 61 $4,905,067,154

Less non-operating value of common stock and value of operating leases 947,498,777

Corrected market value of operating portion $3,957,568,377

10 . The court gains some comfort from the fact that this determination of fair market value is well within the range determined by seven other midwestern states.

11 . The ICC required the railroads to change RRB accounting to depreciation accounting.

12 . Davis testified:

The' methodology used on this last page of plaintiff’s Exhibit 61 is probably the most widely-used method, Your Honor, of allocating any type or any piece of any part of a system value out. For instance, sometimes you’re not trying to allocate out personalty. You may be trying to allocate out realty. You may be trying to allocate out property of a certain state.

' When one goes about allocating something out of a unit value, I think it is clearly understood that we don’t have a good way of doing that. There is no market-driven way that I know of to do that, and so we try to come up with the best substitute.

Around the country, based on my experience, the methodology that is most frequently used is a net book cost method.

As long as you are consistent in calculating your numerator and your denominator, that is, when you look at the system property, as I understood it, the items mentioned there, road and equipment, operating leases, materials and supplies, those are all of the items that are documented that can be found that are part of the system and it is their net book cost.

Below that, under personal property, are those different items, fences, communication systems, signals and so forth, that are considered to be personal property according to this form, and they are listed with their net book cost, so we’re apples to apples. We’re saying the best data we have is to take — put everything on the same basis, net book cost, and simply say, artificially now we will allocate a portion of this system value based on a net book cost allocation method, to allocate out.

******

Some of the confusion, Your Honor, and this is, it’s a hard thing to say as an appraiser, is we try to work as much as we possibly can off of market factor to come up with a system value, and then we have to say, okay, now that we have a system value, how do we allocate a piece of that out.

For instance in a lot of the public utilities that I value, the property is multistate, and so we want to know what state X’s value is out of the total. We generally allocate on a net book cost basis, because we don’t have a better method. We’d love to have a better method, just don’t have one.

13 . From page 7, Exhibit 61—

Net book value of the system $5,798,249,686

Less personal property and operating leases 2,299,220,686

$3,499,029,000

14 . I realize that I am off on a "frolic of my own” in taking this approach but I am convinced it is the best result. I also realize I am combining book value with appraisals of true market value, but I could not accept the dollar for dollar deduction but I know the intangibles contributed to BN’s value.

I also recognize that the valuation of intangibles will pose a serious problem for the Director. The expense of arriving at a value for these assets that have no intrinsic value will be great. But, I do not see any way to avoid acknowledging that they have a value that should not be taxed as part of the real estate.

15 . Net book value $5,798,249,686

Add value of intangibles 672,000,000

Total value $6,470,249,686

Value of intangibles 672,000,000

Total value = 6,470,000,000 = 10.3%

Round to 10%