Washtenaw County v. State Tax Commission

Brickley, J. These consolidated suits involving the State Tax Commission and five Michigan counties concern principally the method of arriving at the state equalized value of real property in two respects: the effect of land contracts and other "creatively financed” transactions on "true cash value,” and the need to employ a "trending” factor to the data base of sales-ratio studies in order to more accurately pinpoint the "true cash value” of property on "the tax day.” We hold that, when the seller of real property has contributed in a significant respect to the buyer’s ability to obtain financing at an interest rate artificially below the prevailing cost of borrowing, a method of valuation that separates the cost of such artificially low financing from the sales price to achieve the "true cash value of such property” is required by the applicable statutes, when they are read, as they must be, to comport with the constitution. We also find that the State Tax Commission has not committed an error of law or applied a wrong principle in refusing to adopt the plaintiffs’ proposed method of trending for the sales-ratio studies. *351 I Background and Statement of Facts Five counties are appealing the final state equalized valuations adopted by the State Tax Commission for each of those counties to be used for their 1982 property tax levy. The Washtenaw, Oakland, and Lapeer County cases came to this Court from the Court of Appeals where they were consolidated. Washtenaw County v State Tax Comm, 126 Mich App 535; 337 NW2d 565 (1983), lv gtd 419 Mich 864 (1984). The Livingston and Ingham County cases were originally joined with the Washtenaw case in the Court of Appeals, but were subsequently separated and not heard by that Court, although they were allowed to file a joint brief as amici curiae. When we granted leave in Washtenaw, Livingston and Ingham Counties sought and were granted their application to appeal the decision of the commission, bypassing the Court of Appeals, Livingston County v Michigan and Ingham County v Michigan, lv gtd 419 Mich 864 (1984), and were consolidated with Washte-naw. In each case, the state equalized value (sev) determined by the commission for residential and certain other classes of property in each of these counties resulted in higher values than those calculated by the county boards of commissioners in the course of establishing their respective county equalized values (cev). 1 The petitioning counties made or requested the commission to make two types of adjustments to *352 the results of their equalization studies. Those adjustments account for the differences between the cev and sev in these cases, and the propriety of those adjustments is at issue here. The commission, in its equalization studies, disallowed the adjustments made or requested by the counties, finding them to be contrary to the State Tax Commission Assessor’s Manual and applicable statutory and constitutional provisions. The commission also found individual studies in those counties to be lacking in other respects. The first adjustment to the equalization studies in question was an effort by the counties to account for the effect of "creative financing.” "Creative financing” occurs where the seller permits the buyer to assume the seller’s existing low mortgage ("assumption”), or the seller pays a lending institution to loan to the buyer at the low market interest rate ("buy down”), or the seller *353 pays "points” to an institution to further an fha-guaranteed, below-market interest rate mortgage loan to the buyer ("fha mortgage”), or the seller sells to the buyer on a land contract ("land contract”), or the buyer both assumes the seller’s old mortgage and borrows more from the mortgagee ("blend”). [Washtenaw County, 126 Mich App 542-543 (opinion of Allen, J.).] Substantial percentages of the sales used in the equalization studies in question involved creative financing. In determining their cev, the county commissioners applied or advocated the application of a discount in order to account for the price enhancement effect believed to be the by-product of seller-extended credit. Several counties used studies to support their contention that sales involving creative financing resulted in prices that were six to ten percent higher than those of cash or conventional term sales. The counties attributed this price enchancement effect to the high rates for institutional mortgages. (In 1982, they varied from 15.75 percent to 18.5 percent.) Typically, it is contended that, when a buyer takes advantage of a legally limited interest rate, assumes seller’s existing low interest rate mortgage, or is offered a buy-down by the seller, the seller accounts for the difference between the low rate and the prevailing interest rate by increasing the selling price of the property. A land contract is an example of a legally limited rate; by law the seller may charge no more than eleven percent interest, MCL 438.31c(6); MSA 19.15(lc)(6). The commission disagreed with the counties’ factual conclusion that creative financing affects sales price, as well as with the legal premise that such an adjustment is allowable under the law, even if the effect were to be shown. In the second type of disputed adjustment, the *354 counties weighted more heavily the sales which occurred closest in time to "tax day,” December 31, 1981. In a period of declining property values, the counties claim that the sales occurring closest to tax day will be the most reliable for purposes of conducting a sales-ratio survey. The counties advocate an adjustment to the sales in their sales-ratio studies, so as to reflect the declining market as of December 31; this adjustment is referred to as "trending.” Unless such an adjustment is made, the counties maintain that the assessment levels would possibly exceed fifty percent. The commission disagrees, arguing that trending will not accomplish its intended purpose. Moreover, because the method is dependent upon a sufficient number of sales in a short period of time, the defendant further argues that it would be applicable in only a few of the more populous counties. Because the facts in each county differ, the particulars are set forth by county. Washtenaw County The County Board of Commissioners for Washte-naw County adopted a 1982 cev of $1,768,505,675 for the residential class of property. The sev for the same class was determined by the commission to be $1,937,707,132, a figure derived from the county equalization study before reduction for creative financing. Washtenaw County conducted several studies, the final conclusions of which were that, of the 1981 residential sales in the county, eighty-nine percent involved creative financing, and that the prices of those sales were consequently enhanced by 8.2 percent. The critical sales-ratio study which resulted in these figures used a sample of 579 disclosed sales, 519 of which were creatively financed. As backup to this study, the county introduced four studies using cash *355 equivalency discounting; a study of 1981 residential land contract sales; thirty-three affidavits from real estate brokers and salespersons; letters from local appraisers; an affidavit showing that, in one school district, alone, on one day, there were twenty-three residential listings giving one price for a cash sale and a higher price for a land contract sale; and finally, evidence of institutional interest rates in 1981. The commission claims that neither the Michigan Constitution nor applicable statutory or administrative provisions permit this type of adjustment for creative financing. Moreover, the commission believes that the Washtenaw County studies were biased by self-interest and were flawed bé-cause they considered only one-half of the total market in Washtenaw County. Relying on a total of twenty-six studies regarding the effect, generally, of creative financing on selling price, the commission concluded that, while there was evidence on both sides, any effect was de minimus and no adjustments were necessary. In response, Washtenaw County criticizes the commission studies, because they rely on undisclosed sales and focus on locations outside Washte-naw County. Washtenaw County claims that the samples were neither representative of the entire state nor of any region encompassing Washtenaw County. Regarding the "trending” issue, or the weighting of sales used in sales-ratio studies that occur closest in time to tax day, Washtenaw County incorporates the arguments advanced by Oakland County. No specific adjustment of this type was made to the Washtenaw County equalized valuations. Lapeer County In advocating the first form of disputed adjust *356 ment, Lapeer County focused solely upon the land contract form of creative financing. It conducted a thirty-month sales-ratio study as well as an appraisal study to arrive at its cev for several classes of property in 1982. For the residential class, the county commissioners adopted a cev of $403,599,-481. The commission determined an sev of $437,411,779 — an 8.3 percent difference. The commission apparently relied on assessments which considered only the actual sales price of property sold within the study period. Because twenty-eight percent of the 1980 and 1981 sales in Lapeer County were by land contract, the county claims that the commission’s reliance upon the face value of those contracts was in error. By contrast, the county advocates a discounting method which takes the interest rate of the individual land contract, divides it by the then current money market rate of interest, multiplies that by the unpaid balance at the time of sale and adds the amount of down payment to arrive at the "correct sale price.” They argue that had this method been used in determining sev, the "total County residential class sev starting base would have been lowered by approximately 5.8% . . . .” The county makes essentially the same argument in regard to the agricultural and developmental classes. The commission apparently employed appraisal studies in arriving at sev for those classes; employing a market data approach to appraisal, those studies used the actual sale prices of land contract sales as comparables. The county claims that it presented "as part of its evidence a letter from the Federal Land Bank Association indicating that for appraisal work, land contracts must be discounted to a cash basis before they can be used.” Moreover, the county argues that any "factfinding” performed by the *357 commission in this case does not apply to Lapeer County, since most of those findings arose from and were presented in conjunction with the Wash-tenaw County hearings. The commission maintains that Lapeer County authorities "had no factual basis for their argument that the state equalized valuations for the agricultural, residential and developmental classes of property should be lower than that proposed by the State Tax Commission.” Lapeer County did not raise the trending issue. Oakland County The Oakland County Board of Commissioners set the 1982 cev for the residential class at $9,075,300,224. The commission adopted an sev for the same class of $9,619,223,876. The difference in this case is due to the county’s "trending” adjustment, used to account for the "drastic decline” in volume of residential sales for 1981, as compared to previous years; the alleged stabilization of the market for the first five months of 1981; and a decline in values during the second half of the year. In order to account for these market phenomena, which the county maintained would not be reflected in either of the two commission-approved study periods of thirty months and twelve months, the county relied on the most current assessment data available in 1981 to determine the 1982 cev. The county supported its methodology with an independent sales study analysis by mathematician Dr. Michael Skaff. The sev adopted by the commission was based on the Oakland County equalization studies which were not adjusted for "trending” of sales on tax day. The commission opined "that the use of any equalization study of less than twelve months duration would produce insufficient data upon *358 which an informed and conscious exercise of value judgment could be made.” It found that a study, such as that conducted by Oakland County "would only exaggerate a temporary fluctuation or statistical accident in the real estate market.” In the Matter of Washtenaw County, Oakland County and Lapeer County 1982 State Equalization, Findings of Fact and Opinion. Moreover, as noted above, the commission argues before this Court that the development of a trending formula would be impracticable as applied in some counties and impossible as applied to others. Oakland County did not make an adjustment for creative financing in its equalization study. However, the county argues in favor of such an adjustment, "[considering that over 50% of the 1981 sales in Oakland County involved land contract financing and that another significant segment of the sales involved assumptions of favorable mortgage terms . . . .” Livingston County The Livingston County Board of Commissioners determined the 1982 cev for the residential class to be $832,009,727. The county advocates adjustment for both creative financing and trending. The commission adopted an sev of $919,348,782, a 10.5 percent difference. Livingston County submitted both a thirty-month and a twelve-month sales-ratio study. The twelve-month study compared the numbers of properties which sold for cash or on conventional terms with those which were creatively financed. Using a cash equivalency formula to reflect the portion of sales price attributable to financing, the board determined that the 1981 sales prices should be lowered by thirteen percent, resulting in a *359 reduction of 8.2 percent in the 1982 assessed values to obtain cev. The commission found that the twelve-month study was not conducted in accordance with the Assessor’s Manual. In addition, it held that the creative financing adjustment was improper, because the thirteen percent discount was applied in an across-the-board manner to sales not involving creative financing, and because the methodology by which the thirteen percent discount was determined was not shown. Another Livingston County study also attempted to account for the declining market in 1981 by collecting sales samples for the three months prior to and three months after tax day. The county argues that if its evidence of a continuing decline in the market is not considered in setting sev, the level of assessment in the county as of tax day may exceed the fifty percent limit. The commission’s position on this issue is identical to that taken in the Oakland County case. The commission maintains that "[i]mperfect value judgment will not be made more reliable by such trending. Assessments will still remain imperfect.” Ingham County The 1982 cev and sev for the residential property in Ingham County are the same: $1,452,045,-574. Although the county argued in favor of a creative financing adjustment to sev, it did not make such an adjustment to its 1982 cev, because the commission-approved methods did not seem to allow for such an adjustment, and because the market data available, comparing cash/conventional sales with creatively financed sales, covered only sixty percent of the residential property in Ingham County. The county, nonetheless, presented its sales data and the testimony of a statis *360 tical analyst before the commission, in an effort to convince the commission that its methods were outdated and that the sev for Ingham County should be adjusted downward. The sales data gathered by Ingham County showed that land contracts constituted approximately fifty percent of the sales in 1981. A comparison of cash/conventional versus land contract terms revealed an average ten percent increase in the sales price for the latter type of financing. Thus, the county advocated discounting all residential cash values by five percent. The commission rejected the county’s arguments, adopting the same figure as that presented by the Ingham County Board of Commissioners, without adjustments. The commission maintains that the expert and opinion testimony presented by Ingham County showed that the difference in ratios between creatively financed and non-creatively financed sales could be due to any number of undefinable factors. Ingham County does not raise the trending issue. Court of Appeals Ruling The Court of Appeals majority agreed with Washtenaw, Oakland, and Lapeer Counties’ creative financing position, relying in part on a study by the American Institute of Real Estate Appraisers that "land contract terms typically create a higher selling price . . . .” The Court noted that the counties had shown "that, to a certain extent, creative financing does artificially enhance the 'sales price.’ ” Washtenaw, supra, p 539, n 2. The majority concluded that the constitution requires uniformity and assessments based on true cash value, and that "a tax assessment system which *361 does not consider creative financing is in fact unconstitutional.” Id., p 541. The Court of Appeals also unanimously agreed with Oakland and Washtenaw Counties that trending provides a more accurate picture of "true cash value.” The Court noted that the commission had recognized the declining market in 1981 by authorizing twelve-month sales-ratio studies. By weighting the sales occurring closer to tax day, the Court felt that the goal of uniformity would be better served. Thus the cases were remanded to the commission "to develop a method to account for creative financing . . . .” Washtenaw County, supra, p 542. In respect to Oakland and Washtenaw Counties, the majority opinion joined Judge Allen’s partial concurrence, which remanded "with instructions that sales-ratio studies extending over not less than a twelve-month period be reviewed and some weight given to sales occurring in the period closest to tax day.” Judge Allen’s opinion also instructed that "[a] formula for weighting to reflect changing market trends close to tax day should be devised by the [commission].” Washtenaw County, supra, p 556 (opinion of Allen, J.). II A. Scope of Review On behalf of the commission, the Attorney General argues that the Court of Appeals exceeded the scope of appellate review by ignoring the commis-. sion’s factual finding to the effect that creative financing has at most a de minimus effect on the value of the property at issue. The commission had arrived at that conclusion by criticizing the studies put forth by the plaintiff counties and by relying on the commission’s own studies. *362 The scope of appellate review is clearly set forth in the constitution: In the absence of fraud, error of law or the adoption of wrong principles, no appeal may be taken to any court from any final agency provided for the administration of property tax laws from any decision relating to valuation or allocation. [Const 1963, art 6, § 28.] We view the commission’s finding of "de mini-mus effect” regarding the issue of creative financing as somewhat disingenuous. The commission has mightily resisted the principle that creative financing may be considered in evaluating the selling price of property. The Attorney General, in resisting that principle, argues that the counties used "unauthorized adjustments ... to recognize the alleged effects of so-called 'creative financing’ ” (emphasis added), and that "there is no method developed by the State Tax Commission, at the present time, to gather the data necessary to account and adjust for creative financing.” The commission opined, as a matter of law, that state law does not permit the exclusion of creative financing from sale prices used in sales-ratio studies. It also found that "it is undeniably true” that half of the sales in these counties are creatively financed. In light of these findings, the commission cannot now shield its decision from review by a factual finding of "de minimus effect.” Furthermore, the record on which the factual "de minimus” finding was based was developed by the commission after a remand by the Court of Appeals in Washtenaw for findings of fact and written opinion. Therefore, the finding would not apply to the claim of the counties of Ingham and Livingston, that were not involved in Washtenaw, *363 but that were also denied relief by the commission on the creative financing issue. By its conclusions of law, the commission has adopted a reviewable principle in this case, and we will consider the legal principle without relying on the commission’s factual findings. Our inquiry in this case will not rest on the degree to which creative financing influences the value of property, but rather on the principle of whether its possible effects must be considered. B. Creative Financing There seems to be no argument that it is the goal of the assessment process to determine, in the abstract, the usual selling price of a given piece of property between a willing buyer and a willing seller and to develop methodologies that make it possible to achieve uniformity in making such determinations within and between counties. The consensus flounders in this case when presented with the question whether the reported sales price of property may be examined to determine if it represents more than the value of the property being sold. The commission argues that the Legislature has set forth, pursuant to Const 1963, art 9, § 3, the definition of true cash value as the "usual selling price,” MCL 211.27(1); MSA 7.27(1), and that even if creative financing has an effect on the sales price of a piece of property, it will in time become part of the usual selling price. The commission points out that it has been their policy to allow for the exclusion of individual sales from sales-ratio studies if the terms of financing are unusual. This, they contend, protects the ratio studies and the equalization figures from distortion by such unusual transactions. The commission thus concludes that when, as is admitted here, "the market con *364 tains a preponderance of sales which are 'creatively’ financed, then those sales and their attendant 'prices’ are the market and the 'usual selling price’ or 'cash value’ . . . The counties argue that an amount added to the sale of property to compensate for obtaining below-market financing can never be part of the sales price of the property, regardless of how usual it becomes. The authority and framework for the assessment and equalization of property is founded in the constitution. Const 1963, art 9, § 3 provides in part: The legislature shall provide for the determination of true cash value of such property; the proportion of true cash value at which such property shall be uniformly assessed, which shall not, after January 1, 1966, exceed 50 percent; and for a system of equalization of assessments. The Legislature has determined the "true cash value” referred to in Const 1963, art 9, § 3 in § 27(1) of The General Property Tax Act: "[C]ash value” means the usual selling price at the place where the property to which the term is applied is at the time of assessment, being the price which could be obtained for the property at private sale, and not at forced or auction sale. [MCL 211.27(1); MSA 7.27(1).] The constitution clearly states that the Legislature shall determine the "true cash value of such property” (emphasis added). The implementing legislation refers to "the price which could be ob *365 tained for the property” (emphasis added). It is evident to us from these plain words of both the constitution and the statute that any method of assessment or equalization of real property that ignores significant and ascertainable components of a sales price which are relevant to the cash value of the property — be they usual components or not — is flawed. There is no dispute that the cost of borrowing in 1981, with mortgage rates in excess of fifteen percent, contributed to the depressed value of property that was reflected in the sales-ratio studies of the plaintiffs. It is also a safe assumption that the disparity between the prevailing cost of a mortgage and a land contract, limited by law to an interest rate of eleven percent, explained the prevalent use of that most common form of creative financing in 1981. In contrast to the more conventional transactions, where the buyer buys the property from the seller and finances the purchase through a financial institution, the buyer under a land contract or other form of creative financing purchases both the real property and the financing from the seller. While an individual property owner/seller is not normally in a position to earn on a secured investment an amount equal to the mortgage lending rate of a financial institution, in 1981 it was possible, if not probable, that a land contract vendor was giving up something of value by loaning money at eleven percent interest. It is also possible that the vendor was selling property that, without creative financing, would not have sold at all or that would have sold for a lower price. As to the land contract purchaser, it is not speculative to say that borrowing at eleven percent, instead of *366 at fifteen percent or more, represented considerable value. 2 It is obvious that in 1981 a land contract at eleven percent had a positive value to the purchaser and a negative or, at best, a neutral value to the vendor. We cannot gainsay the possibility that any such value for the below market rate financing would normally be reflected in the sales price of the property. There are many factors external to property that influence and determine its real value; they include neighborhood conditions, location, property tax, general economic conditions, and the cost of borrowing. However, unlike those factors that influence value but are external to the transaction, creative financing by its nature is part of the transaction between the parties to the sale. And, to the extent that it represents something of value to one of them, it is not part of the real property and cannot be included in the determination of the true cash value of property. We have made it clear in the past that the sales price of property does not necessarily determine the true cash value of the property. Fisher-New Center Co v State Tax Comm, 380 Mich 340, 362; 157 NW2d 271 (1968). In analyzing a somewhat analogous problem, in Antisdale v City of Galesburg, 109 Mich App 627; 311 NW2d 432 (1981), rev’d 420 Mich 265; 362 NW2d 632 (1984), where the Michigan Tax Tribunal 3 had refused to factor out unusually low inter *367 est rates on a federally subsidized mortgage in arriving at its value, we reversed the Michigan Tax Tribunal and stated: To say that the mortgage balance had a value of $1,500,000 when it could be entirely assumed by giving up the use of $433,232 is to give an entirely new and foreign meaning to the word value. By failing to discount the outstanding mortgage balance of the comparable properties to the actual cost to the investor the Tax Tribunal adopted a wrong principle. [Antisdale, supra, pp 282-283.] Even though that case involved the assessment of an individual piece of property, rather than equalization, the principle announced there applies with equal force here. The same general rules of valuation apply to both assessment and equalization. See MCL 211.27(3); MSA 7.27(3). The common denominator for either assessment or equalization is market value. Assessor’s Manual, Chapter VI, p 127. We note with interest that in Woodgate Associates v City of Lansing, 1 MTTR 488 (Docket No 6719, June 13, 1978), where, in establishing the *368 petitioner’s assessment, comparable sales that included creative financing were not discounted accordingly, the Tax Tribunal, in interpreting the statutory definition of true cash value, MCL 211.27(1); MSA 7.27(1), said: This statutory limitation is obviously intended to avoid an undervaluation of property as a consequence of using, for comparison properties, sales that are of a distress [sic] nature, where the sale price is apt to be below the normal market value. If artificially low sale prices are not indicators of true cash value, then it follows that artificially high sale prices are equally offensive and inappropriate. [Id., p 491. Emphasis added.] The Tax Tribunal further noted, It makes little difference whether the comparative sale price is artificially inflated (because of a trade off of favorable terms at sale) or artificially depressed (because of a forced sale), because both result in something other than true cash value as defined by statute and case law. If the net effect is that a reported sale price is either artificially depressed or elevated, and if that sale price directly affected the computation of true cash value (and therefore assessment) of another property, such assessment is erroneous. [Id., p 492.] The objectionable "favorable terms” referred to in Woodgate included assumption of outstanding mortgages and land contracts by the buyer at considerably lower interest rates than what were available at the dates of sale, balloon payments, and a payback to a buyer from a seller of a portion of the reported sale price as a "commission fee.” [Id., p 491.] The Attorney General makes a further argu *369 ment to the effect that since the constitution has given the Legislature responsibility for "defining” true cash value, the Legislature has done so without allowing for across-the-board discounting of creative financing. Therefore, such discounting is not permissible. In furtherance of that argument, the defendant points to another statutory provision, §27(3) of The General Property Tax Act, MCL 211.27(3); MSA 7.27(3), which deals specifically with the cost of financing included in the sales price of property. That statute, in relevant part, provides: Beginning December 31, 1978, a city or township assessor, a county equalization department, or the state tax commission before utilizing real estate sales data on real property purchases, including purchases by land contract, for the purpose of determining assessments or in making sales ratio studies for the purpose of assessing or equalizing assessments shall exclude from the sales data the following amounts allowed by subdivisions (a), (b), and (c) to the extent that the amounts are included in the real property purchase price and are so identified in the real estate sales data or certified to the assessor as provided in subdivision (d): (a) Amounts paid for obtaining financing of the purchase price of the property or the last conveyance of the property. (d) The purchaser of real property, including a purchaser by land contract, may file with the assessor of the city or township in which the property is located 2 copies of the purchase agreement or of an affidavit which shall identify the amount, if any, for each item listed in subdivisions (a) to (c). One copy shall be forwarded by the assessor to the county equalization department. The affidavit shall be prescribed by the state tax commission. [Emphasis added.] *370 The defendant claims that the constitutional language "The legislature shall provide for the determination of true cash value” is a grant of power to the Legislature to alter or in general define the meaning of true cash value. The defendant has advanced no authority, nor are we aware of any, that would permit such an expansion of the words to "provide for the determination of.” The general meaning of true cash value predated the Constitution of 1963, and it is not likely that the drafters would incorporate that phrase, with its long history of interpretation and settled meaning, only to have its future left to the whim of the Legislature. 4 The constitution specifies that property shall not be assessed beyond fifty percent of true cash value and to hold that true cash value can be defined by the Legislature would, for all practical purposes, make the fifty percent limitation meaningless. As Justice Cooley said, "A constitution is made for the people and by the people. The interpretation that should be given it is that which reasonable minds, the great mass of the people themselves, would give it. 'For as the Constitution does not derive its force from the convention which framed, but from the people who ratified it, the intent to be arrived at is that of the people, and it is not to be supposed that *371 they have looked for any dark or abstruse meaning in the words employed, but rather that they have accepted them in the sense most obvious to the common understanding, and ratified the instrument in the belief that that was the sense designed to be conveyed.’ (Cooley’s Const Lim 81).” [Quoted in Traverse City School Dist v Attorney General, 384 Mich 390, 405; 185 NW2d 9 (1971). Emphasis deleted.] Limitations on taxes and assessments often play a prominent role in deliberations leading to a constitution’s adoption or rejection. It is not likely or probable that the voters, in adopting the Constitution of 1963, would understand that true cash value would be what the Legislature said it was, rather than its historical if not literal meaning. 5 We thus examine § 27(3) with the view that, if it derogates from art 9, § 3, it cannot be valid. And, our interpretation remains consistent with the principle that, whenever possible, interpretations that result in constitutional invalidity will be avoided. People v McLeod, 407 Mich 632, 657; 288 NW2d 909 (1980). See Thoman v City of Lansing, 315 Mich 566, 576; 24 NW2d 213 (1946) ("presumption of constitutionality following taxing statutes is stronger than applies to laws generally . . .”). Under § 27(3), "[a]mounts paid for obtaining financing of the purchase price of the property” when "identified in the real estate sales data or certified to the assessor as provided in subdivision (d)” (emphasis added) must be excluded when "utilizing real estate sales data” in the process of *372 equalizing assessments. The key words here are "real estate sales data.” The commission argues that "it is clear that individual sales in a sales ratio study may be adjusted for amounts paid by the purchaser for obtaining financing ... if the purchase price includes such amounts and if such amounts are either certified to the assessor by affidavit, provided to the assessor by means of delivering copies of the purchase agreement, or included on the face of the deed or contract.” (Emphasis deleted.) The defendant would, therefore, narrowly construe this phrase to mean the documents of an individual transaction specifically identifying such costs, whereas the plaintiffs construe it as referring to the kind of real estate data they have accumulated to support their figures which purport to identify a higher selling price for creatively financed transactions. The defendant also construes § 27(3)(a) to refer to mortgage initiation fees, appraisal costs, credit checks, attorney fees, and so forth. We find the plaintiffs’ position more persuasive for the following reasons. First, the statute clearly and unequivocally refers to excluding the costs of financing "before utilizing real estate sales data . . . for the purpose of assessing or equalizing assessments . . . .” This certainly suggests that such exclusions of the cost of financing would have to include across-the-board adjustments, because equalizing almost always does just that. The equalization process does not deal with individual assessments. Second, "real estate sales data” is the term used most frequently, even in the commission’s Assessor’s Manual, supra, Chapter VI, p 129, to mean the kind of data relied upon in the studies offered by the plaintiffs in support of their finding of price inflation due to the cost of financing. Possible *373 sources for obtaining data are listed in the manual as realtors, lending institutions, and individual property owners. Third, the term "real estate sales data” is not the most precise and, therefore, the most likely way to describe the documents used in a real estate transaction. The words, in and of themselves, argue against the more narrow interpretation that they refer to an individual transaction. Finally, when this statute was adopted in 1976, subsection (d) employed the word "shall.” In 1978, "shall” was changed to "may.” 1978 PA 25. Subsection (d), in its original form, then, provided that all purchasers of real property would furnish copies of the purchase agreement or an affidavit identifying, among other things, the amounts paid for financing of the purchase price, as required in subsection (a). If by using "sales data” the Legislature meant the data of an individual transaction, then, adding the words "or certified to the assessor as provided in subdivision (d)” would have been repetitious. Under the defendant’s interpretation, the Legislature would only have had to refer to data identified in a real estate transaction, whether or not furnished to the assessor in subsection (d), or, if they had intended sales data to be confined to information that was furnished to the assessor, they would only have had to refer to the costs of financing identified in the information furnished to the assessor under subsection (d). A more logical and literal reading of § 27(3) is that amounts paid for financing shall be excluded, if they can be identified in real estate sales data that are gathered in assessment and equalization studies or if they can be identified in the documents of individual transactions. Reading § 27(3) in its entirety, consonant with the constitutional mandate to achieve true cash value, it is our *374 conclusion that § 27(3) excludes the cost of creative financing from the valuation of property. 6 We thus find that both § 27(1) and § 27(3) are within the constitutional mandate to determine true cash value and that both, the former by implication and the latter by expression, require the exclusion of creative financing from the equalization process to the extent such financing affects the sales price of real property. C. Trending The remaining issue in these cases is the propriety of statistically weighting sales study data to give greater weight to sales occurring closer in time to "tax day,” which by statute is the December 31 preceding each tax year. MCL 211.2; MSA 7.2. Prior to the 1982 tax year the commission allowed the use of a thirty-month sales-ratio study in arriving at a county’s equalized valuation. For three periods — the first six months of the year *375 prior to the tax year (which for these cases would be the first six months of 1981) and the previous two years — a ratio of the prior assessed value of the property to the value as determined by the sales study is computed. The three ratios are then added together and divided by three, the six-month period holding equal weight with each of the prior two years. In this manner, any wide fluctuation in one period between the aggregate values of property as assessed and as determined by a sales study is cushioned by the ratio of the two other periods. This final ratio is then used in the equalization process. Beginning with the 1982 tax year, the commission authorized a one-year sales study. This one-year study of sales occurring in 1981 could be utilized if a sufficient number of sales occurred within the locality so that the study would be statistically reliable and if a substantial decline in property values was indicated. Thus, the commission allowed such localities to immediately reflect the declining market in local property values, instead of having that decline cushioned by prior years. It is the contention of the three counties raising this issue that, as a result of the above statute, the use of even the optional one-year sales study is illegal and unconstitutional. Their premise is that all property must be valued for tax purposes as to its value as of the December 31 preceding the tax year. Therefore, they reason, by failing to statistically weight even the one-year study to give greater effect to sales occurring near to the tax date, the study is flawed. Furthermore, the real property market of 1981 was declining toward the end of the year, and since the one-year study gives equal weight to all sales throughout the year, the result of the equalization process, in their view, *376 was to equalize the property within each county above the maximum of fifty percent of true cash value. The Court of Appeals agreed with the counties, stating that "some such adjustment is needed to accurately reflect the value on tax day. While the commission correctly argues that insufficient sales occur on tax day for accuracy, and that a broad study may be necessary to encompass the fluctuations of the real estate market, there is no justification for ignoring evidence of an existing market trend.” Washtenaw, supra, p 555. While they expressed their preference for a system of weighting, the Court of Appeals did say they would "oppose use of any study of less than 12 months and also oppose extending the study into the next calendar year.” Id., p 556. The Court concluded that "[o]nly by permitting such weighting can an accurate picture of 'true cash value’ be presented.” Id. Unlike its position on the creative financing issue, here, the commission has neither denied the desirability of achieving up-to-date assessments, nor, in our view, stated an erroneous principle of law. Rather, the commission disputes the counties’ belief that the weighting scheme proposed will accomplish the desired result. They contend that there would only be a few populous counties in which there would be a sufficient volume of sales for which a weighting factor could be utilized. It will have an insignificant impact in a handful of populous counties and no impact whatsoever throughout the rest of the state. It will bring no appreciable precision to imperfect estimates of value. The "tax day” statute, MCL 211.2; MSA 7.2, in part provides: *377 The taxable status of persons and real and personal property shall be determined as of each December 31, which shall be deemed the tax day, any provision in the charter of any city or village to the contrary notwithstanding. No assessing officer shall be restricted to any particular period in the preparation of the assessment roll but may survey, examine or review properties at any time prior to or after the tax day. While tax day is the December 31 preceding the tax year, the assessment roll need not be completed until the first Monday of March of the tax year itself. MCL 211.24; MSA 7.24. This statute provides that "[o]n or before the first Monday in March” the local assessor "shall estimate, according to his best information and judgment, the true cash value of every parcel of real property and set the same down opposite such parcel.” Between the mandate to determine the "status” of property on December 31, and the mandate to estimate true cash value of every parcel "[o]n or before the first Monday of March” is an uncertainty of the day or period for which property should be valued, which has not been addressed or argued by the parties. If there were such a date on which value was to be established, it would, given the state of the art of assessing, have to be a mythical value at best. Nonetheless, the Legislature, by specifying that a survey may be conducted both before and after December 31 and that values shall be estimated on or before the first Monday in March, has at least specified a general indication of a time framework within which to determine value. We do not think it is within the role of the reviewing courts to attempt to prescribe methodologies that may or may not be more accurate than that employed by the administrative agency whose *378 decision is being reviewed. Given the legislative imprecision as to the time of fixing value and the willingness of the commission to use a shorter study period where applicable, we think it has not been shown that the commission has applied a wrong principle or committed an error of law in refusing to use the "trending” methodology preferred by the plaintiffs. Ill Conclusion Thus, as to the first issue, we affirm the holding of the Court of Appeals in Washtenaw, but modify the Court’s judgment to strike that portion of the judgment which remands these cases to the commission. The commission’s decision in regard to Livingston and Ingham Counties is left undisturbed. The rule announced herein shall apply to all assessments conducted after January 1, 1986 and all equalizations based on those assessments. Prospective application of our holding on this issue is appropriate. As we have noted in the past, The benefit of flexibility in opinion application is evident. If a court were absolutely bound by the traditional rule of retroactive application, it would be severely hampered in its ability to make needed changes in the law because of the chaos that could result in regard to prior enforcement under the law. [Tebo v Havlik, 418 Mich 350, 360; 343 NW2d 181 (1984), quoting Placek v Sterling Heights, 405 Mich 638, 665; 275 NW2d 511 (1979).] In this case, the local governments have already collected and spent the 1982 tax levies in question; state aid, such as the school fund and revenue sharing, has already been allocated on the basis of those figures. It would represent a considerable *379 administrative burden to require recalculation of the 1982 equalized valuations, especially in light of the fact that no method currently exists for taking the creative financing effect into account. While we are not persuaded by the commission’s equal protection argument in behalf of the counties who are not involved in this case, we are sympathetic to the possibly unfair effect that such a decision might have in those counties which may also have experienced the effects of creative financing in the tax year in question. 7 Livingston and Ingham Counties argue that, in the face of a prospective ruling in this case, counties will be more hesitant in the future to challenge the practices of the commission. We are not persuaded by that argument, because the counties will indeed benefit from the prospective applica *380 tion of the rule in this case, albeit not as much as from a retrospective holding. Part of the relief sought by the counties was a correction of the commission’s view of the applicable law. We expect the commission to change its interpretation of the law on the basis of our decision, and all counties will stand to benefit in future tax years from the methods which the commission will develop and implement. Thus, it is difficult to perceive that any real disincentive to litigate will result from our decision in this case to avoid the pitfalls and possible hardships of retroactivity. Regarding the second issue, we reverse the judgment of the Court of Appeals and uphold the commission’s correct interpretation of the law and decision not to utilize the proposed trending methodology. Williams, C.J., and Levin, Ryan, Cavanagh, and Boyle, JJ., concurred with Brickley, J. Riley, J., took no part in the decision of this case. Footnotes [1] Article 9, §3 of the 1963 Michigan Constitution requires the Legislature to provide for a system of equalization. Generally, equalization at both levels seeks to achieve uniformity of property tax assessment among the cities or townships within a county, in the case of intracounty equalization, and among all counties within the state, in the case of intercounty (state) equalization. *352 The process of equalization begins with the completion of property-assessments by the local assessors. Property must be assessed uniformly at fifty percent of its true cash value. Const 1963, art 9, § 3; MCL 211.27a(l); MSA 7.27(1)(1). Local assessors are required to use the commission’s Assessor’s Manual in preparing assessments, MCL 211.721; MSA 7.40, and they use sales-ratio studies or appraisal studies, rather than assessments of all individual pieces of property, in fulfilling their function. A sales-ratio study compares the sales prices of recent typical sales within a given property classification with the prior year’s assessed values for those same parcels. An appraisal study is similar, but is used in situations where there is an insufficient number of recent sales. Appraisal studies compare actual appraisals of a sampling of properties to the previous year’s assessments. The resulting ratio is applied across the board to all properties within the classification to obtain the assessed values for the current year. After assessment is completed, the county boards of commissioners meet to perform county equalization. The commissioners must "examine the assessment rolls of the townships or cities and ascertain whether the real and personal property in the respective townships or cities has been equally and uniformly assessed at true cash value”. MCL 211.34(2); MSA 7.52(2). That determination is made through the use of an equalization study, also in the form of a sales-ratio or appraisal study, and outlined in the commission’s Assessor’s Manual, Chapter XVI. See MCL 211.721; MSA 7.40. [2] This is not to say that any disparity between an interest rate facilitated by the seller and the prevailing rate will necessarily add to the purchase price. It is possible that the advantage to both the vendor and vendee will be equal in such a situation. If there is no creative financing effect on sales price, then, of course, the method of valuation need not take it into account. [3] The tax tribunal was created by the Tax Tribunal Act, 1973 PA 186, MCL 205.701 et seq.; MSA 7.650(1) et seq. Pursuant to the act, the tribunal has jurisdiction over *367 (1) "[a] proceeding for direct review of a final decision, finding, ruling, determination, or order of an agency relating to assessment, valuation, rates, special assessments, allocation, or equalization, under property tax laws,” MCL 205.731(a); MSA 7.650(31)(a), and; (2) "[a] proceeding for refund or redetermination of a tax under the property tax laws,” MCL 205.731(b); MSA 7.650(31)(b). Matters formerly reviewable by the commission or a circuit court were made reviewable by the tribunal under the act. MCL 205.741; MSA 7.650(41). In Emmet County v State Tax Comm, 397 Mich 550, 554; 244 NW2d 909 (1976), we held that equalization matters which both arise under the property laws and were appealable to the commission prior to the enactment of 1973 PA 186 are within the jurisdiction of the tribunal. Thus, "equalization” under MCL 205.731(a); MSA 7.650(31)(a), was construed to mean only intracóunty equalization. Intercounty or state equalization, at issue in this case, is thus not administratively reviewable. [4] The constitutions of 1850 and 1908 both required assessments to be made at "cash value.” See Const 1850, art 14, § 12; Const 1908, art 10, §7. Contemporaneous decisions of this Court emphasized that the constitutional standard required assessment at "true cash value.” See, e.g., Avery v East Saginaw, 44 Mich 587, 589; 7 NW 177 (1880); Hayes v City of Jackson, 267 Mich 523, 531; 255 NW 361 (1934). In Perry v Big Rapids, 67 Mich 146, 147; 34 NW 530 (1887), this Court elaborated on the meaning of cash value in the constitution: "The Constitution requires assessments to be made on property at its cash value. This means not only what may be put to valuable uses, but what has a recognizable pecuniary value inherent in itself, and not enhanced or diminished according to the person who owns or uses it.” (Emphasis added.) [5] We quote with approval from the brief of Lapeer County. “It is plain that to 'provide for’ something is not the same as to 'define’ it. To claim the contrary, when that something is designated 'true,’ is an absurdity of language. Truth is an absolute. It simply is. Truth can give meaning to definitions, but definitions can never give meaning to truth. The [commission’s] construction would mean that the legislature is empowered by the Constitution to define truth. The obvious function of the word 'true’ is to negate that construction.” [6] The Legislature has apparently had an opportunity, in the form of Senate Bill 890 of 1982, to amend § 27(3) to make specific provision for creative financing. The bill did not pass. Judge Allen, in dissent in the Court of Appeals in Washtenaw, relying on a Senate Analysis of another 1982 bill which did not include a creative financing provision and which also did not pass, found the Legislature’s inaction on these occasions to be persuasive evidence of legislative intent not to direct the commission to exclude creative financing factors from sales prices under § 27(3). See Washtenaw, supra, pp 551-552. We cannot agree. There are a myriad of reasons why legislators may oppose or fail to vote in favor of a particular bill. Reliance upon legislative inaction in these instances, alone, is an insufficient basis for a finding of legislative intent at the time of the enactment of § 27(3). It is equally as plausible that SB 890 was intended as clarification of § 27(3) and that some legislators did not believe such clarification was necessary, the words themselves being clear. The portion of the Senate Analysis quoted by Judge Allen premised its argument on the fact that "a general downward adjustment to compensate for the use of creative financing is not permitted by the Tax Commission.” This is hardly evidence of earlier legislative intent, especially in light of the plain language of the statute to the contrary, read in a manner consistent with the constitutional mandate of true cash value. [7] The Attorney General argues that any retroactive application of a decision to allow the consideration of creative financing will ipso facto render the state equalization for 1982 nonuniform. This would result from the fact that only several counties would be able to utilize it, while others who may be similarly affected would not. We also note the Attorney General’s argument, unresponded to by the counties, that there are taxing units such as the Huron-Clinton Metropolitan Authority encompassing properties within five counties, only three of which are involved in this case, and at least one school district whose boundaries fall within two counties, only one of which is involved in this case, which would result in a situation where some taxpayers in those two taxing units would be taxed on an equalized value that reflects creative financing and the others would not. The parties seem to agree that within those counties who would receive relief through a lower sev based on consideration of creative financing, school districts would be entitled to an increase in their 1982 school aid allocation and local units an increase in revenue sharing. There has been no discussion, however, as to whether those increases would come from corresponding reductions from the remaining units of government throughout the state or an increased appropriation from the state Legislature. We are mindful of our citation in Shavers v Attorney General, 402 Mich 554, 609, n 33; 267 NW2d 72 (1978) to Robinson v Cahill, 62 NJ 473; 303 A2d 273 (1973), cert den 414 US 976 (1973), in which the New Jersey Supreme Court, in applying a prospective remedy when it declared an education financing scheme unconstitutional, stated: "The present system being unconstitutional, we come to the subject of remedies. We agree with the trial court that relief must be prospective. The judiciary cannot unravel the fiscal skein.”