OPINION

PETERSON, Judge

After LSR Acquisition Corp. (LSR), a wholly owned subsidiary of Landry's Seafood Restaurants, Inc. (Landry's), acquired a majority of the shares of Rainforest Café, Inc., (Rainforest I), LSR and Rainforest I merged to form respondent Rainforest Café, Inc. (Rainforest II). Appellants, State of Wisconsin Investment Board (SWIB) et. al., owned shares of Rainforest I. Appellants dissented from the merger and asserted dissenters' rights under Minn. Stat. § 302A.473, subd. 6 (2002), seeking payment for the fair value of their shares, and respondent initiated an appraisal proceeding under Minn. Stat. § 302A.473, subd. 7 (2002), seeking a determination of the fair value of Rainforest I shares before the merger. Following a trial, the district court determined that the fair value of each share was $3.25. On appeal, appellants argue that their shares are worth more than $3.25 per share and that the district court erred by relying upon only some of the evidence presented by the parties and by failing to consider and reconcile conflicting expert testimony. We affirm.

FACTS

Rainforest I owned, operated, and licensed three categories of large, high-volume, restaurant facilities in what is known as the "eatertainment" industry. The facilities reflected a rainforest theme and included mall units, which were located primarily in shopping malls and contained between 300 and 450 restaurant seats; icon units, which were located in high-traffic tourist areas and contained 400 to 600 restaurant seats; and international units, which were located outside the United States and paid Rainforest I licensing fees and royalties. Rainforest I became a publicly traded company in April 1995.

Between 1994 and 2000, Rainforest I opened approximately 30 units across the United States and licensed additional units internationally. The units typically opened to high business volume followed by a decline in same-store sales after one year due to a lack of repeat business. The company implemented incentive programs, promotions, menu changes, and advertising campaigns to reduce the decline in repeat business. These efforts continued through 1999 and 2000, but same-store sales declines continued. In its quarterly and annual public filings, the company reported same-store sales declines for seven consecutive quarters. Same-store sales declined 12.7% in the second quarter of 2000 compared to the second quarter of 1999 and 13.4% in the third quarter of 2000, compared to the third quarter of 1999. However, in spite of the declining same-store sales, overall revenues of the company increased from 1995 through 1999 and into 2000, due to new store openings.

During its expansion, the company entered into long-term leases that included a base rent, plus additional rent computed as a percentage of a restaurant's sales. The leases had no provisions for early cancellation or termination by the company. By early 2000, Rainforest I's long-term lease obligations totaled approximately $175 million, but the company continued to generate positive cash flow. The company implemented cost-control measures to improve financial performance, but its ability to open new stores was limited by restrictions placed on it by certain leases and by its ability to find high-traffic tourist locations.

In 1999, Rainforest I's CEO, Lyle Berman, began seeking a buyer for Rainforest I. Lakes Gaming, Inc. tendered an offer of $4.62 per share in December 1999. At that time, Berman was chairman of the board of directors for both Rainforest I and Lakes Gaming, and Rainforest I shares were trading at four dollars per share.

In February 2000, Landry's offered to purchase Rainforest I for a combination of Landry's stock and cash. Landry's stock price at the time made the offer worth $5.23 per Rainforest I share. SWIB publicly opposed the Landry's offer, and the publicity generated by SWIB's opposition convinced Rainforest I management that they would not receive sufficient shareholder votes to approve the purchase. The Landry's offer was terminated in April 2000. By that time, the offer's value had fallen to a little over four dollars per share as a result of a decline in Landry's stock price.

In the summer of 2000, Rainforest I closed one of its stores, and in July 2000, it took a $101.5 million asset-impairment and store-closing charge, which reflected the declining performance of its mall location stores and the exit costs for closing the one store. In the summer and fall of 2000, the company unsuccessfully attempted to renegotiate its leases. Rainforest I lost key members of its management team and a number of other employees, and management contemplated closing 20 additional stores. During the fall of 2000, the company contemplated a bankruptcy filing to terminate its leases for stores that management anticipated would soon become unprofitable. The company retained bankruptcy counsel, but did not disclose the possibility of bankruptcy to shareholders.

In September 2000, Landry's proposed a two-step merger that LSR would commence with a cash tender offer to acquire 100%, but not less than a majority, of Rainforest I common shares for $3.25 per share. If LSR acquired a majority of Rainforest I common shares, Rainforest I shareholders would vote on a merger, and after approval of the merger, any Rainforest I shares not tendered to LSR, except shares of shareholders that exercised and perfected their right to dissent from the transaction, would be exchanged for $3.25 per share. On September 26, 2000, the Rainforest I board approved a resolution in favor of the Landry's offer. Rainforest I stock was trading in the two dollars per share price range for several weeks before Landry's second offer. After the offer was announced, the stock price jumped one dollar, but the price never rose above $3.25 per share.

SWIB publicly opposed the offer. In response, Rainforest I issued a press release inviting a better offer, but no other offer was made. Approximately 60% of Rainforest I shareholders accepted Landry's tender offer and tendered their shares to Landry's for $3.25 per share. On December 1, 2000, 80% of Rainforest I shareholders either agreed to the merger of LSR and Rainforest I or did not pursue their dissenters' rights. Appellants did not vote their Rainforest I shares in favor of the Landry's offer and asserted and perfected their statutory dissenters' rights.

On April 27, 2001, respondent petitioned the district court under the Minnesota Business Corporation Act to determine the fair value of Rainforest I stock. During a trial to the court, appellants' experts testified that the fair value was $6.10 per share and respondent's experts testified that the fair value was three dollars per share.

Christopher Mercer, appellants' expert, analyzed the value of Rainforest I stock using three different methods. First, Mercer determined that the net asset value on the valuation date was $3.81 per share (excluding Rainforest I's excess assets of $22,130,000), but considered this value to be only a benchmark in determining the fair value of the shares and gave it zero weight. Second, Mercer used the guideline-company method to arrive at a share value. This method compares the subject company to similar publicly traded companies to arrive at valuation conclusions, typically by determining various multiples of the guideline company's stock, and then applying an adjusted multiple to the subject company. Under Mercer's guideline-company-method analysis, Rainforest I stock was valued between $4.80 per share and $4.91 per share, without including excess assets. Mercer assigned weights of 37.5% and 25%, respectively, to these values. Finally, Mercer valued the shares by applying the discounted-cash-flow (DCF) method, the most commonly used income valuation approach. Using Landry's financial projections, Mercer valued the company at $5.23 per share and assigned a weight of 37.5 percent to this valuation. After applying the different weights to the different valuations, Mercer arrived at a final share value of $4.98 as of the valuation date. After taking into consideration Rainforest I's excess assets, including cash and a federal income-tax receivable, Mercer calculated the value of each share to be $5.95. Finally, Mercer calculated a share value after including the value of severance payments incurred in the merger, as well as the tax benefits that accrued to Landry as a result of the merger, and concluded that the fair value of Rainforest I stock was $6.10 per share.

In an initial valuation conclusion rendered on December 27, 2000, approximately two years before his final report, Mercer opined that the market pricing of Rainforest I stock was "`rational' given declining fundamentals," and that "[f]air value [was] likely in the range of the transaction price." He also recognized that earnings for the third quarter of 2000 were "significantly augmented" by the lower depreciation charges following the asset-impairment charge taken by the company. Mercer's initial valuation report also included a section on how to achieve a higher price than the $3.25 per share offering price. Mercer suggested that the company must prove the "reality" of the third quarter earnings, overcome the issue of same-store sales declines, and overcome the taint or questionable concept of themed restaurants.

Don Nicholson, respondent's rebuttal expert, opined that Mercer's final valuation report was inaccurate because Mercer (1) did not interview Rainforest I's or Landry's management; (2) failed to consider Rainforest I financial information for October and November 2000; (3) used guideline companies that bore little resemblance to Rainforest I; (4) failed to consider the public stock price of Rainforest I as an indicator of the market's valuation; and (5) used financial projections prepared by Landry's for its lenders and did not take into account changes Landry's made based on later information. Nicholson opined that Mercer's use of Landry's projections was inappropriate because it assumed a company under Landry's management and control with the availability of Landry's resources — which were not available to Rainforest I.

Respondent's expert, Donald Erickson, determined the fair value of Rainforest I shares using the DCF method and applying it to four different scenarios based on Rainforest I's financial projections. Two of the scenarios assumed that Rainforest I reorganized under bankruptcy protection and two did not. The other variables were: (a) variations in the number of store closings; (b) variations in the outlook for Rainforest I's Fisherman's Wharf store; (c) variations in the outlook of Rainforest I's foreign operations; and (d) variations in the discount rate.

Erickson considered using a market-based comparative approach but concluded that Rainforest I was unique, particularly with regard to the magnitude of same-store sales declines, such that comparing Rainforest I's performance to guidelines companies would yield unreliable results. Erickson also rejected book value as an indication of fair value, in part because it does not take into consideration contingent liabilities (such as expenses related to future store-closing) that others in the marketplace consider when evaluating the company. Erickson testified that the book value as of the valuation date was in the five dollars per share range.

Richard May testified for appellants to rebut Erickson's testimony. May testified that Erickson's valuation was flawed because it projected future sales out 12 years rather than the standard five years; used an inappropriate discount rate for the level of risk associated with Rainforest I or the eatertainment industry as a whole; assumed an unreasonably low long-term growth rate; and erroneously assumed the closure of too many stores. By correcting each of Erickson's alleged errors, May testified that the minimum fair value of Rainforest I shares as of the valuation date was between $4.69 and $5.06 per share. May testified that after applying the guideline-company method to Erickson's own selected guideline companies, the fair value for Rainforest I shares was $7.23 per share. However, May also testified that the price of stock that is traded in volume in a free and active market by informed persons best reflects the consensus of the investing public. Also, both Mercer and May testified that the average control premium in a merger is approximately 40%, which is less than the 62% obtained by Rainforest I shareholders in the merger.

The district court concluded that the fair value of appellants' shares of Rainforest I stock as of the valuation date was $3.25 per share. In arriving at this value, the district court determined that the experts' conflicting testimony provided no aid in determining the fair value of the stock, finding that appellants' experts were overly optimistic and respondent's experts were overly pessimistic. The district court determined that the market-based and transactional evidence was a more reliable and relevant indicator of the stock's value. The court also found that book value was not a reliable indicator of fair value.

Appellants made no posttrial motions and now appeal from the May 2003 judgment, arguing that the district court abused its discretion by (a) failing to resolve the differences between the experts' opinions and rejecting all expert testimony introduced at trial; (b) rejecting book value as a measure of value that should set a floor for the value of the shares; (c) failing to consider comparable transactions offered by appellants' expert; and (d) ultimately determining that the fair value of Rainforest I stock was $3.25 per share.

ISSUES

1. Did the district court abuse its discretion by rejecting expert testimony and relying instead upon market-based and transactional evidence to determine the fair value of the dissenters' shares?

2. Did the district court abuse its discretion by finding that book value was not a reliable indicator of Rainforest I's fair value?

3. Did the district court abuse its discretion by rejecting the analysis of appellants' expert's analysis as overly optimistic?

4. Did the district court err in concluding that the fair value of Rainforest I stock was $3.25 per share?

ANALYSIS

On appeal from a judgment where there has been no motion for a new trial, appellate review is limited to "whether the evidence sustains the findings of fact and whether such findings sustain the conclusions of law and the judgment." Gruenhagen v. Larson, 310 Minn. 454, 458, 246 N.W.2d 565, 569 (1976). But, a motion for a new trial is not a prerequisite for appellate review of a substantive question of law that was previously considered and addressed by the district court. Alpha Real Estate Co. of Rochester v. Delta Dental Plan of Minn., 664 N.W.2d 303, 311 (Minn. 2003).

"Findings of fact, whether based on oral or documentary evidence, shall not be set aside unless clearly erroneous, and due regard shall be given to the opportunity of the trial court to judge the credibility of the witnesses." Minn. R. Civ. P. 52.01. "Findings of fact are clearly erroneous only if the reviewing court is left with the definite and firm conviction that a mistake has been made." Fletcher v. St. Paul Pioneer Press, 589 N.W.2d 96, 101 (Minn. 1999) (quotation omitted). If there is reasonable evidence to support the district court's findings, we will not disturb them. Id. An appellate court views the record in the light most favorable to the district court's judgment. Rogers v. Moore, 603 N.W.2d 650, 656 (Minn. 1999). A district court's conclusions of law are not binding on this court and are reviewed de novo. MCC Invs. v. Crystal Props., 451 N.W.2d 243, 246 (Minn.App. 1990), review denied (Minn. Mar. 27, 1990).

Upon a fundamental corporate change, such as a merger, Minnesota law allows minority shareholders to dissent from the corporate action. Minn. Stat. § 302A.471, subd. 1 (2002). Dissenting shareholders that comply with the statutory conditions for dissenting to the proposed action are entitled to receive the fair value of their shares. Minn. Stat. § 302A.473, subds. 3, 6, 7 (2002).

The court . . . shall determine the fair value of the shares, taking into account any and all factors the court finds relevant, computed by any method or combination of methods that the court, in its discretion, sees fit to use, whether or not used by the corporation or by a dissenter.

Id. at subd. 7.

[F]air value . . . means the pro rata share of the value of the corporation as a going concern. To determine fair value, the trial court may rely on proof of value by any technique that is generally accepted in the relevant financial community and should consider all relevant factors, but the value must be fair and equitable to all parties.

1. Appellants argue that the district court committed reversible error when it disregarded all of the expert testimony. Citing Thomas v. Thomas, 407 N.W.2d 124, 126 (Minn.App. 1987), which states, "[w]hen conflicting opinions of expert witnesses have a reasonable basis in fact, the trier of fact must decide who is right," appellants contend that the district court had a responsibility to resolve the conflicting experts' analysis and that its failure to do so was prejudicial and an abuse of discretion. But appellants ignore the next sentence in Thomas, which states, "The trial court is not bound by the opinion of any witnesses concerning values." Id. (citing Lehman v. Hansord Pontiac Co., 246 Minn. 1, 7, 74 N.W.2d 305, 310 (1955)). The weight and credibility of expert testimony is for the fact-finder to determine. Shymanski v. Nash, 312 Minn. 304, 308, 251 N.W.2d 854, 857 (1977). The "opinions of expert witnesses are only advisory and the [fact-finder] may weigh such evidence in the light of all the facts and opinions presented to it and draw its own conclusions." Housing Redevelopment Auth. v. First Ave. Realty Co., 270 Minn. 297, 306, 133 N.W.2d 645, 652 (1965). Furthermore, the dissenters' rights statute expressly states that the court "shall determine the fair value of the shares, taking into account any and all factors the court finds relevant . . . whether or not used by a corporation or by a dissenter." Minn. Stat. § 302A.473, subd. 7. The district court considered the experts' opinions but determined that the opinions were not helpful because they were overly optimistic or overly pessimistic, which is specifically permitted by the statute.

Appellants argue that they were prejudiced by the court's disregard of their experts' testimony because, had the court adopted their experts' testimony, the outcome would have been vastly different. See Jenson v. Touche Ross Co., 335 N.W.2d 720, 725 (Minn. 1983) (stating "before an error in the exclusion of evidence may be grounds for a new trial, it must appear that such evidence might reasonably have changed the result of the trial if it had been admitted"). Appellants confuse the concept of prejudice as applied to appellate review of a trial court's evidentiary rulings with a fact-finder's determination of the weight and credibility of expert testimony. The issue here is not the admission or exclusion of evidence; it is the weight that the fact-finder attributed to the evidence. The weight that the district court attributed to the experts' testimony, which was admitted at trial, might reasonably have changed the result of the trial, but that does not mean that attributing no weight to the testimony was prejudicial error.

Appellants next assert that the district court arbitrarily rejected the rebuttal testimony of their expert, May. Appellants argue that by using the same valuation techniques as respondent's expert, but with less pessimistic assumptions and projections, May addressed the district court's concerns about respondent's overly pessimistic analysis and, therefore, the court should have adopted May's valuation.

The district court's findings of fact and memorandum demonstrate that it did not arbitrarily reject May's testimony. The court found that May successfully pointed out the weaknesses of Erickson's valuation, which was the purpose of May's rebuttal testimony. But the court also found that both of appellants' experts were overly optimistic and that their reliance on positive third-quarter earnings was unfounded, in part because they failed to recognize that the up-tick in third-quarter earnings was due to the impairment charge taken during the second quarter of 2000 (resulting in lower depreciation expenses during the third quarter), and the boost from a temporary decrease in accrued bonus payments as a result of the departure of key employees who would eventually have to be replaced. The court also found that both of appellants' experts assumed perpetual growth but presented no evidence to support that assumption, especially in light of the company's continuing declines in same-store sales.

2. Appellants argue that the district court erred when it rejected book value as a floor value for determining the fair value of Rainforest I shares. Appellants acknowledge that book value is often an unreliable measure of a company's fair value but assert that in this case, the value of the company should not be less than the book value. See Warner v. E.C. Warner Co., 226 Minn. 565, 576, 33 N.W.2d 721, 727 (1948) (stating book value is only minor element to be considered in arriving at true value of stock and that book value is entitled to little, if any, weight in determining true value of stock).

Appellants argue that when Rainforest I took a $101.5 million impairment charge against its assets in the second quarter of 2000, it acted to reflect the liquidation value of its assets. Consequently, appellants argue, the normal criticisms against using book value are not present, and the book value after the impairment charge was taken should function as a floor for fair value. Therefore, appellants conclude, the district court erred in reaching a fair value for the shares that was below book value.

Appellants are asking this court to reweigh the evidence. The role of the court of appeals is to correct errors, not to find facts. Sefkow v. Sefkow, 427 N.W.2d 203, 210 (Minn. 1988). The district court found that book value was not a reliable indicator of fair value. Nicholson, Erickson, and May all testified that book value often bears little relationship to the price at which a company is bought or sold, primarily because book value is a measure derived from a company's balance sheet, which may not fully reflect the market value of a company's assets. Even Mercer, appellants' expert, analyzed Rainforest I's book value but gave it no weight. Also, there was evidence that at the time of the merger, Rainforest I stock never traded anywhere near book value, and caselaw from Minnesota and other jurisdictions overwhelmingly recognizes that book value is not a reliable indicator of fair value. Thomas, 407 N.W.2d at 126 (stating book value is often an erroneous figure and easily subject to manipulation); see also Beerly v. Dep't of Treasury, 768 F.2d 942, 946 (7th Cir. 1985) (stating book value is "virtually meaningless" index of share value). The district court's rejection of book value as a floor value is not clearly erroneous.

3. Appellants argue that the district court erred by failing to consider the comparable-transaction and guideline-company methods proffered by Mercer. While appellants acknowledge that the district court acted within its discretion in finding that Mercer's income approach (DCF method) for valuing the shares was overly optimistic due to its reliance on unwarranted projections of growth, appellants contend that the court's criticism does not apply to Mercer's guideline-company analysis because that methodology did not rely on projections. Therefore, appellants argue, the court's rejection of Mercer's guidelines methodology was an abuse of discretion.

Appellant's argument disregards the court's finding that the companies Mercer identified as guidelines bore little resemblance to Rainforest I. Few, if any, of the companies were in the eatertainment segment of the restaurant industry, most or all projected future growth, and none had the type of declining same-store sales that Rainforest I had. Mercer conceded that Rainforest I was unique and that it was "fundamentally different from most of the guideline companies." Contrary to appellants' argument, the district court presented a reasonable basis for rejecting Mercer's guideline-company analysis.

4. Finally, appellants argue that the district court erred in concluding that the fair value of appellants' stock was $3.25 per share. Appellants contend that after rejecting the expert testimony, there was little evidence of fair value left, other than anecdotal evidence.

The district court found that the fact that the negotiations, tender offer, and merger occurred at arms length on the open market indicated that the $3.25 share price was fair and the product of independent market forces. The court also found that Rainforest I's heavy trading volume on the NASDAQ stock exchange was indicative of fair value. The court noted that the premerger stock price of two dollars was a relevant indicator that fair value did not exceed $3.25 per share. The court also noted that Landry's offered a 62% premium over the publicly traded price, which was greater than the 40% premium typically paid. Finally, the court found that the acceptance of Landry's tender offer by 80% of the shareholders, including Rainforest I's senior management, was a relevant indicator of fair value.

Citing cases from foreign jurisdictions, appellants argue that although market price can be a relevant consideration in the valuation context, market price accurately reflects fair value only when the market has reliable and current information. Therefore, appellants contend, because the market was not aware of Rainforest I's significantly improved third-quarter results in 2000 when the stock was trading around two dollars per share (immediately before Landry's offer), the trading price did not accurately reflect the fair value of the shares, and the district court's reliance on the measure was misplaced.

But the district court found that the up-tick in the third-quarter earnings was not the result of improvements in the company's core business, but rather was the result of lower depreciation expenses associated with the impairment charge Rainforest I took in the second quarter, as well as the lower administrative costs associated with the departure of key employees who would have to be replaced. The record supports the district court's finding, and press releases reported the departure of key employees and the impairment charge to the investing public before Landry's offer. Moreover, the district court found that due to the seasonal nature of the restaurant business, the most relevant comparison compared one quarter with the same quarter in the prior year, not one quarter with the following quarter. Same-store sales declined by 13.4% between the third quarter of 1999 and the third quarter of 2000.

Appellants argue that the district court's reliance on the 62% premium paid by Landry's was erroneous because the market was not aware of the positive third-quarter 2000 performance results, which depressed the market price of the stock. Therefore, appellants contend, the fair value of the shares was higher than the two dollars per share market price, and the premium paid by Landry's was likely less than the 40% market average. But appellants' argument relies on their portrayal of the third-quarter-2000 performance as a significant improvement. As we have already determined above, the district court articulated its basis for rejecting appellants' characterization of the third-quarter-2000 performance as a significant improvement.

Appellants also argue that the district court's reliance on the market and transactional evidence eviscerates the appraisal statute. Appellants contend that there would be no need for the valuation statute if all that a court needed to do was determine whether the market price was close to the tender-offer price. Appellants mischaracterize the district court's decision. The record demonstrates that the district court considered all of the evidence presented; it did not rely solely on the market price. Minnesota caselaw recognizes market price as a relevant factor in determining the value of a corporation. Spinnaker Software Corp. v. Nicholson, 495 N.W.2d 441, 444-45 (Minn.App. 1993), review denied (Minn. Mar. 30, 1993).

Finally, appellants argue that relying on the number of shareholders who approved the transaction when determining fair value defeats the purpose of the statute, which is to protect minority interests. See MT Props., Inc. v. CMC Real Estate Corp., 481 N.W.2d 383, 388 (Minn.App. 1992) (recognizing legislative aim of statute is to protect minority dissenting shareholder). We agree. The fact that a substantial majority of Rainforest I shareholders voted to approve the merger is not a relevant factor in determining the fair value of the corporation because the essence of appellants' claim is that the shareholders who approved the merger incorrectly determined the fair value of the shares.

But we conclude that the remaining factors considered by the district court provide a sufficient basis for its determination of fair value. Market price is a generally accepted "technique" employed by the financial community as a relevant measure of value when a stock is heavily traded. See Follett, 615 N.W.2d at 290 (stating court may rely on proof by any technique generally accepted by financial community in determining fair value); see also Beerly, 768 F.2d at 946 (stating market price of thinly-traded stock may not be reliable guide to current value); BNE Mass. Corp. v. Sims, 588 N.E.2d 14, 21 (Mass App. Ct. 1992) (stating when valuing corporation, courts should use purchase price where that price was negotiated in arms-length transaction). Given the evidence in this case, the district court's reliance on the market price of Rainforest I shares is not clearly erroneous. The record supports the district court's findings of fact, and those findings support its conclusion that the fair value of the dissenters' shares is $3.25 per share.

DECISION

In determining the fair value of the dissenters' shares, the district court did not err by rejecting the valuation opinions of expert witnesses and basing its determination of fair value on a valuation method that was not used by an expert witness. There is reasonable evidence to support the district court's determination of fair value.

Affirmed.