Clinchfield R. Co. v. Lynch, 605 F. Supp. 1005 (E.D.N.C. 1985)
March 22, 1985
Mark G. LYNCH, Secretary of Revenue of the State of North Carolina; Douglas R. Holbrook, Director, Ad Valorem Tax Division of the North Carolina Department of Revenue; Catawba County; Forsyth County; Granville County; Iredell County; Robeson County; Rowan County; Rutherford County; Vance County; Wilson County; Beaufort County; Bertie County; Bladen County; Buncombe County; Burke County; Caswell County; Chatham County; Chowan County; Craven County; Davie County; Duplin County; Edgecombe County; Franklin County; Gates County; Henderson County; Jackson County; Jones County; Lincoln County; Martin County; McDowell County; Mitchell County; Moore County; Nash County; New Hanover County; Onslow County; Pasquotank County; Pender County; Perquimans County; Polk County; Randolph County; Rockingham County; Sampson County; Scotland County; Stanly County; Transylvania County; Warren County; Wayne County; Wilkes County; and Yancey County, Defendants.
United States District Court, E.D. North Carolina, Raleigh Division.
*1006 *1007 Armistead J. Maupin and Charles B. Neely, Jr., Raleigh, N.C., Courtney L. George, Jacksonville, Fla., Everett B. Gibson and Gregory G. Fletcher, Memphis, Tenn., William C. Antoine and James W. McBride, Washington, D.C., Edward C. Winslow, III, Greensboro, N.C., for plaintiff.
George W. Boylan, Raleigh, N.C., Hamlin L. Wade, Charlotte, N.C., and county attys. for each North Carolina county, for defendants.
MEMORANDUM OF DECISION
This action is the second in a series of suits brought by plaintiff railroads, alleging *1008 discriminatory taxation of their real and personal property in North Carolina for the 1981 tax year in violation of Section 306 of the Railroad Revitalization and Regulatory Reform Act (the 4-R Act), 49 U.S.C. § 11503. The original defendants in this action were Mark G. Lynch, Secretary of Revenue of the State of North Carolina, and Douglas R. Holbrook, Director of the Ad Valorem Tax Division of the North Carolina Department of Revenue. Of the eighty-five counties in North Carolina which contain railroad property and against which plaintiffs sought relief, twenty-four were allowed to intervene as defendants without objection by plaintiffs. Fifteen of those counties subsequently withdrew from the action, leaving nine intervenor counties: Catawba, Forsyth, Graville, Iredell, Robeson, Rowan, Rutherford, Vance and Wilson.
Plaintiffs contend that all of these counties assess and tax both real and personal commercial and industrial property at less than 100 per cent of actual market value, thus discriminating against the railroads, whose property by stipulation is assessed at 100 per cent of market value. The action came to trial before the court on November 27-29, 1984 and January 7-8, 1985. Having received and reviewed the parties' post-trial filings and the trial transcript, the court now incorporates into this memorandum of decision its findings of fact and conclusions of law. Rule 52(a), F.R.Civ.P.
BACKGROUND OF THE CASE
The facts of this case cannot properly be reviewed without first discussing the prior action brought by the same plaintiffs challenging defendants' assessment practices for the 1980 tax year. In Clinchfield Railroad Company v. Lynch, 527 F. Supp. 784 (E.D.N.C.1981) (Clinchfield I), this court found that plaintiffs successfully met their prima facie case of de facto discrimination against all railroad property by virtue of their sales assessment ratio study which "established the ratio of assessment of locally-assessed commercial and industrial real estate to the true market value of such real estate for the 1980 tax year," id. at 786. This court accordingly entered an order enjoining defendants from assessing plaintiffs' property at ratios exceeding those set out in plaintiffs' sales assessment ratio study.
On appeal, the Fourth Circuit affirmed this court's decision, but on slightly different grounds. Clinchfield Railroad Company v. Lynch, 700 F.2d 126 (4th Cir. 1983). The court held that plaintiffs did make out their prima facie case of discrimination for all property through their sales assessment ratio study, but that defendants could rebut this presumption with respect to personal property by presenting evidence of higher ratios of assessment for personal property and by showing the breakdown by percentage of the respective worths of the railroads' real and personal property.
We are satisfied that once North Carolina was shown to have practiced discrimination with regard to real property under its statute which applies a single undifferentiated assessment to both real property and personal property, the state assumed the burden of establishing facts sufficient both to warrant a different conclusion with respect to personal property and to enable the district court to fashion a decree that would not frustrate efforts to alleviate the discrimination already proven as to real property. (Footnote omitted.)
Id. at 131.
Based on this analysis, the court found that even if it was assumed that defendants' *1009 statistics showed ratios of assessment for personal property at or near 100 per cent, defendants had failed to present any evidence regarding the breakdown of railroad real and personal property. Thus, it concluded, defendants failed to afford the district court the means to fashion a decree which could properly alleviate the discrimination already proven by plaintiffs.
In the present action the parties, having the benefit of the Court of Appeals' opinion in Clinchfield I, have simplified the issues to be presented before this court. They have stipulated to the 1981 levels of assessment of locally assessed commercial and industrial real property in relation to the market value of such property. Those assessment ratios are attached to this opinion as Appendix A. Stipulations filed September 14, 1982, ¶ 8, and Final Pre-Trial Order, Stipulation 6. They have stipulated that all centrally-assessed real and personal property, including rail transportation property, is assessed on a county-by-county basis at 100 per cent of actual market value. Stipulations filed September 14, 1982, ¶ 5, and Final Pre-Trial Order, Stipulations 3 and 9. They have agreed to postpone the question of the breakdown of railroad real and personal property until a later trial, if such is necessary. Thus, the sole question before this court is at what rate is locally-assessed commercial and industrial personal property valued. The burden remains on defendants to show by a preponderance of the evidence that that rate should be different than the ratios stipulated for real property. Clinchfield I, 700 F.2d at 131-134. As the thirty-nine counties listed in Footnote 1 of this opinion presented no evidence regarding locally-assessed business personal property, plaintiffs are entitled to judgment as a matter of law with respect to those counties. However, with respect to the nine remaining counties which did present evidence, the court must review that evidence on a county-by-county basis.
Plaintiffs chose not to rely solely on the stipulated ratios for real property, but also presented evidence regarding the assessment of personal property statewide and in the individual intervenor counties. They presented their evidence through three basic sources: Dr. Dick Netzer, an economist specializing in property taxation; Douglas R. Holbrook, Director of the Ad Valorem Tax Division of the North Carolina Department of Revenue; and through cross-examination of the county tax officials who testified. Dr. Netzer was their first witness. He testified that except for real property taxes, virtually all taxes are self-reporting. Furthermore, under any self-reporting tax system, the taxpayer will be inclined to underreport income and undervalue property. With real property, there is no such problem, since real property can be mapped and easily identified. Furthermore, since all real property transactions are publicly recorded, there is no difficulty in ascertaining the ownership of a particular piece of property. By contrast, personal property transactions are generally private and there is no uniform system to insure that each item is listed on the individual's property tax return. Volume I, pp. 41-43, 48.
As a result of the problems with self-reporting personal property, verification procedures must be developed to insure that the property is accounted for and valued. Dr. Netzer suggested several verification procedures for personal property taxation, including internal checks of the tax forms themselves, matching the returns against other evidence, such as income tax returns, and selective on-site auditing of taxpayers to check for the presence of assets and to compare the returns with the taxpayers' books and records. He stated that on-site audits are particularly important, because that method is frequently the only way to discover assets not reported on tax returns. The problem, he said, is not that there is wholesale cheating, but that people make judgment calls. As a result, the taxpayer *1010 gives himself the benefit of the doubt when listing the nature and value of his property. For example, a taxpayer may be permitted by the Internal Revenue Service to expend certain purchases of equipment, rather than capitalize them. Nevertheless, that equipment should be listed on the taxpayer's property tax return for ad valorem tax purposes. The taxpayer who fails to list this property may legitimately feel that it need not be included, but absent an on-site audit, that property might never be listed and taxed by the assessing authority. Volume I, pp. 43-49.
There also are problems with valuing particular types of property. For example, it is difficulty to assess a value for machinery and equipment because there frequently is no second-hand market for such equipment. There is also an enormous variety in types of equipment. The standard method of valuing machinery and equipment has been to take the taxpayer's purchase price and then resort to depreciation schedules to determine its value given its age. It is also important to consider such factors as inflation, which will tend to raise the value of equipment. Although this method results, at best, in a rough approximation of true value, Dr. Netzer admitted that it is the only available realistic means of valuing machinery and equipment. Volume I, pp. 57-60.
Dr. Netzer pointed to inventory as a particularly troublesome area with respect to valuation. The problem, he said, is that inventory is not all traded goods. Instead, it frequently consists of one-third raw materials, one-third goods in process, and one-third finished goods. There is no market for goods in process and there consequently is no valid way to determine its value. Dr. Netzer concluded that of the most common methods of valuing inventory, that is, LIFO, FIFO and lower of cost or market, the FIFO method would most nearly approximate market value. Volume I, pp. 61-65.
The balance of Dr. Netzer's remaining testimony involved a study that he was asked to perform by the plaintiffs. Plaintiffs sought to determine whether a method could be developed to compare the listed values of personal property in each county in North Carolina with their actual market values. Dr. Netzer testified that during the preparation of his 1966 book Economics of the Property Tax, he began developing such a method. Volume I, p. 77. The key to this study was to find a way to estimate actual market value for all personal property in each county. Dr. Netzer computed this estimate by gathering data published by the United States Department of Commerce regarding the total value nationwide of machinery and equipment and of inventory in each of thirty-seven industrial categories. Volume I, pp. 82-90. He then determined the total number of employees nationwide in each category, and determined the value-per-employee of machinery and equipment and of inventory in each county by dividing the aggregate values by the total employees nationwide. Volume I, p. 84. This figure was multiplied by the number of employees in each industry within each of the nine intervenor counties, resulting in an estimated aggregate value of each industry in each county. Volume I, p. 84. The aggregate values in each county were added up, and the results were compared with the actual assessed values of business personal property in each county. The results showed the following *1011 ratios of locally-assessed tax value to total estimated value in each of the nine counties (Plaintiffs' Exhibit No. 4, Table 7):
Catawba - 67.5% Forsyth - 45.6% Granville - 48.2% Iredell - 66.9% Robeson - 42.8% Rowan - 70.9% Rutherford - 61.8% Vance - 51.3% Wilson - 59.6%
As can be seen by comparing these ratios with those on Appendix A, only Rowan County had an estimated level of assessment for personal property higher than the stipulated level of assessment of business real property.
On cross-examination, though, defendants pointed to a problem with Dr. Netzer's study involving his valuation of inventory. In the Bureau of Census figures, all inventory was valued as of December 31, 1980. However, under the North Carolina ad valorem tax laws, taxpayers may report their inventory as of the close of their fiscal year, which may not be the end of each calendar year. Dr. Netzer testified that taxpayers generally have their lowest inventory value at the end of their fiscal year. Thus, Dr. Netzer's estimated inventory values would have been greater than the actual value of inventory in the nine counties. Seven of the counties (Granville, Iredell, Robeson, Rowan, Rutherford, Vance and Wilson) later produced evidence showing that of their top twenty-five taxpayers (who accounted for over fifty per cent of the total business personal property tax base in each county), the majority operated on a fiscal year basis other than January 1December 31. Thus, Dr. Netzer's valuations of inventory may have been overstated by a sizable percentage. Volume I, pp. 151-161.
These and other problems make it clear that Dr. Netzer's ratios do not accurately reflect the ratios of locally-assessed tax value to total actual fair market value of business personal property in each county. Nevertheless, the methodology in Dr. Netzer's report was basically sound, and the report accordingly does provide some evidence that business personal property in the nine intervenor counties is not assessed at 100 per cent of fair market value.
Plaintiffs' second witness, Douglas R. Holbrook, Director of the Ad Valorem Tax Division, testified about three areas of property assessment which plaintiffs contend show that personal property is not assessed at fair market value. Those areas involve (1) verification procedures, (2) inventory valuation, and (3) machinery and equipment valuation.
With respect to verification procedures, Mr. Holbrook testified that such procedures are necessary to insure that taxpayers do not accidentally or intentionally fail to report taxable property. Volume II, p. 155. The traditional tool for verifying property tax returns in North Carolina has been to compare those returns with the taxpayer's returns from the previous year (called a desk audit). Using this procedure, tax officials can determine whether there have been any major changes in the valuation of a taxpayer's property which should be further checked. The disadvantage to the system, though, is that it would not discover property purchased within the preceding year but not reported on the latest return. Another tool for auditing a taxpayer's property tax return is to compare it with his income tax return. Volume II, pp. 55-56. This system has resulted in a significant *1012 amount of discoveries when employed on a systematic basis, and the Department of Revenue accordingly has recommended its use by the county tax offices. Volume II, pp. 55-56.
However, there are deficiencies with this method of verification, as well. First, inventories on income and property tax forms may be different, since the LIFO method of accounting inventories is permissible for income tax purposes but not for property tax purposes. Instead, FIFO is required. Volume II, p. 45. As many taxpayers do not note on their income tax returns what accounting method they use, tax officials may be forced to contact the taxpayer to determine whether LIFO or FIFO is used. Second, a review of income tax returns will not aid in discoveries if the property is reported on neither return. And third, most multi-county taxpayers do not break down their property county by county on their income tax returns. Therefore, the income tax returns of those taxpayers are generally useless for verification purposes. Volume VI, p. 100.
Mr. Holbrook testified that the best verification method is through the use of field audits, where both the taxpayer's physical plant and his records are reviewed. In a well-run assessment program, twenty per cent of the business taxpayers should be audited each year, with field audits being the preferable method. In the counties where a systematic audit program has been established, significant amounts of unreported property have been discovered. Volume II, pp. 62-64. It should be noted that none of the counties involved in this action had such a program in 1981.
Plaintiffs' second area of contention involves inventory valuation. There are three basic methods to account for inventory: (1) last-in, first-out (LIFO), (2) lower of cost or market, and (3) first-in, first-out (FIFO). The Department of Revenue requires, for ad valorem tax purposes, that inventories be valued at 100 per cent of cost using the FIFO method of accounting. Volume II, p. 32. LIFO and lower of cost or market are not permitted, since in times of inflation they tend to understate market value. As a result, it is important for tax assessors to adopt verification procedures to insure that inventory is not reported based on LIFO values. Volume II, p. 36. Such procedures should include a specific prohibition against LIFO on the property tax forms and instructions. Plaintiffs contend, though, that the intervening counties have not adopted sufficient procedures to guard against the use of LIFO in valuing inventories on property tax returns. Their specific contentions will be discussed later when the court reviews the evidence of the individual counties.
Plaintiffs' third area of contention relates to the valuation of machinery and equipment. Mr. Holbrook testified that the Department of Revenue has recommended the use of trending manuals to determine the value of machinery and equipment, because these manuals take both depreciation and inflation into account. Volume II, p. 147. Nevertheless, a number of counties did not use the trending manuals in 1981, but simply depreciated the equipment. As a result, in periods of high inflation such as was the case in the late 1970's and early 1980's, machinery and equipment in those counties would have been substantially undervalued. Volume II, pp. 51-52.
Plaintiffs assert that these facts rebut defendants' contention that the business personal property in the nine intervenor counties is assessed at or near 100 per cent of fair market value. However, a review of each county's evidence is necessary, as assessment and verification methods vary widely among the counties.
1. Rowan County
Appearing for Rowan County was Wallace Peeler, its tax supervisor. He testified *1013 that Rowan has a population of approximately 100,000 people, and consists of one city and nine towns. The county is predominantly rural with textile manufacturing being its principal industry. In 1981, the county listed 3,996 business personal property taxpayers. Mr. Peeler's office consisted of eleven full-time employees, including two business personal property appraisers. Volume II, pp. 208-212.
Mr. Peeler generally described the methods by which his office goes about locating and assessing business personal property in Rowan County. Those methods include: (1) checking the yellow pages and the city directory for new businesses opened within the previous year; (2) checking the Department of Motor Vehicles list to insure that every vehicle in the county is listed; (3) cross-checking with taxpayers' income tax returns; (4) cross-checking with the list of mobile homes from each mobile home park; (5) reviewing the warehouse reports of property stored; (6) reviewing each airport's list of planes stored as of the first of each year; (7) reviewing taxpayers' returns from the previous year for consistency (desk audit); (8) driving through the county to find new businesses (called a windshield audit); (9) checking out vehicles claimed to be listed in another county by contacting the tax assessor from that county; and (10) checking leased equipment (which must be included on the lessee's property tax form) to see if the lessor had listed it. Volume III, pp. 14-15.
The value of all business personal property, except farm machinery, is determined by Mr. Peeler's office. Extensive sets of form letters and returns have also been developed to insure that each taxpayer knows his responsibilities regarding listing business personal property. An automatic ten per cent penalty is also imposed on any taxpayer who is caught not listing or underlisting his property. Volume II, p. 218; Volume III, pp. 11, 34-63, 85-86; Rowan Exhibits 1-33.
With regard to the county's verification procedures, however, Mr. Peeler's office does not review income tax returns on a systematic basis. Volume III, pp. 20-21. They also do not use tax returns to determine the method of accounting for inventories. Volume III, p. 71. Therefore, if a taxpayer reports inventory on his income tax return using the LIFO method of valuation, that return would not reflect the actual value of a taxpayer's inventory for ad valorem tax purposes. Furthermore, the county's 1981 property tax form did not contain language prohibiting the use of LIFO values, as has since become the practice on all county forms. Volume III, pp. 71-72. As a result, taxpayers may end up listing their inventories on their property tax forms using the LIFO method, thus understating their value.
Also, as Mr. Holbrook testified, the most accurate verification procedure is an indepth on-site audit of the taxpayer's property, books and records. However, Rowan County has never employed this procedure, and Mr. Peeler admitted that there is no one in his office who is even qualified to conduct such an audit, even though there are seventy or more business taxpayers in the county who list more than $1,000,000 in personal property. Volume III, pp. 75-76.
Finally, Rowan County's practice of permitting farmers to value their own equipment most likely results in undervaluation of that property. As virtually every witness admitted, the taxpayer will almost always give himself the benefit of the doubt when listing or valuing his property. Although the value of farm machinery may not constitute a great percentage of total business personal property, its undervaluation will reduce the total tax base in the county. For these and the other reasons discussed hereinbefore, the court concludes that Rowan County has failed to sustain its burden of proving that business personal property assessed by the county is assessed at 100 per cent of actual fair market value.
2. Forsyth County
The second county to present evidence was Forsyth County. It was the most populous and most urban of the counties *1014 presenting evidence, and consequently has the largest and most sophisticated property tax assessment office. Testifying for the county were Harvey Pardue, its tax supervisor, and Jackson Cumby, who audits business personal property. These gentlemen testified that Forsyth generally follows the same procedures as Rowan County to verify the accuracy of their business personal property tax returns. Forsyth Exhibits 1-41. However, Forsyth County also (1) sues out warrants when taxpayers fail to list; (2) checks UCC filings recorded in the Register of Deeds; and (3) frequently assesses property at a rate higher than taxpayers would like, forcing the taxpayers to appeal to the Board of Equalization and Review and to the state appellate courts. Volume III, pp. 112, 116, 121-126. Forsyth County also included in 1981 a prohibition on its tax form against using LIFO to value inventory. Volume III, p. 163.
Since 1982, Forsyth County has begun a program of conducting periodic audits of all business taxpayers. However, no such program was in effect in 1981. Of the field audits that were conducted in 1979 through 1981, no discoveries resulted. As the later periodic program did result in a number of discoveries, the court must assume that the institution of such a program in 1981 would have resulted in more discoveries being made.
Forsyth County's method of valuing machinery and equipment in 1981 also was flawed, in that it included no trend factor to take inflation into account. As the years 1980-81 were a period of high inflation, this failure may have resulted in significant undervaluing of business machinery and equipment.
Perhaps most important, though, is plaintiffs' contention that the state's practice of exempting forty per cent of tobacco in storage from taxation discriminates against the railroads, whose property is taxed at 100 per cent of fair market value. Such a factor is especially important in Forsyth County, where almost thirty per cent of business inventories in 1981 ($271,000,000 out of $947,000,000) consisted of tobacco in storage. Volume III, pp. 183-185. Defendants cite ACF Industries, Incorporated v. State of Arizona, 714 F.2d 93, 94 (9th Cir.1983), for the proposition that untaxed property should not be included in an average of assessed value for taxed property. However, the property in ACF Industries which was exempt was business inventories. It was not limited to one particular industry. Furthermore, the exemption was total, not partial.
By contrast, N.C.G.S. § 105-277(a) is clearly designed to benefit the tobacco industry particularly those businesses which warehouse tobacco. While most businesses, including railroads, have inventories, most businesses do not store tobacco. Whether the property tax discrimination is de jure or de facto, it is prohibited by Section 306. Clinchfield I, 527 F. Supp. at 785 n. 5. By requiring tobacco warehousers to pay tax on only sixty per cent of their tobacco in storage, the state has discriminated against the railroads. As a result, in counties where tobacco in storage forms a substantial portion of business inventories, those counties will not assess all business personal property at actual market value. As almost thirty per cent of Forsyth County's business inventories in 1981 consisted of tobacco in storage, it has failed to sustain its burden of showing that locally-assessed commercial and industrial personal property in that county is assessed at 100 per cent of actual market value.
3. Robeson County
Robeson County presented evidence through James A. Jacobs, its tax supervisor. He testified that Robeson follows the same general procedures as Rowan and Forsyth, with some differences due to the fact that it is a largely rural county. Volume III, p. 223; Robeson Exhibits 1-15. *1015 Furthermore, unlike Forsyth County, Robeson used the trend manual in 1981 to value machinery and equipment. However, rather than including a residual of twenty-five per cent as recommended by the Department of Revenue, Robeson used a twenty per cent residual. Volume III, p. 233. The decision to lower the residual was not based on any actual study of market conditions, but was based on general conversations with business personal property taxpayers in the county. Volume III, pp. 248-249.
Robeson County also did not include a LIFO prohibition on its 1981 property tax form. Instead, the listing form instructed the taxpayer to list inventories at "100% cost, as adjusted for income tax purposes." Volume III, pp. 236, 246. Therefore, taxpayers who reported inventory on their income tax forms using the LIFO method probably would also use it on their property tax forms, thus undervaluing inventory.
Perhaps the weakest part of the county's assessment practices comes in the area of verification. The county has no periodic system of checking income tax returns. Although 135 returns were checked in 1979 and 128 were checked in 1980, only four were checked in 1981, the relevant tax year, and in 1982 and 1983 no income tax returns were reviewed. Volume III, p. 253. Of the four returns checked in 1981, Mr. Jacobs did not know whether the taxpayers had reported inventory on a LIFO or FIFO basis. Volume III, p. 253. The tax office has never conducted an audit of a taxpayer's books and records. Volume III, p. 255.
Robeson County also has over $1.6 million of tobacco in storage which, like Forsyth County, was valued for property tax purposes at only sixty per cent of fair market value. There was no evidence placed in the record as to what percentage of business inventories tobacco in storage represented.
Therefore, for the reasons set out herein, the court concludes that Robeson County failed to sustain its burden of showing that all or most locally assessed business personal property is assessed at 100 per cent of actual fair market value.
4. Rutherford County
Rutherford County, which was represented at the trial by Von Q. Holland, its tax supervisor, is a small (55,000 people) predominately rural county. In 1981 it had 1,267 business personal property taxpayers. Volume IV, p. 9. The principal businesses in the county are textile manufacturing and used car dealerships. Volume IV, pp. 9-10. Mr. Holland testified that Rutherford generally follows the same procedures as the other counties. Rutherford Exhibits 1-24. However, his office performs all valuations on farm machinery and livestock. Volume IV, p. 14. It also uses the trend manual to value machinery and equipment. Volume IV, p. 25. As used car dealerships comprise a large portion of the county's business, the tax office takes extra steps to insure that all of these automobiles are listed and properly valued. On January 1 of each year Mr. Holland drives around the county and counts all of the cars on each dealer's lot. Volume IV, pp. 38-39. Vehicles in the county are assessed at wholesale value due to the large number of dealerships. Volume IV, p. 30.
However, Rutherford has not developed any systematic method of auditing tax returns. They do look at each taxpayer's prior year's listing, but if there are no irregularities apparent on the face of the listing form, no further inquiry is made. Volume IV, p. 74. Mr. Holland's office does not regularly request income tax returns from the Department of Revenue. In 1981, only two or three returns were requested. Volume IV, p. 74. The tax office has never conducted on-site audits of the books and records of business personal property taxpayers. Volume IV, p. 73.
Based on these reasons, the court concludes that Rutherford County has failed to carry its burden of showing that all locally assessed commercial and industrial personal property is assessed at 100 per cent of actual fair market value.
*1016 5. Catawba County
Testimony for Catawba County was elicited from Bobby R. Miller, its tax administrator, and Jackie Spencer, its personal property auditor. Catawba County's population is a mixture of rural and urban, with approximately 100,000 people. In 1981, there were 4,076 business personal property taxpayers. Volume V, p. 2. Mr. Miller testified that the county generally performs the same procedures as the other intervening counties. Catawba Exhibits 1-2. With respect to inventories, the 1981 listing form specifically prohibited LIFO. Volume IV, p. 101, Volume V, p. 2. Desk audits are performed on all business property tax returns. Volume IV, p. 92. The county also systematically checks the income tax returns of twenty per cent of the corporations in the county every five years and matches them against their property tax returns, and did so in 1981. Volume IV, p. 93.
Like Rutherford and unlike Rowan, the Catawba County tax office assesses the value of farm machinery and equipment. However, the method used in 1981 for valuing machinery and equipment did not take inflation into account. Volume V, p. 9. In 1983, the county began using the trending manual, resulting in substantial increases in assessment of personal property. Volume V, p. 9.
Catawba County's system of verification through the use of income tax returns also had a problem in that the data from income tax returns could not be used to verify taxpayers' reported cost of machinery and equipment. Volume V, p. 26. No effort was made to determine whether LIFO was being used when reporting inventory for income tax purposes. The county also did not conduct any on-site audits of taxpayer's books and records for verification purposes.
Thus, while like the preceding counties, Catawba appears to make a concerted effort to assess all business personal property at 100 per cent of fair market value, it has failed to carry its burden of showing that it does, in fact, do so.
6. Iredell County
Testifying for Iredell County was Howard Vanderford, the county's business personal property appraiser. He testified that the county has a population of 82,000 and that in 1981 there were 2,671 business personal property taxpayers. Volume V, pp. 34-35. Iredell generally follows the same procedures as the other counties regarding the listing of business personal property. Volume V, p. 39; Iredell Exhibits 1-13.
One difference, though, involves the valuation of machinery and equipment. In 1981, machinery and equipment was valued using useful-life tables developed by the Department of Revenue a number of years before. This system utilized a year life cycle for various categories of businesses, applying a straight line depreciation with a twenty per cent residual. The depreciation is based on the life cycle of machinery and equipment in the particular industry. Some time before 1981, this system was abandoned by the Department of Revenue in favor of the trending method. Volume V, pp. 40-44. The advantage to using the trend manuals is that they take inflation into account, while the system utilized by Iredell County did not. As the years 1980-81 were in a period of high inflation, the county's system of valuing machinery and equipment probably resulted in the undervaluation of that property. Volume V, pp. 64-67.
Iredell also failed to include in 1981 an instruction on the listing form prohibiting the use of LIFO to value inventories. The form simply stated that inventories should be listed at "100% cost as adjusted for North Carolina income tax purposes." The form also failed to instruct the taxpayer that inventory cost includes freight and sales tax as well as invoice cost. Volume V, pp. 60-65. Thus, as with the other counties, inventories may have been undervalued.
Finally, while all prior property tax returns were reviewed, Iredell County had no systematic program for comparing property *1017 tax returns with income tax returns to look for discrepancies. Only thirty-nine income tax returns were reviewed in 1981, and no field audits were performed. Based on these reasons, the court concludes that Iredell County has failed to meet its burden of showing that all business personal property in the county is assessed at 100 per cent of fair market value.
7. Granville County
Granville County, which presented evidence through its tax supervisor, Daniel M. Faucette, is a relatively small rural county with a population of 34,000. In 1981, there were 632 business personal property taxpayers. Faucette's office generally follows the same procedures regarding listing and verification as the other tax assessors' offices. Granville Exhibits 1-24. The prohibition against LIFO is included on the listing form. Farm machinery and livestock are valued by the tax office, not the taxpayer. Volume V, pp. 77-81, 88-89.
However, all machinery and equipment were valued in 1981 using an eight per cent straight line depreciation down to a twenty per cent residual. No consideration was given to the effects of inflation. The listing form also failed to instruct the taxpayer that the "cost" of machinery and equipment included sales tax, freight and cost of installations. Volume V, pp. 102-103. Thus, Granville County's methods may have understated the value of machinery and equipment in 1981.
Granville County also has very limited verification procedures. Prior property tax returns are regularly checked to see if there are any major changes, but only nine income tax returns were requested of the Department of Revenue in 1981. Income tax returns are requested on a random basis and follow no set pattern. The tax office has never conducted a field audit to review taxpayers' books and records. Volume V, pp. 105-106. Mr. Faucette testified that he needed additional personnel and had requested funding for another position, but that request had been denied. He also testified that although all business personal property of which he knows is listed, he believes that there is property in his county which is not listed. He thinks that his county can do a better job if he can get another person. Volume V, pp. 107-108.
Based on these findings, the court concludes that Granville County has failed to carry its burden of showing that locally-assessed business personal property in the county is assessed at 100 per cent of actual market value.
8. Vance County
Vance County was represented at trial by Wilton Worthem, its tax supervisor. He testified that Vance County is predominantly rural, with a population of 36,700. Volume VI, p. 4. In 1981, the county listed 983 business personal property taxpayers. Volume VI, p. 5. Mr. Worthem testified that his office generally follows the same procedures as the other tax supervisors' offices with respect to listing business personal property. Volume VI, pp. 9, 21; Vance Exhibits 1-4. Inventories must be listed by the taxpayers at cost and they are specifically prohibited from using the LIFO method of valuation. Volume VI, p. 21. All machinery, equipment and livestock are listed by the taxpayer but assessed by the tax office. Volume VI, pp. 15, 18-20.
There is a problem, though, with the listing of machinery and equipment. No year of acquisition is requested, making it difficult for the tax office to properly depreciate property. A ten-year life is assigned to all equipment except computers, which are assigned a five-year life. Thus, the tax office will undervalue any equipment which has a recognized useful life longer than ten years. No attempt is made to include a trend factor to account for inflation. Volume VI, pp. 12-15, 28-32.
There also are problems with the county's verification procedures. In 1981, no income tax returns were checked and no field audits were performed. The standard procedure was to review each property tax return from the prior year. If no serious *1018 discrepancy appeared, no further checks were made. Mr. Worthem admitted that there have been few discoveries made in the past few years. Volume VI, pp. 11, 32-35.
Consequently, the court must conclude that Vance County has failed to sustain its burden of showing that all locally-assessed business personal property in the county is assessed at 100 per cent of actual market value.
9. Wilson County
The final county to present evidence was Wilson County. Testifying for the county was William L. Jacobs, its tax supervisor. The population of Wilson County is predominantly rural. There are approximately 61,000 people, and in 1981 there were 2,691 business personal property taxpayers listed. Mr. Jacobs testified that his office follows the same general procedures as the tax offices in the other testifying counties. Volume VI, p. 44; Wilson Exhibits 1-11. The 1981 listing form instructed taxpayers that the LIFO method of valuing inventory was prohibited. Volume VI, p. 63. Machinery, equipment and livestock are all assessed by the tax office. The tax office also makes comparisons of inventories and machinery and equipment between similar businesses of the same size. If there are any significant discrepancies, they will investigate further.
With respect to inventories, though, there is a problem with the 1981 listing form. In the list taker's guide, taxpayers are instructed that "inventory values should agree with the amount in the taxpayer's latest fiscal year income tax return and will be checked against their state return." Volume VI, p. 64. Thus, even though there is a prohibition against using LIFO on the property tax return, it is likely that a taxpayer who uses LIFO for income tax purposes will state the same value for inventory on their property tax form, thus understating the value of their inventory.
Wilson County also listed $3,735,072 in market value of tobacco in storage in 1981. Volume VI, p. 70. As in the other counties, this tobacco was assessed by statute at sixty per cent of actual market value. Although Wilson County has no control over the assessment rate of tobacco, this rate nevertheless discriminates against the railroads and consequently violates Section 306 of the 4-R Act.
Furthermore, Wilson County's method of valuing machinery and equipment in 1981 failed to take inflation into account. Instead, it used a ten per cent straight line depreciation down to a twenty per cent residual. Volume VI, pp. 54, 68. As the years 1980 and 1981 were in the middle of a period of high inflation, this method would have resulted in machinery and equipment being undervalued.
The county's verification procedures also are limited. In 1981, between 110 and 120 business tax returns were requested from the Department of Revenue, but there is no systematic program of checking a certain number each year. Field audits of taxpayers' books and records are not conducted. Of the returns that were reviewed in 1981, the tax officials compared inventories on approximately fifty. They found that forty per cent of those returns showed a discrepancy in the inventory figures, resulting in about $38,000 in discoveries. Mr. Jacobs testified that he needs more employees in the business personal property section of his office, but that his request for a budget increase had been turned down. Volume VI, pp. 70-78.
Based on the foregoing reasons, the court concludes that Wilson County has failed to carry its burden of showing that locally-assessed business personal property is assessed at 100 per cent of actual market value.
11. Other Evidence
Defendants also called Mr. Roger Ellis, a property valuation specialist for the Ad *1019 Valorem Tax Division, to testify. At plaintiff's request, Mr. Ellis had been asked to secure information about the nine intervening counties in order to compare 1981 income tax valuations with business personal property tax valuations for the same year. The returns requested represented those of the top ten taxpayers in each county plus a random selection in each county of the third taxpayer from each letter of the alphabet. However, due to the limited amount of time available before trial, plus other problems, Mr. Ellis was only able to get meaningful figures from Rutherford, Iredell and Granville Counties. Rutherford Exhibit 25; Iredell Exhibit 16; Granville Exhibit 25; Volume VI, pp. 95-98.
The reason that such a cross-check is valid is that there is no advantage for the taxpayer to show a reduced value of his investment on his income tax return. In fact, the higher his investment on that return, the greater his depreciation expense deduction will be. Thus, if investment values on income tax returns are near to or the same as values on property tax returns, it provides some evidence that taxpayers are not deliberately undervaluing property for property tax purposes. A review of the three studies shows that investment values on business personal property tax returns were close to or above investment values on income tax returns. Thus, the studies appear to lend support to defendants' position that business personal property is being properly valued. Volume VI, p. 99.
However, the results of this test do not convince the court that its conclusion with respect to the level of assessment of business personal property in Rutherford, Iredell and Granville Counties was wrong. First, defendants could compare the results of only about half of the income tax returns in each county with business personal property tax returns, due to the fact that some multi-county taxpayers did not list their personal property by county, or because no income tax file was available for a particular taxpayer. See, e.g., Granville Exhibit 25. Furthermore, as defendants admitted, the study does not show actual market value since the book values shown on the income tax returns do not necessarily relate to market value. Volume VI, pp. 103-104. Therefore, while the study constitutes some evidence in support of defendants' position, it is not sufficient to show that Granville, Iredell and Rutherford Counties assess business personal property at 100 per cent of actual market value.
CONCLUSIONS OF LAW
The Court of Appeals held in Clinchfield I that once plaintiffs showed discrimination through a valid sales assessment ratio study with respect to defendants' assessment of their real property, the burden was on the state to show that plaintiffs' business personal property should be assessed at a different rate. Defendants chose to meet that burden in this case by presenting evidence which they assert shows that locally-assessed business personal property is assessed at 100 per cent of actual market value, and that plaintiffs therefore suffer no discrimination with respect to their personal property. However, for the reasons already set out herein, the court has concluded that defendants have failed to show that locally-assessed business personal property in the nine intervening counties is indeed assessed at fair market value.
Furthermore, defendants did not attempt to show ratios of assessment somewhere between 100 per cent and those stipulated for real estate, thus failing "to adduce evidence which would ... allow a meaningful alternative remedy to be fashioned." 700 F.2d at 132. Given the problems with Dr. Netzer's study, already discussed herein, and given the counties' obvious desire to assess business personal property at 100 per cent of fair market value, the court has no doubt that the actual levels of assessment of business personal property are greater than those set out in the stipulations. However, those ratios clearly are not 100 per cent, and defendants have not provided the court with a "meaningful alternative" to the real estate ratios. Accordingly, the court must adopt those ratios *1020 with respect to all commercial and industrial property in the nine counties. Furthermore, with respect to the thirty-nine counties which did not intervene or present evidence and against whom plaintiffs continue to seek relief, the court will also apply the stipulated ratios to all commercial and industrial property in those counties. The ratios for those forty-eight counties are set forth in Appendix A to this opinion.
Accordingly, the court will issue an order permanently enjoining defendants from assessing or collecting property taxes on all railroad operating property, both real and personal, at ratios of assessed value to actual market value in excess of the ratios established by stipulation for locally-assessed commercial and industrial real property in North Carolina for the 1981 tax year.
County Ratio Catawba 67.54 Forsyth 81.40 Granville 63.96 Iredell 71.17 Robeson 78.70 Rowan 56.21 Rutherford 58.19 Vance 60.07 Wilson 64.35 * * * * * Beaufort 81.18 Bertie 88.00 Bladen 46.75 Buncombe 49.50 Burke 93.83 Caswell 78.87 Chatham 73.59 Chowan 44.66 Craven 81.73 Davie 90.75 Duplin 71.50 Edgecombe 56.34 Franklin 50.27 Gates 36.63 Henderson 51.70 Jackson 67.43 Jones 47.41 Lincoln 72.90 Martin 58.41 McDowell 90.09 Mitchell 76.12 Moore 80.41 Nash 78.50 New Hanover 80.74 Onslow 84.48 Pasquotank 50.90 Pender 74.25 Perquimans 37.84 Polk 86.60 Randolph 69.08 Rockingham 77.44 Sampson 74.25 Scotland 85.69 Stanly 70.18 Transylvania 87.78 Warren 46.75 Wayne 75.35 Wilkes 79.30 Yancey 69.25NOTES
 Of the sixty-one remaining counties which did not intervene, plaintiffs withdrew their claims against twenty-two, leaving thirty-nine counties which did not present evidence and against which plaintiffs continue to seek relief. Those counties are: Beaufort, Bertie, Bladen, Buncombe, Burke, Caswell, Chatham, Chowan, Craven, Davie, Duplin, Edgecombe, Franklin, Gates, Henderson, Jackson, Jones, Lincoln, Martin, McDowell, Mitchell, Moore, Nash, New Hanover, Onslow, Pasquotank, Pender, Perquimans, Polk, Randolph, Rockingham, Sampson, Scotland, Stanly, Transylvania, Warren, Wayne, Wilkes and Yancey. Final Pre-Trial Order, Stipulation 5.
 Centrally-assessed property is property which is assessed by the State Department of Revenue, as opposed to assessment by county tax authorities.
 Motor vehicles, on the other hand, do have an established second-hand market. As a result, trending manuals and depreciation schedules for motor vehicles have tended to be much more accurate.
 With respect to the listed values of personal property, Dr. Netzer simply applied the amounts assessed by the individual counties and supplied to plaintiffs through discovery.
 The Department of Commerce figures gave total inventory value at book value. Since government statistics show that book value generally understates market value of manufacturing inventories by approximately 29 per cent, retail inventories by approximately 9 per cent, and wholesale inventories by approximately 20 per cent, Dr. Netzer increased his figures in each of the three inventory categories by these percentages. Volume I, pp. 68-69, 87.
 Business personal property is divided into three categories: motor vehicles, inventories and machinery and equipment. All of the defendants presented evidence that they receive each year a list from the North Carolina Department of Motor Vehicles listing every vehicle registered in their county. Thus, the defendant counties have a method of insuring that every vehicle in each county is listed on personal property tax returns. Furthermore, there are pricing guides which value vehicles by make and year. As there is a ready market for used cars, these guides are considered to be fairly accurate. Plaintiffs did not seriously dispute any of this evidence. Thus, it appears to the court that all of the nine intervening counties are listing and assessing their motor vehicles at 100 per cent of actual market value.
 A "discovery" can occur when the tax office finds property not listed, when it assesses property at a value higher than that given by the taxpayer, or simply when it assesses a penalty for failing to list at all.
 In 1984, Forsyth County did begin to use the state's trend manual to value machinery and equipment.
 See N.C.G.S. § 105-277(a).
 Farmers are permitted to assess a value to smaller farm implements. However, the court considers the value of such property to be so insignificant as to have a de minimis effect on the total value of business personal property.