The plaintiff taxpayer brings this appeal under RSA 541:6 and RSA 76:16-a, V (Supp. 1986) from an order of the New Hampshire Board of Tax and Land Appeals denying an abatement of tax on a Somersworth shopping center for the tax year 1982. The taxpayer claims that the board committed five errors: in failing to accord adequate significance to a judgment of the superior court requiring abatements of taxes on the same property for the two prior years; in failing to assign adequate weight to the income method of valuation; in erroneously rejecting the taxpayer’s appraisal evidence prepared in reliance upon that method; in ruling against the weight of the evidence; and in failing to support its decision with adequate findings of fact. We find no merit in these assignments of error and affirm.
The property in question comprises a strip-type shopping center and three separate commercial buildings situated on nineteen acres of land, all of which the taxpayer bought in 1974 for $2,025,000. In response to earlier petitions under RSA 76:17 (Supp. 1986) seeking abatements of taxes for the tax years 1980 and 1981, the superior court ordered relief after finding that the fair market value of the property was $1,661,854 in 1980, and $1,783,403 in 1981. While that proceeding was pending, the defendant employed an independent firm to do a general reappraisal throughout the city, which resulted in a valuation of the property at $2,643,300 for the tax year 1982. The taxpayer disputed this figure, and when the city assessors refused to abate the resulting tax under RSA 76:16, the taxpayer sought relief, not from the superior court under RSA 76:17 (Supp. 1986) as in the earlier cases, but from the board of tax and land appeals under RSA 76:16-a (Supp. 1986).
Because of the general reappraisal, the parties stipulated that property in Somersworth for the tax year 1982 was appraised at one hundred percent of fair market value. Therefore, in order to obtain relief on the ground that its property was assessed “at a higher percentage of fair market value than the percentage at which property [was] generally assessed in the [city],” Appeal of Town of Sunapee, 126 N.H. 214, 217, 489 A.2d 153, 155 (1985), the taxpayer had the burden to prove that the city’s valuation exceeded fair market value.
In its effort to carry this burden, the taxpayer presented the testimony of an expert appraiser who had valued the property using the capitalization of income method. The board also had before it four other appraisal reports, two of them likewise based upon the income method, and two reflecting the cost approach to valuation.
The board first denied relief by an order that was later vacated by this court because the board had purported to act without a quorum. Appeal of Net Realty Holding Trust, 127 N.H. 276, 497 A.2d 865 *797(1985). On remand, a quorum of the board considered the case on the original record, see Appeal of Seacoast Anti-Pollution League, 125 N.H. 708, 716, 484 A.2d 1196, 1202 (1984); RSA 541-A:19 (Supp. 1986), and when the board again ruled against the taxpayer, this appeal followed.
As its first issue, the taxpayer claims that the board erred in its treatment of the legal significance of the superior court judgment ordering partial abatements for the two prior tax years, 1980 and 1981. Although the board received the record of the superior court’s judgment into evidence and likewise admitted the master’s report on which the judgment was based, the board stated in its opinion that it “determined not to give . . . great weight [to the earlier decision] due to errors in the court’s calculations and methodology.” While the taxpayer believes that such treatment of the earlier judgment and findings was somehow mistaken, the taxpayer’s several attempts to formulate its position suggest how elusive the exact nature of the supposed error really is.
To begin with, we find no plausible ground even to allege error when we consider the argument of the taxpayer’s counsel before the board. There, counsel first argued that “the prior findings of the Superior Court certainly deserve to be considered ... with respect to the issue of the appropriate fair market value to be applied to this property.” This, as we read the record, is exactly what the board did.
The taxpayer suggests a second formulation of the issue when it argues in its brief that the board’s decision “has the effect of increasing the fair market value of the subject property in one year’s time by almost one million dollars, even though no additions, alterations or improvements were made to the property . . . [,] a result ... at odds with both common sense and economic reality.” This statement will not do, however. No one pretends that there was any such increase in market value from April 1, 1981, to April 1, 1982. The board simply decided that the superior court’s valuations for the earlier years were wrong and that the taxpayer had not proven that the city’s assessment of its property for 1982 was disproportionate.
The taxpayer states a third position when it argues in its brief that “the board erred when it ruled that it was not bound by the . .. superior court’s finding as to the value of the . . . property in 1980 and 1981,” a position which the taxpayer raised when its counsel argued before the board that it would be “improper” to “relitigate” the superior court’s decision. As we understand this argument, the taxpayer means to assert two related propositions: first, that each party to the superior court’s earlier determinations of fair market value for 1980 and 1981 must accept those determinations as conclu*798sive for those respective years; and second, that in a later proceeding the value for 1982 could be found to differ from the value in 1981 only to the extent that the evidence established intervening changes in the factors that determine such value. Because it may be assumed that no evidence in this case justified finding an $859,897 increase in value, it would follow from this argument that the city’s 1982 assessment was disproportionate.
By the first proposition the taxpayer seeks to apply the principle of collateral estoppel to bar the defendant from relitigating in a later proceeding a fact that the defendant actually litigated in the earlier action. See Caouette v. Town of New Ipswich, 125 N.H. 547, 554, 484 A.2d 1106, 1111-12 (1984). By the second proposition, however, the taxpayer would carry the traditional collateral estoppel rule one step further, to bar a fact-finder in a later proceeding from reaching a conclusion inconsistent with a relevant evidentiary fact as determined in the earlier proceeding. In urging us to adopt these legal propositions, however, the taxpayer runs up against Appeal of Public Service Co. of New Hampshire, 120 N.H. 830, 424 A.2d 197 (1980). There we held that in a later proceeding in which the value of property may be in issue, the judgment in an earlier tax appeal does not conclusively establish the value of that property at the earlier date, even as between the parties to the earlier appeal. Id. at 832, 424 A.2d at 198.
While the taxpayer has not overlooked Public Service, it attempts to distinguish the case on the ground that the earlier proceeding considered in that case had determined only that the earlier year’s assessment was not disproportionate, whereas the earlier proceeding considered in the present case did find disproportionality, which it predicated on a specific finding of fair market value. The taxpayer thus argues that the earlier proceeding in Public Service raised no estoppel or preclusion bearing on the issue of value because it never determined specifically what the value of the property was. In this case the superior court did reach a specific finding of value which, the taxpayer submits, should be given preclusive effect to the extent of its later relevance. See Lethin v. Dept. of Revenue, 278 Or. 201, 205-06, 563 P.2d 687, 689-90 (1977).
The distinction is foreclosed, however, by the reasoning in Public Service. This court did not hold that there was no later preclusion because there had been no earlier specific finding; it held there was no preclusion because any earlier finding of value in a tax abatement case is not a finding on a matter in issue, i.e., disproportionality, but rather on a matter merely in evidence, e.g., fair market value, as to which there is no issue preclusion under the traditional *799concept of collateral estoppel. Public Service, supra at 833, 424 A.2d at 199. This reasoning was underscored by the court’s reliance on Winnipiseogee etc. Co. v. Laconia, 74 N.H. 82, 85, 65 A. 378, 380 (1906), where a finding of specific value in an earlier tax abatement proceeding was held to lack any conclusive force as between the same parties in an abatement proceeding covering a later year. The taxpayer can not prevail on this issue, therefore, unless we overrule Public Service, which we are not about to do.
We are aware, to be sure, of the criticism that has been directed to the distinction between matters in issue and matters in evidence, for purposes of applying the rule of collateral estoppel. See Restatement (Second) of Judgments § 27, comment j at 260-61 (1982). The taxpayer has not argued that we should discard this distinction, however. And even if we were to abandon it, we would continue to follow the rule that a determination of value under an earlier abatement petition does not bind the parties even as to the value in the earlier year, in a later abatement proceeding as to a later year.
We so limit the effect of such a determination simply because we are convinced that the ideal of fair and proportionate taxation can be approached only through a constant process of correction and adjustment of assessments. RSA 75:8, indeed, requires selectmen and assessors to engage in just such continual revision by examining appraisals for error each year. This candid statutory recognition of the need for constant corrective effort is antithetical to any legal doctrine that would invest a given valuation of property with preclusive effect for the future, so that any error would affect subsequent assessments indefinitely. Where a single taxpayer was the losing party in the first proceeding, the preclusive effect of such an error could be disastrous to him. Where the public was the loser, preclusive effect would just as certainly perpetuate unfairness. The risk of such results is too high a price to pay for the judicial economy and consistency that collateral estoppel is designed to provide.
We therefore affirm the rule expressed in Appeal of Public Service Co. of New Hampshire, 120 N.H. at 832, 424 A.2d at 198. In the case before us, the board properly applied this rule and accorded only such persuasive force to the earlier judgment as it thought the judgment deserved. We accordingly find no merit in the taxpayer’s position on this first issue.
The taxpayer next raises the related claims that the board generally gave inadequate weight to the income method of valuation and specifically erred in rejecting the testimony of the taxpayer’s *800appraiser, on the ground of his supposed errors in applying that method in his appraisal. The first general claim need not keep us long. The capitalization of income is one of four accepted approaches to real estate valuation, Dartmouth Corp. of Alpha Delta v. Hanover, 115 N.H. 26, 28, 332 A.2d 390, 391 (1975), which the board by no means denied. In its written decision the board simply took note of the conditions that may qualify the soundness of applying the income approach in a given case, including the risk that an appraiser will fallaciously capitalize actual income figures that fall below the level of income obtainable in the market. See Demoulas v. Town of Salem, 116 N.H. 775, 782, 367 A.2d 588, 593 (1976). There is nothing exceptionable in such observations. It is obvious that the taxpayer’s real complaint is with the board’s rejection of the conclusions offered by the taxpayer’s expert, who based his appraisal on the income approach.
The board rejected the taxpayer’s appraisal because, inter alia, it was not satisfied with the expert’s figures stating the income to be capitalized or with the capitalization factor applied to that income. The board found that the taxpayer had not demonstrated that the income figures represented the rental income available on the market as distinguished from contract rent, that is, actual income derived from insufficiently profitable leases. The board rejected the capitalization factor employed by the expert because it reflected the depreciation of land as well as the depreciation of improvements constructed on the land. We find no error on either point.
Although the taxpayer’s expert claimed that the taxpayer’s rental income represented market levels, he failed to present any evidence of comparable rentals to substantiate his claim. In fact, the board noted that the master’s report in the superior court proceeding covering the years 1980 and 1981 clearly indicated that at least some of the rental rates had fallen below the market level, as long-term leases had grown increasingly disadvantageous to the taxpayer. The board also considered the wide disparity in lease charges per square foot among the various tenants in the shopping center in question; those figures revealed no pattern related to the size of premises or other relevant valuation factors. Accordingly, the board was left to infer that some, at least, of the disparate rates fell below the market level. These were legitimate grounds to reject the taxpayer’s application of the income method.
There were equally sound grounds to reject the expert’s conclusion as reflecting a capitalization formula that depreciated land. The expert’s own testimony on cross-examination is instructive:
*801“Q Now, you’re using that 15 percent to capitalize the entire property; is that correct?
A That’s correct. The entire income.
Q [. . .] Now, does this property consist of both land and buildings?
A It does, sir.
Q Then aren’t you depreciating land?
A No, sir.
Q Well, if you have a three percent as part of your total depreciation rate —
A I come down to a value, a total value of the real estate of $1,135,000.
Q [...]
A From that $1,135,000 I would allocate a land value if the building were demolished on the land. But in the case we have land and building, improved; we have land improved with a building. The two values merge into one unit, and any separation is arbitrary and is not susceptible of proof.
Q And that’s a position that most appraisers take, then, depreciated land?
A It’s the position I take.”
The expert thus denies that he depreciated the land. But when he says that the values of land and improvements “merge” into the “entire property” for which he allows depreciation in the capitalization formula, the board has every reason to find that the expert did indeed depreciate the taxpayer’s land. As against this treatment by the expert, the board temperately observed that
“[l]and is a non wasting asset and will have a residual value at the end of the building life; in fact, historically, land has generally appreciated with the passage of time. It is improper to provide for its recapture in the income-capitalization technique employed by the appraiser for Net Realty Holding Trust in this case ....”
That is, the board rejected a valuation methodology that implied land would have no value at the end of the depreciable life of improvements constructed on it. We see nothing unreasonable about the board’s position, and the taxpayer has cited no authority contrary to it.
*802Before leaving this third issue, we should note, albeit briefly, the argument raised by taxpayer’s counsel in an effort to rescue the witness’s testimony. In his brief, counsel submits that the income considered by the expert reflects only the improvements to the land, so that depreciation of the entire income-producing aspect of the property cannot result in depreciation of land. Besides suffering from lack of evidentiary support, this argument could not under any circumstances carry the taxpayer anywhere but out of the frying pan into the fire. If the argument were well taken, the result would be nineteen acres of commercial land to which no income would be attributed, which accordingly would be left with zero value for tax purposes. That result would not rehabilitate the plaintiff’s witness, and there was, in sum, no error in rejecting the expert’s purported application of the income method of valuation.
In the wake of this discussion there may seem little call to comment on the taxpayer’s fourth claim, that the board’s decision is inconsistent with the weight of the evidence. On this point we should note only that the taxpayer relies upon two Vermont cases, Jeffer v. Town of Chester, 138 Vt. 478, 417 A.2d 937 (1980) and Welch v. Town of Ludlow, 136 Vt. 83, 385 A.2d 1105 (1978). Each explains the Vermont rule, that in a tax abatement case the taxing authority’s appraisal has the status of an evidentiary presumption. Although that presumption places on the taxpayer a burden of going forward with evidence, the burden is satisfied and the presumption disappears upon the “introduction of credible evidence fairly and reasonably tending to show that the property was assessed at more than fair market value or that the listed value exceeded the percentage of listed value actually applied to the general mass of property in the community.” Welch, supra at 86, 385 A.2d at 1107.
While the taxpayer argues that application of such a rule would entitle his expert witness to carry the day, it is enough to say in reply that this case does not present an opportunity to consider the merits of the Vermont rule. As expressed in the passage quoted, it would require “credible evidence fairly and reasonably tending to show” over-assessment before the presumption would disappear. It is obvious in this ease that the board did not find the taxpayer’s evidence credible, or reasonably tending to show that the city’s appraisal produced disproportionate taxation. If there is any value in articulating a Vermont presumption, it will have to wait for another case. See N.H. R. Ev. 301.
A final issue remains, the taxpayer’s claim that the board’s written decision fails to measure up to standards of administrative law, which require findings of basic fact in support of ultimate conclusions. See N.H.-Vt. Health Serv. v. Comm’n of Ins., 122 N.H. 268, 273, *803444 A.2d 508, 510 (1982); Society for Protection of N.H. Forests v. Site Evaluation Comm., 115 N.H. 163, 173, 337 A.2d 778, 786 (1975); RSA 541-A.-19 (Supp. 1986). The claim has no merit.
The sufficiency of an administrative agency’s findings must be judged, necessarily, in relation to the ultimate relevant conclusion that the agency draws. In this case the board ultimately concluded that the taxpayer failed to carry its burden to demonstrate the disproportionality of the 1982 assessment. Given the parties’ stipulation of a 1:1 ratio of assessed valuation to fair market value generally, the taxpayer’s burden to prove disproportionality was a burden to prove that the city had assessed the taxpayer’s property above the fair market level. Consequently, the board’s finding that the taxpayer had failed to prove disproportionality required findings in the nature of reasons to reject the taxpayer’s evidence of fair market value. (A case like this is thus distinguishable from one in which the board does find disproportionality, which presupposes findings of specific value, see Stevens v. City of Lebanon, 122 N.H. 29, 440 A.2d 451 (1982), and from one in which the taxing authority concedes that the assessment is wrong, so that the board must make some finding about specific value. See Appeal of Loudon Road Realty Trust, 128 N.H. 624, 517 A.2d 843 (1986).)
It should be clear from our discussion of the board’s reasons for rejecting the taxpayer’s appraisal evidence that the board did state sufficiently detailed reasons to indicate why the taxpayer had failed to carry its burden. The board found that the taxpayer had failed to offer anything but conclusory allegations that its income figures reflected market as distinguished from contract rents; it accepted evidence that some rents were below market level; and it found persuasively that the taxpayer’s purported application of income capitalization analysis was flawed further by the depreciation of land. While these examples do not exhaust the board’s findings or its articulated reasoning, they are sufficient to support the final conclusion. The taxpayer is therefore flatly mistaken in analogizing the board’s findings to “a bald conclusion that the taxpayer has failed to satisfy his burden of proof,” Appeal of Harvey Road Assoc’s, 122 N.H. 24, 26, 440 A.2d 11, 13 (1982), which we found inadequate most recently in Appeal of Loudon Road Realty Trust supra. The findings are adequate; there is no apparent error.
Affirmed.
Thayer, J., did not participate in the decision; the others concurred.