The principal issue in this appeal is whether the terms of a merger between Mergenthaler Linotype Company ("Mergenthaler") and the Electric Auto-Lite Company ("Auto-Lite") were fair to Auto-Lite's minority shareholders. We hold that the merger terms were fair and reverse the judgment of the district court.

* This is the second time that this case has come before us. In order to place this appeal in its proper perspective, we will briefly review the case's history.

Prior to 1960 Auto-Lite primarily manufactured automotive parts and equipment. Because of changes in the automobile industry in the late 1950's, the continuation of its traditional business was threatened and it began a program of diversification into other industries. Nonetheless, in 1963 a substantial percentage of its sales remained tied to the automobile industry.

Mergenthaler produced and distributed typesetting equipment. It began to purchase Auto-Lite stock in 1957 and by March 1962 had acquired 54.2 percent of Auto-Lite. At that point Auto-Lite became a subsidiary of Mergenthaler and Mergenthaler obtained control of the Auto-Lite Board of Directors.

In early 1963 Mergenthaler decided to attempt a merger of itself and Auto-Lite into a new company to be called "Eltra Corporation." The Auto-Lite Board voted to accept the proposed merger on May 28, 1963 and on May 29 a request for proxies was sent to shareholders accompanied by a statement in which the Board endorsed the merger. The holders of approximately thirteen percent of the minority shares had to approve the merger in order to secure the two-thirds vote necessary for ratification. The merger was approved at a shareholders' meeting on June 27, 1963 and became effective on June 28, 1963.

Plaintiffs, who were Auto-Lite shareholders representing themselves and all other minority Auto-Lite shareholders, filed suit on June 26, 1963 in the district court for the Northern District of Illinois. They sought to set aside the merger on the ground that the proxy statement was deceptive because it endorsed the merger without clearly disclosing that the Auto-Lite Board was controlled by Mergenthaler. The district court agreed with plaintiffs' theory and held that the proxy statement violated section 14(a) of the Securities Exchange Act of 1934. It also held that the plaintiffs had shown a causal relationship between the proxy statement and the consummation of the merger. 281 F. Supp. 826 (N.D.Ill.1967).

This court affirmed the district court's holding that the proxy statement was illegally deceptive but reversed on the causality issue, holding that no relief could be appropriate if the defendants could show "by a preponderance of probabilities" that the merger would have been approved even if the proxy statement had not been deceptive. 403 F.2d 429, 436 (7th Cir. 1968).

The Supreme Court reversed and reinstated the judgment of the district court. It held that "(w)here there has been a finding of materiality, a shareholder has made a sufficient showing of causal relationship between the violation and the injury for which he seeks redress if, as here, he proves that the proxy solicitation itself, rather than the particular defect in the solicitation materials, was an essential link in the accomplishment of the transaction." 396 U.S. 375, 385, 90 S. Ct. 616, 622, 24 L. Ed. 2d 593 (1970).

The Court went on to consider the relief to which plaintiffs were entitled. It noted that the merger did not need to be set aside because of the deception in the proxy statement, though a court could order such action if it were warranted by equitable considerations. It then discussed the possibility of monetary relief and stated:

(W)here, as here, the misleading aspect of the solicitation did not relate to terms of the merger, monetary relief might be afforded to the shareholders only if the merger resulted in a reduction of the earnings or earnings potential of their holdings. In short, damages should be recoverable only to the extent that they can be shown. If commingling of the assets and operations of the merged companies makes it impossible to establish direct injury from the merger, relief might be predicated on a determination of the fairness of the terms of the merger at the time it was approved. These questions, of course, are for decision in the first instance by the District Court on remand, and our singling out of some of the possibilities is not intended to exclude others. 396 U.S. at 388-89, 90 S.Ct. at 624.

Finally, the Court held that the plaintiffs were entitled to be reimbursed by the corporation for litigation expenses and reasonable attorneys' fees.

The case then moved back to the district court for a determination of the appropriate relief. The court first held that the merger should not be rescinded. It then determined that the merger terms were unfair to plaintiffs and awarded the class they represent $1,233,918.35, as well as approximately $740,000 in prejudgment interest. It further held that plaintiffs' attorneys should be compensated out of this award.

Both parties now appeal from the district court's judgment. Plaintiffs contend that the amount of damages was too low and that their attorneys' fees should be assessed against defendants rather than against the damages awarded. Defendants assert that no damages should have been awarded because the terms of the merger were fair.

The district court considered two possible theories of damages in attempting to follow the Supreme Court's mandate: (1) to compensate plaintiffs for the reduction of the earnings potential of their holdings in Auto-Lite as a result of the merger; or (2) an award based on a determination of the fairness of the terms of the merger at the time it was approved. It rejected the use of the first theory under the circumstances of this case and adopted the second. We shall evaluate both whether the district court was erroneous in its choice of remedies and whether it correctly applied the second theory.

In order to perform this evaluation, it is first necessary to describe the merger terms. They called for the minority Auto-Lite shareholders to receive 1.88 preferred shares of Eltra for each share of Auto-Lite common that they held and the Mergenthaler shareholders to receive one common share of Eltra for each share of Mergenthaler common that they held. Eltra preferred shares were convertible into common shares on a one-to-one basis for the first two years following the merger and on a slightly decreasing basis for the next three years.2 At the time of the merger Mergenthaler common paid a dividend of $1 per share and Auto-Lite common paid a dividend of $2.40 per share. Under the merger terms Eltra common was to pay a dividend of $1 per share and Eltra preferred a dividend of $1.40 per share. The dividend received by the Auto-Lite minority shareholders was therefore increased as a result of the merger by twenty-three cents for each share of Auto-Lite that they had held, because 1.88 X $1.40 = $2.63.

The preferred Eltra stock was clearly worth more than Eltra common because it paid a higher dividend and represented a more secure investment if the new corporation encountered financial difficulties, yet was convertible into common stock. During the month following the merger, the average market value of Eltra preferred was $31.06 per share. Consequently the Auto-Lite minority shareholders received stock worth $58.39 on the market for each share of Auto-Lite that they had previously held, because 1.88 X $31.06 = $58.39. The average market value of Eltra common for this month was $25.25 per share. Since the Mergenthaler shareholders received one share of Eltra common for each share of Mergenthaler common, the Auto-Lite shareholders received stock for each share of Auto-Lite that they held worth $58.39/$25.25 = 2.31 times as much on the market as the stock that the Mergenthaler shareholders received for each share of Mergenthaler that they held. We therefore hold that the exchange ratio for the merger was effectively 2.31 to 1.3

A theory of damages based on the "reduction of the earnings or earnings potential" of the Auto-Lite minority shareholders caused by the merger is an attempt to discern, by looking at the postmerger performance and activities of the Auto-Lite subsidiary in comparison to the other components of Eltra, whether the value placed on the Auto-Lite shares at the time the merger took place was fair to those shareholders. Plaintiffs contend that the postmerger record of Eltra demonstrates the unfairness of the merger in two significant ways: first, by showing that Eltra appropriated for use in its other divisions liquid assets held by Auto-Lite prior to the merger; and second, by showing that Eltra continually siphoned off Auto-Lite's postmerger earnings.

Even if plaintiffs' assertions are true, they cannot form the basis for an award of damages. Plaintiffs assume that it would be unfair to the former minority Auto-Lite shareholders if, after the merger, the Eltra management "weakened" the Auto-Lite divisions by shifting liquid assets or earnings to other divisions. This assumption is incorrect. After the merger the former Auto-Lite shareholders had become Eltra shareholders and had no more interest in the Auto-Lite divisions than any other Eltra shareholder. Therefore, they could not be injured by intra-corporate transfers of assets designed to strengthen the corporation as a whole. The interests of the former Auto-Lite shareholders and former Mergenthaler shareholders coincided after the merger, and the Eltra management could not possibly take actions that benefitted "its" shareholders at the expense of the Auto-Lite shareholders.

Plaintiffs rely heavily on Affiliated Ute Citizens v. United States, 406 U.S. 128, 92 S. Ct. 1456, 31 L. Ed. 2d 741 (1975); Gerstle v. Gamble-Skogmo, Inc., 478 F.2d 1281 (2d Cir. 1973); and Janigan v. Taylor, 344 F.2d 781 (1st Cir.), cert. denied, 382 U.S. 879, 86 S. Ct. 163, 15 L. Ed. 2d 120 (1965), to support their theory of postmerger damages. They contend that under these decisions the defendants must be compelled to disgorge any "profits" realized as a consequence of the deceptive proxy statement and that those profits should be measured by the sum of assets which moved from Auto-Lite to the other divisions of Eltra following the merger. However, plaintiffs misconceive the holdings of these cases. The cases all involved situations where defendants had realized demonstrable profits, at the expense of plaintiffs, through misrepresentation. Under such circumstances all three courts permitted a postmerger valuation of property which plaintiffs had been fraudulently induced to sell. But whether the exchange of Auto-Lite stock for Eltra stock profited Mergenthaler at the expense of the Auto-Lite minority shareholders is precisely the issue we are trying to resolve. Damages must be based on evidence that the Auto-Lite minority shareholders were not paid a fair price and the fact that Eltra shifted assets away from Auto-Lite following the merger is not satisfactory evidence of unfairness at the time the merger was consummated.

Plaintiffs ignored the only theory on which relief based on the postmerger performance of Eltra might be granted. If the ratio of the postmerger earnings of the Auto-Lite subsidiary of Eltra to the postmerger earnings of the Mergenthaler subsidiary were unusually high given the terms of the merger, it would be evidence that those terms were unfair to the Auto-Lite minority shareholders. This is true not because the former Auto-Lite minority shareholders are entitled to any percentage of the earnings of the Auto-Lite subsidiary after the merger, but because a high ratio would indicate that in retrospect the merger terms underestimated Auto-Lite's value as an enterprise.

Eltra's financial records show that for the ten year period beginning in 1963 and ending in 1972 the Auto-Lite divisions earned $122,501,632 while the Mergenthaler divisions earned $58,827,595. Before the merger Auto-Lite had 1,159,265 shares outstanding while Mergenthaler had 2,698,822 shares outstanding. Arguably, therefore, Auto-Lite would have had average yearly earnings per share of $10.57 between 1963 and 1972 if there had been no merger because its average yearly earnings would be $122,501,63 2/10 = $12,250,163 and $12,250,16 3/1,159,265 = $10.57. By the same reasoning, Mergenthaler's average yearly earnings per share for the same period would be $2.18 because $58,827,59 5/10 = $5,882,759 and $5,882,75 9/2,698,822 = $2.18. The ratio of $10.57 to $2.18 is 4.85, indicating that the actual effective exchange ratio of 2.31 underestimated the value of Auto-Lite in comparison to Mergenthaler.

An award of damages based on the comparative postmerger earnings of Auto-Lite and Mergenthaler, however, depends upon the assumption that the two subsidiaries continued to function independently after the merger was consummated. If the assets or operations of the two subsidiaries were commingled, it would be improper to utilize the postmerger performance of either one as evidence that at the time of the merger Auto-Lite was a stronger company than the merger terms indicated, because Auto-Lite's increase in earnings following the merger might have been the result of input from Mergenthaler that it would not have received without the merger.

The district court held that there was substantial commingling of the assets and operations of Auto-Lite and Mergenthaler during the period following the merger. It found that the plaintiffs' statistics failed to measure the change in the quality of management that flowed to the Auto-Lite divisions of Eltra as a consequence of the merger or the economies of scale that the merger produced. It also found the statistics to be misleading because the postmerger Eltra statements underestimated the expenses of the Auto-Lite divisions, making the earnings of those divisions appear higher than they really were. These findings are supported by substantial evidence and we affirm them.

Given the fact that significant commingling occurred, the postmerger earnings of Auto-Lite and Mergenthaler cannot supply a reliable guide to whether the merger terms were fair to the Auto-Lite minority shareholders. Even in the absence of commingling, postmerger evidence can only create a rebuttable inference of unfairness because it is impossible to know with certainty whether the increase in earnings of one partner to a merger should have been predictable at the time the merger took place. In this case the ratio of the earnings per share of the two companies for the four years prior to and including 1963 were all at or below the effective exchange ratio of 2.31 to 1.4 Plaintiffs have not shown that the management of Mergenthaler should have known in 1963 that Auto-Lite's earnings were going to increase faster than Mergenthaler's during the next decade. The more plausible inference is that, insofar as Auto-Lite's business became more productive because of factors unrelated to commingling, they were unforeseeable at the time the merger was consummated. Accordingly, we hold that the district court did not abuse its discretion in refusing to award damages based on postmerger data.

The district court based its award of damages on an assessment of the fairness of the merger terms at the time the merger took place. It evaluated five criteria in making this assessment: (1) the market value of each corporation's stock; (2) each corporation's earnings; (3) the book value of each corporation's assets; (4) the dividends that each corporation paid on its stock; and (5) other "qualitative factors" indicating the strength of each corporation. The court found market value to be an unreliable criterion and discounted the importance of dividends. It found that the comparative earnings and book values of each corporation were significant and demonstrated that the merger terms were unfair to the Auto-Lite minority. The court did not indicate what significance it was attributing to "qualitative factors."

Based on these findings the court held that the merger would have been fair if the Auto-Lite minority shareholders had received the equivalent of 2.35 shares of Eltra common for each share of Auto-Lite that they held and Mergenthaler shareholders had received one share of Eltra common for each share of Mergenthaler that they held. It also found that the effective exchange ratio for the actual merger, where the Auto-Lite minority shareholders received 1.88 shares of Eltra preferred for each share of Auto-Lite common that they held, was 2.25 to 1 in terms of Eltra common.5 It then awarded damages of $1,233,918.35 to plaintiffs based on the differential of .10 between the effective exchange ratio of 2.25 to 1 and the fair exchange ratio of 2.35 to 1.6B.

The district court discounted the significance of the comparative market values of Auto-Lite and Mergenthaler stock during the five year period preceding the merger because it found that purchases of Auto-Lite stock by Auto-Lite itself and by Mergenthaler, and of Mergenthaler stock by the American Manufacturing Company,7 made market value an unreliable indicator during that period of the true worth of the two parties to the merger. Defendants challenge the district court's assessment while plaintiffs contend that it was correct, at least for the period after 1960. We agree with defendants.

The district court's holding depends upon the validity of two premises: first, that the inter- and intra-company purchases substantially affected the market value of either corporation's stock immediately prior to the merger and second, that any effect which these purchases did have caused the price of Auto-Lite stock to fall relative to the price of Mergenthaler stock.8 We find neither premise to be supported by the evidence. The great portion of the inter- and intra-company purchases took place before the six month period immediately preceding the merger, and the effect of those purchases during that period was negligible.9 There was no manipulation of stock prices immediately before the merger by any of the companies involved in an effort to secure more favorable merger terms. Moreover, the price ratio of the two stocks was the same during 1961 and 1962, when there were substantial purchases by Auto-Lite, Mergenthaler, and American, and the first half of 1963, when there were not.10 This is a strong indication that the price of both corporations' stock was responding to factors other than these transactions.

Even if we assume that the purchases had an effect, it is difficult to see how it could be detrimental to the price of Auto-Lite relative to the price of Mergenthaler. Since all of the transactions at issue were purchases, they would have tended to raise rather than lower the price of the stock that was traded. But there was a greater volume of Auto-Lite stock purchases by Auto-Lite itself and by Mergenthaler than of Mergenthaler purchases by American.11 Second, during the eighteen months prior to the merger the ratio of the inter- and intra-company purchases of Auto-Lite to the total amount of Auto-Lite stock traded was greater than the ratio of American's purchases of Mergenthaler to the total amount of Mergenthaler stock traded.12 Finally, a greater proportion of the Auto-Lite purchases took place in the period between 1961 and 1963, when they were likely to have a more substantial effect on price than earlier purchases.13 The combination of these three factors indicates that the inter- and intra-company purchases pushed the price of Auto-Lite stock upward relative to Mergenthaler stock rather than producing the opposite effect.

We therefore hold that the inter- and intra-company transactions did not unfairly distort the relative market prices of Auto-Lite and Mergenthaler for purposes of determining the fairness of the merger. We must now decide what period of time should be used in calculating a price ratio between each corporation's stock. Since prices from the period immediately preceding the merger are the most likely to reflect the actual value of each corporation at the time the merger was consummated, we begin with a presumption that a short period is appropriate. Accordingly, we hold that the average market value for approximately the six month period preceding the merger should be used unless there are special factors indicating that this period is unreliable. Six months is long enough so that very short term price fluctuations will not play an unfairly important role and short enough so that the calculated ratio does not reflect business conditions that have substantially changed as of the time of the merger.

In this case the ratio between the average price of Auto-Lite and the average price of Mergenthaler during 1963 prior to the formulation of the merger terms in late May was 2.1. Our confidence that this figure accurately reflects the relative worth of the two corporations is bolstered by the fact that the ratio for 1962 was also 2.1 and was 2.0 for 1961. The similarity of these numbers is evidence that the ratio immediately preceding the merger was not the result of a short term anomaly caused either by the merger itself or by other factors.

Plaintiffs argue, however, that market prices were an unreliable indicator of the true worth of Auto-Lite and Mergenthaler between 1961 and 1963 but did provide a valid measure of each corporation's value between 1958 and 1960. We find these arguments to be unpersuasive. Plaintiffs first assert that Mergenthaler used its control over Auto-Lite in 1961-63 to compel it to pay unusually high dividends, with the effect of depressing the price of Auto-Lite stock by draining Auto-Lite of capital and raising the price of Mergenthaler stock by giving Mergenthaler funds with which to pay higher dividends. This reasoning is inconsistent. If the payment of higher dividends by Mergenthaler increased the value of Mergenthaler stock, it would follow that the payment of high dividends by Auto-Lite increased the attractiveness of Auto-Lite stock on the market. There is no reason to believe that a reduction in the Auto-Lite dividend would have resulted in raising the price of Auto-Lite stock. It is more likely that such a dividend reduction would have decreased the value of each corporation's stock. Accordingly, we hold that Auto-Lite's high dividends between 1961 and 1963 did not unfairly distort the relative market values of Auto-Lite and Mergenthaler.

Plaintiffs also urge that market prices between 1958 and 1960 provided a more accurate picture of each corporation's actual value than prices during the later period because a substantial change in the nature of Auto-Lite's business following 1960 temporarily depressed the price of its stock. We find this to be an argument for using the later prices rather than the earlier ones. The fact that Auto-Lite's traditional business was eroding by 1961 renders unreliable market values based on that business. In 1961 the market began to reflect the uncertainty in Auto-Lite's future which most observers perceived. That uncertainty remained at the time of the merger and should properly have been a factor in determining the merger terms.14

After finding that market value provided an inaccurate measure of the true worth of Auto-Lite and Mergenthaler, the district court determined whether the merger terms were fair on the basis of comparative earnings and book value. Given our conclusion that market prices were an accurate gauge of actual value, we must decide whether the other criteria on which the district court relied should properly be considered in evaluating whether the merger was fair.

We hold that when market value is available and reliable, other factors should not be utilized in determining whether the terms of a merger were fair. Although criteria such as earnings and book value are an indication of actual worth, they are only secondary indicia. In a market economy, market value will always be the primary gauge of an enterprise's worth. In this case thousands of shares of Auto-Lite and Mergenthaler were traded on the New York Stock Exchange during the first part of 1963 by outside investors15 who had access to the full gamut of financial information about both corporations, including earnings and book value. If we were to independently assess criteria other than market value in our effort to determine whether the merger terms were fair, we would be substituting our abstract judgment for that of the market. Aside from the problems that would arise in deciding how much weight to give each criterion, such a method would be economically unsound.

We turn now to a determination of whether the merger terms were fair, based on the comparative market price of each corporation's stock during the first part of 1963. The simplest method of resolving this issue would be to compare the price ratio, in this case 2.1, to the effective exchange ratio, which we have previously established as 2.31. Under this framework the merger would be fair since the effective exchange ratio gave the Auto-Lite minority shareholders more Eltra stock than they were entitled to in the judgment of the market.

This method of calculation, however, assumes that the new corporation that results from a merger is worth exactly as much as the sum of what its two component parts were worth before the merger. As Professors Brudney and Chirelstein have cogently pointed out, this assumption is usually false because a merger produces a synergistic effect resulting in the merged corporation being worth more than the sum of the two old corporations. Brudney & Chirelstein, Fair Shares in Corporate Mergers and Takeovers, 88 Harv.L.Rev. 297, 308-09 (1974). They demonstrate that fairness requires that minority shareholders be compensated not only for the market value of their shares in the old corporation but also for the share of the synergism generated by the merger that is proportionate to the interest that those shares represented in the combined premerger value of the two old corporations. Id. at 313-25.

We adopt the approach formulated by Professors Brudney and Chirelstein and will attempt to apply it to this case. At the time of the merger there were 532,550 minority shares of Auto-Lite and 2,698,822 shares of Mergenthaler outstanding. During the first part of 1963 the average market price of Auto-Lite was $52.25 per share and the average market price of Mergenthaler was $24.875 per share. Thus, the premerger value of the minority holdings in Auto-Lite was 532,550 x $52.25 = $27,825,737 and the premerger value of Mergenthaler was 2,698,822 X $24.875 = $67,133,197. The combined premerger value of the two corporations was $27,825,737 k $67,133,197 = $94,958,934.16

In the month following the merger, Eltra common stock had an average market value of $25.25 per share. Eltra preferred stock had an average market value of $58.39 per 1.88 shares, the amount of stock which Auto-Lite shareholders had received for each share of Auto-Lite that they had held. The postmerger value of Eltra was therefore (2,698,822 X $25.25) k (532,550 X $58.39) = $68,145,255 k $31,095,594 = $99,240,849. The difference between the combined premerger value of Auto-Lite and Mergenthaler and the postmerger value of Eltra, which was $99,240,849 - $94,958,934 = $4,281,915, can be attributed to the synergism generated by the merger.

According to the fairness formula devised by Professors Brudney and Chirelstein, the minority shareholders of Auto-Lite should have received Eltra stock worth at least as much as the premerger market value of their holdings in Auto-Lite and a share of the synergism produced by the merger proportionate to the percentage of the combined premerger value of Auto-Lite and Mergenthaler which their holdings represented. The premerger value of the Auto-Lite minority shares was $27,825,737, which represented 29.3 percent of $94,958,934, the combined premerger value of Auto-Lite and Mergenthaler. Thus, to satisfy the constraints of fairness, the Auto-Lite minority shareholders should have received stock worth at least $27,825,737 k (.293 x $4,281,915) = $29,080,338. This would be equivalent to 1,151,696.5 shares of Eltra common at $25.25 per share. Had this many shares been distributed to the Auto-Lite minority shareholders, the exchange ratio would have been 1,151,696. 5/532,550 = 2.16 to 1.17

The Auto-Lite minority shareholders actually received preferred stock worth $58.39 on the market for each share of Auto-Lite that they had held. As a group, their Eltra holdings were worth 532,550 x $58.39 = $31,095,594. This was $31,095,594 - $29,080,338 = $2,015,256 more than fairness required. This result can be expressed in terms of Eltra common shares. Since the effective exchange ratio of the merger was 2.31 to 1, the property given the Auto-Lite minority was worth 2.31 - 2.16 = .15 shares of Eltra common per share of Auto-Lite more than what a fair amount would have been.

We therefore hold that the terms of the merger were fair and that plaintiffs should recover no damages. A numerical example may help to show the justice of this result. In early 1963, an Auto-Lite shareholder with one hundred shares and a Mergenthaler shareholder with 210 shares each owned stock worth approximately $5225. After the merger, the former Auto-Lite shareholder had 188 shares of Eltra preferred worth approximately $5839 while the former Mergenthaler shareholder had 210 shares of Eltra common worth approximately $5302. Both individuals benefitted from the merger, but the former Auto-Lite minority shareholder benefitted more.

The district court held that plaintiffs should recover their attorneys' fees and other litigation expenses out of the damages which the court awarded them. Since we have found that plaintiffs were treated fairly in the merger and are not entitled to damages, we must face the question of whether and how much they nonetheless should be compensated by Eltra for their fees and expenses.

When this case was before the Supreme Court, the Court held "that petitioners, who have established a violation of the securities laws by their corporation and its officials, should be reimbursed by the corporation or its survivor for the costs of establishing the violation." 396 U.S. at 389, 90 S.Ct. at 624 (footnote omitted). The Court later stated that "(w)hether petitioners are successful in showing a need for significant relief may be a factor in determining whether a further award should later be made. But regardless of the relief granted, private stockholders' actions of this sort 'involve corporate therapeutics,' and furnish a benefit to all shareholders by providing an important means of enforcement of the proxy statute." Id. at 396, 90 S.Ct. at 627 (footnotes omitted).

We interpret this language to mean that plaintiffs should recover from Eltra all fees and expenses related to the establishment of a violation of the federal securities laws. This would encompass the litigation through the decision in the Supreme Court. The Court has indicated that the award is not contingent upon the existence of a fund created from any damages eventually awarded the plaintiffs.18

The Court left open the question of whether plaintiffs should also be compensated for fees and expenses incurred in a trial and subsequent appeals on the issue of damages, implying that whether or not plaintiffs obtained damages should be a factor in resolving that question. We hold that plaintiffs are not entitled to be compensated for fees and expenses they have incurred since the Supreme Court's decision in this case. In Alyeska Pipeline Co. v. Wilderness Society, 421 U.S. 240, 95 S. Ct. 1612, 44 L. Ed. 2d 141 (1975), the Supreme Court held that, absent statutory authority, the only exceptions to the general American rule that a litigant must pay his own fees and expenses are in situations where the litigant confers a common benefit on the class that he represents or where a losing party acts in bad faith. Since the Securities Exchange Act provides no authorization for fee-shifting in this type of case and the bad faith exception is inapposite here, plaintiffs cannot recover fees and expenses unless they demonstrate that their work on the issue of damages produced a common benefit for the former Auto-Lite minority shareholders. Plaintiffs cannot meet this burden. They have conferred no benefit upon the class they represent beyond the "corporate therapeutics" produced by their showing that the proxy statute was violated, and they must pay their own fees and expenses for their unsuccessful attempt to obtain damages for that violation.

The judgment of the district court is reversed and the cause is remanded for further proceedings consistent with this opinion.