The Counties of King, Grays Harbor, and Pacific appeal from a summary judgment in their declaratory judgment action challenging the Department of Revenue's method of valuing forest land for purposes of 1978 property tax assessments. The Thurston County Superior Court awarded summary judgment in favor of the Department and the timber companies whose lands were affected. We affirm.
The facts involve arcane principles of forest land valuation. Each county assessor has the duty of ascertaining values for forest land so a tax can be collected from those landowners as well as from the owners of other kinds of property within each county. In 1971, the Legislature passed a law creating a new system of valuing forest lands. RCW 84.33. In RCW 84.33.040, the Legislature exempted the timber itself from ad valorem taxation; instead, RCW 84.33.071 imposes an excise tax on the timber at the time it is harvested.
As for the land itself, "forest land" is defined in RCW 84.33.100(1) as meaning land "which is primarily devoted to and used for growing and harvesting timber and means the land only." Pursuant to RCW 84.33.120(1), the Department was required, as of January 1 of each year, to
determine the true and fair value of each grade of bare forest land and [to] certify such values to the county assessors. Such values shall be determined on the basis that the only use of the land is for growing and harvesting timber, and other potential uses shall not be considered in fixing such values.[1]
*619The forest land values are to be adopted by rule according to the Washington administrative procedure act, RCW 34.04. County assessors use the values thereby established in implementing their ad valorem taxation process.
From 1972 through 1976, the Department established forest land values for ad valorem tax purposes by means of the abstraction appraisal method.2 For 1977, the Department substituted a new method of valuation, which we will call the weighted average approach. It combined 75 percent of the value achieved by abstraction analysis with 25 percent of the value computed by multiple regression analysis. The resulting weighted average value was $82 per acre of average forest land in western Washington, as published in WAC 458-40-19100.3 That WAC was challenged in a prior action in Thurston County Superior Court cause 56251 which resulted in a decision in 1978 that the 1977 forest land values were unfair, and a remand for a reappraisal of values. The Department then used the abstraction method to arrive at a revised value for 1977 of $66 per acre.
Before trial in cause 56251, the Department was required to determine forest land values for 1978. Mr. Hoffmann, the forest tax supervisor, used the abstraction method and arrived at a figure of $132 per acre. The Director of the Department of Revenue, Mr. Hodde, believed that such a drastic increase in appraised value from one year to the next was "absolutely unreasonable," and ordered Mr. Hoff-mann to "take another look at it." The upshot was a trend*620ing formula, in which the valuation forester, Mr. Smith,
divided the total purchase price by the total acreage of all sales contained in the forest land sales data base for 1978 to determine an average sales price per acre; performed the same calculation with respect to the sales data for 1977; compared those two average sales price per acre figures and derived a percentage adjustment figure of 4.01% for western Washington.
In setting values for 1978, the Department first applied the trending factor to its original 1977 weighted average value of $82 and arrived at a value of $85 for an acre of forest land; but after the court in cause 56251 disapproved of the weighted average approach and the abstraction method was used to revise the value for an acre of land to $66, the Department applied the 4 percent trending factor to the revised base and arrived at a value of $69 per acre for 1978.
In this case, the Counties sought a declaratory judgment that the 1978 valuation in WAC 458-40-19101 "exceeds the statutory authority of the agency" (RCW 34.04.070), because it is less than the true and fair value as required by RCW 84.33.120.
A summary judgment should be granted pursuant to CR 56(c) if the record before the court shows that there is no genuine issue as to any material fact. A summary judgment may rest on a presumption in the absence of prima facie evidence to overcome it. Bates v. Bowles White & Co., 56 Wn.2d 374, 353 P.2d 663 (1960); Hoskins v. Kirkland, 7 Wn. App. 957, 503 P.2d 1117 (1972). A presumption of validity of the valuation of property is built into RCW 84.40.0301(1):
Determination of value by public official — Presumptions — Review—Revaluation. (1) Upon review by any court, or appellate body, of a determination of the valuation of property for purposes of taxation, it shall be presumed that the determination of the public official charged with the duty of establishing such value is correct but this presumption shall not be a defense against any correction indicated by clear, cogent and convincing *621evidence.
(Italics ours.) Thus, the issue is whether the Counties have created a genuine issue of material fact as to the 1978 valuation by presenting a prima facie case with clear, cogent, and convincing evidence that would overcome the presumption of validity. Cf. Chase v. Daily Record, Inc., 83 Wn.2d 37, 43, 515 P.2d 154 (1973) (if the plaintiff "has offered evidence of a sufficient quantum to establish a prima facie case, and the offered evidence can be equated with the standard or test of 'convincing clarity' . . ., the motion for summary judgment should be denied").
We first consider the Counties' argument that a material issue of fact exists regarding the validity of the trending method of forest land appraisal. Even crediting the doubts regarding that method, however, it would only be of concern if it produces a land valuation not fair and true as required by the statute. See Xerox Corp. v. King Cy., 94 Wn.2d 284, 617 P.2d 412 (1980). "True and fair value" states a standard by which valuation is to be determined, but it makes no single method of appraisal mandatory. Pier 67, Inc. v. King Cy., 78 Wn.2d 48, 469 P.2d 902 (1970). An assessor, or the entity otherwise charged with establishing value, has broad discretion to determine the proper method of appraisement. Chief Seattle Properties, Inc. v. Kitsap Cy., 86 Wn.2d 7, 541 P.2d 699 (1975). Here, although Mr. Hoffmann and Mr. Conklin, an economic analyst with the Department, were strong advocates of the abstraction method and the multiple regression method, respectively, both agreed with Mr. Hodde's characterization of the 1978 valuation obtained by trending as reasonable in the circumstances.
The Counties also seize upon a memo written to Mr. Hodde on October 26, 1977 by Mr. Conklin and Mr. Demo-rest, the assistant director of the Department's Forest Tax Division, in which they said that trending's "only weakness is that it reflects the change in the per acre value of land and immature trees rather than of land alone.” Appellants *622argue that any trending valuation is not confined to the bare forest land alone as required by RCW 84.33.100(1) and .120(1), and therefore exceeds the Department's statutory authority. We observe in the depositions, however, that Mr. Conklin assumed, and Mr. Demorest was of the general opinion, that the 1977 and 1978 data bases underlying the trending formula were reasonably comparable, and that this similarity obviated the possible weakness. Although neither of them was certain of this comparability, Mr. Hoffmann was able to state that the bases differed only by approximately 1 percent. Mr. Hodde was aware of the staff study so concluding. Mr. Demorest said in his deposition that trending is a valid technique for setting forest land value, and that the values derived by applying the 4 percent trending figure to the 1977 values produced true and fair results. Mr. Conklin, while stating trending was not the best procedure, said it produced a reasonable estimate of true and fair forest land values for 1978.
The Counties emphasize the following answer by Mr. Hoffmann in his deposition:
Q. I want to know whether you believe that the values that were adopted by the Department of Revenue using the 4.01 percent accurately reflect the true and fair value of bare forest land in 1978?
A. Personally, in my own personal opinion, I do not.
The statement, however, must be read in light of the entirety of Mr. Hoffmann's deposition, in which he came out as a strong advocate of the abstraction valuation technique. His seemingly unequivocal refutation of trending is tempered by his bias in favor of abstraction as the best method. Indeed, he stated later in his deposition that it was reasonable to adjust the 1977 values for 1 year based on the 4 percent increase if the data showed an average increase of approximately 4 percent for each year from 1972 through 1977. Even though he did not state whether by a reasonable value he meant a true and fair value, the total effect of Mr. Hoffmann's deposition and affidavit in the record does not create a prima facie case to overcome the statutory pre*623sumption.
Appellants point out that Mr. Smith stated on deposition that the trending method was only accurate enough to demonstrate that forest land values were rising or falling but not to pinpoint the degree of change. The Counties stipulated in superior court, however, that Mr. Smith's statement, taken alone, "is not a material issue of fact that would affect summary judgment in this case." They should be held to the stipulation in this court, inasmuch as we, like the trial court, do not find any other basis for denying summary judgment. Although some of the Department's valuation experts believed that methods other than trending were preferable means of valuing forest lands, the record reflects that it produced true and fair values for 1978. The Counties have failed to present a prima facie case of clear, cogent, and convincing evidence to overcome the presumption of validity accorded to the Department's 1978 valuations. Thus, there is no genuine issue of material fact as to whether the agency exceeded its statutory authority in its determination of true and fair values.
Finally, the Counties contend the trial court improperly considered evidence outside the record in the form of Mr. Hoffmann's testimony in the prior case, cause 56251, to which the court alluded in its letter opinion granting summary judgment.4 The error, if any, is harmless in that the letter opinion does not reflect anything which is not appar*624ent from Hoffmann's deposition in the record.
The summary judgment is affirmed.
Petrie and Worswick, JJ., concur.
RCW 84.33.120 has been amended since the litigation in this case.
The abstraction method is described in the appellants' opening brief as follows:
"In essence the abstraction method of valuing bare forest land involves two steps: (1) collection of a data base of forest land sales and (2) analysis of that data to eliminate that portion of the sales prices attributable to trees rather than land. The data base is developed by collecting all of the sales of forest land in the state of Washington for the five (5) year period preceding the appraisal date. In other words, the data base for the 1978 assessment would include all sales of forest land between January 1,1973 and late 1977."
For purposes of this opinion, we will refer only to the valuation of average forest lands, and not to the valuations of lands considered to be above or below average in terms of timber-growing capacity and accessibility.
The court's challenged reference to Mr. Hoffmann's testimony is as follows: "After reading Mr. Hoffmann's deposition the court concludes that he is not really critical of the trending technique applied in this case. The court knows from having read his deposition and additionally, from having heard him testify at great length in Arden Tree Farms Inc., et al, vs. the State of Washington (Thurston County #56251), that he is a proponent of the abstraction process for obtaining values equivalent to true and fair value for bare forest land. It is obvious that 'trending' is not his favorite method for obtaining values but I do not conclude from his deposition that he is of the belief that it is not a valid appraisal method or that it was incorrectly applied in this case. Mr. Hoffmann does testify that trending is a valid appraisal method when making a one year adjustment from a known value. While he may have some difficulty accepting the court approved values for bare forest land in 1977 as the known value that is a figure that cannot be impeached in this trial."