ISQUITH v. MIDDLE SOUTH UTILITIES, INC.

KING, Circuit Judge:

Plaintiffs, who purchased common stock of Middle South Utilities, Inc. during the period from March 30, 1983 to August 16, 1985, appeal a summary judgment entered against them in their proposed class action against Middle South, its officers and directors, certain of its subsidiaries, the underwriters of a common stock offering that occurred during that period, and its independent auditors. For the reasons set forth below, we vacate the summary judgment and remand the case to the district court for further development.

I.

A. Background Information1

Middle South is an investor-owned public utility holding company which owns all the outstanding common stock of four operating utility subsidiaries: Arkansas Power & Light Co., Louisiana Power & Light Co. ("LP & L"), Mississippi Power & Light Co. ("MP & L"), and New Orleans Public Service, Inc. In the late 1960's and early 1970's, Middle South embarked upon a program to expand the power generating facilities of its operating subsidiaries. LP & L was given the responsibility for constructing Waterford, a 1,104,000 kilowatt nuclear power unit. Initially, MP & L was given the responsibility for constructing Grand Gulf Station, a generating facility composed of two nuclear units: Grand Gulf 1 and Grand Gulf 2. However, when it became apparent in 1974 that none of the operating subsidiaries alone possessed sufficient resources to finance, construct, and operate Grand Gulf Station, Middle South formed another subsidiary, Middle South Energy, Inc. ("MSE"), for that purpose.

Originally, MSE estimated that Grand Gulf 1 would begin commercial operation in 1979, and that the total cost of the unit would be $600 million. Actually, construction of Grand Gulf 1 was completed six years behind schedule and greatly over budget; commercial operation of the facility — which had an actual installed cost of $3.29 billion — finally began in July of 1985. Waterford experienced similar problems. When plans for the unit began in 1970, Waterford was slated for commercial operation by January, 1977 at a total cost of $230 million. Reality, however, was that commercial operation did not occur until September, 1985, at which time Waterford's installed cost had increased to approximately $3 billion.

Middle South managed its operating subsidiaries as an integrated system. As part of this system, the electricity generated by each power plant — regardless of the facility's actual owner — was pooled for use by the operating subsidiaries according to their individual needs. In addition to, and in fact because of, this sharing of energy, the operating subsidiaries also shared in the cost of the system's energy-producing facilities. The ability of the operating subsidiaries to repay obligations incurred to finance the nuclear power plants and to earn a return on their investments in those plants depended upon their ability to obtain increases in the retail rates charged to their customers. However, as public utilities, the operating subsidiaries' retail rates are controlled by state and local regulatory bodies. Accordingly, the extent to which the operating subsidiaries could successfully shift the cost of the nuclear power plants to their respective customers rested, at least in part, in the hands of the relevant regulatory bodies, which could grant, deny, or delay adequate rate relief.

In the early 1980's, however, the state regulatory authorities became hostile to the Grand Gulf 1 and Waterford projects. This hostility stemmed from concern that the demand for electricity which was projected for the Middle South area at the inception of the nuclear power plant construction projects was significantly overestimated. The regulatory authorities were also concerned over the dramatically escalating costs of those projects.

In the summer of 1985, the regulatory authorities began denying rate increases which the operating subsidiaries needed in order to meet their financial obligations. The authorities' actions precipitated serious financial crises for the individual operating subsidiaries and, predictably, for Middle South. During the week of August 16, 1985, Middle South's common stock declined in price from $12 per share to $9.125 per share. On August 16, 1985, the Securities and Exchange Commission disclosed that Chapter 11 was a "very real possibility" for both Middle South and its operating subsidiaries unless emergency rate relief was granted soon. During the week of August 26, 1985, all of the operating subsidiaries withheld payment of their common stock dividends to Middle South. As a result, on August 29, 1985, Middle South omitted its third quarter dividend on common stock to its shareholders.

B. The Complaint and Defendants' Responses Thereto

The five suits covered by this appeal, which have been consolidated for pretrial purposes, ensued. Briefly stated, plaintiffs' seventy-one page consolidated amended complaint challenged — under the federal securities laws and Louisiana state law — the adequacy of more than forty excerpts from twelve documents Middle South filed with the SEC and disseminated to its stockholders from 1982 to 1985. Basically, plaintiffs alleged that to pay for the construction and related costs of Waterford and Grand Gulf Station, the Middle South system needed debt and equity capital from the investing public. The Middle South system's ability to raise this needed capital, however, was dependent on the maintenance of stockholder and investor confidence, in Middle South and its operating subsidiaries. To insure that necessary confidence, plaintiffs posited, Middle South and the other defendants embarked upon a "scheme, plan and course of conduct to issue and disseminate reports and statements which artificially inflated the market price" of Middle South's common stock. Plaintiffs charged that, in the twelve documents principally at issue here, defendants misrepresented to, and concealed from, the investing public a substantial number of material facts in violation, depending upon the document involved, of section 11 or 15 of the Securities Act of 1933 (the "1933 Act"), or section 10(b) or 20 of the Securities Exchange Act of 1934 (the "1934 Act"). The misrepresentations and omissions concerned five areas of disclosure: (1) the Middle South system's need for the electricity to be generated from the Waterford and Grand Gulf Station units; (2) the ability of the operating subsidiaries to obtain adequate and timely rate relief from state regulatory agencies to meet the financial obligations they incurred to construct the nuclear power plants; (3) the factual bases for the various estimates of costs and dates of completion which Middle South disclosed to the public during the construction of Waterford and Grand Gulf 1; (4) the effect of the operating subsidiaries' cost-recovery problems on Middle South's ability to continue to pay dividends to its common stockholders; and (5) the propriety of reporting Equity-AFUDC2 as income in Middle South's financial statements during the construction of the Waterford and Grand Gulf Station facilities.

None of the defendants filed an answer to the complaint. Instead, on April 18, 1986, the Middle South defendants filed a 2 page "Motion To Dismiss" and a 127 page "Memorandum of Law In Support of Defendants' Motion to Dismiss." In their motion, the Middle South defendants

For our purposes, the Middle South defendants asserted that the plaintiffs had failed to state a claim under both the federal securities laws and state law because the documents upon which the complaint focused, when taken in their entirety, as a matter of law fully disclosed without misrepresentation all of the facts which the plaintiffs claimed the documents either misrepresented or omitted. To prove this point, the Middle South defendants filled approximately twenty-one pages with "disclosure examples" taken principally from the documents at issue in plaintiffs' complaint. For purposes of comparison, the excerpts were grouped into five categories which, according to the defendants, roughly corresponded to the categories drawn in the plaintiffs' complaint. Because the Middle South defendants were unsure, however, whether the district court would treat their motion as a motion to dismiss or as a motion for summary judgment, they repeated many of these excerpts in verified form. Along with their motion, the Middle South defendants filed a three page "Statement of Uncontested Facts;" accompanying this statement was an instruction that it was to be used by the court "[i]f and only if and to the extent that" the court elected to treat the defendants' motion as a motion for summary judgment. The statement itself, however, set forth only two "facts": first, that the affidavit of Edward Lupberger attached to the statement contained undisputed facts and second, that the exhibits attached to Lupberger's affidavit were authentic. The Lupberger affidavit did little to expand the base of undisputed facts — it asserted only that Lupberger was president and chairman of the board of Middle South, and that the documents attached to his affidavit were copies of documents which Middle South had filed with the SEC and which had been "widely disseminated to institutional investors and analysts." These documents included ten of the twelve documents which plaintiffs' complaint challenged as misleading. To these ten documents, the Middle South defendants added six more documents filed with the SEC during the relevant time period. Not one of these sixteen documents, however, was a complete copy — instead, from each lengthy document, the Middle South defendants chose to include only two to eight pages of excerpts.

Shortly after the Middle South defendants filed their "Motion to Dismiss," the other defendants filed motions to adopt and join the Middle South defendants' motion. One defendant added other reasons to dismiss that were not ruled on by the district court and are not relevant to this appeal.

On August 4, 1986, plaintiffs filed a 102 page "Memorandum of Law in Opposition to Defendants' Motions to Dismiss." In it, plaintiffs argued that it would be inappropriate to convert the defendants' motions to dismiss into motions for summary judgment. If, however, the court decided to do so "at this early stage," the plaintiffs requested notice and leave to submit affidavits or other materials permitted by Rule 56. Plaintiffs filed no supporting material with their Memorandum in Opposition. For another four months, the parties traded reply, surreply, and response briefs. The end result was more than a thousand pages of detailed allegations and responses, exhibits and memoranda of law.

C. The District Court's Decision

In January 1987, in a four page "Order and Reasons," the district court granted summary judgment to all defendants. The court began by acknowledging "the brevity of these reasons," but declared that since "[a]n appeal is virtually certain, [] it would be a waste of judicial resources for for [sic] us to write at length, or even at all, on every point raised by these motions." In fact the court did not address any point raised by the defendants' motions except in broad generalities. Essentially, the district court gave three reasons for its decision. First, while specifically declining to "parse each statement or to scrutinize each alleged omission upon which plaintiffs rely," the district court found that "a reasonable potential investor who would have read any of [the] documents [upon which plaintiffs base their claims] as a whole would have been fairly informed about the negative investment factors which plaintiffs contend were misstated or omitted by defendants." Consequently, the court found, "no reasonable fact-finder could conclude that the information published by defendants was misleading to such an extent as to make them liable to purchasers of [Middle South's] stock." In reaching this conclusion, the district court reported that it had considered all the excerpts set out in the complaint, accepted as true all the complaint's allegations, and reviewed those sections of the documents which the Middle South defendants had filed with their motion. Second, the court attacked one theory upon which plaintiffs' action was based — that the manner of disclosure, as distinguished from the fact of disclosure, can lead to liability under the securities laws. The court noted that:

From this observation, the court concluded that complaints about the way in which a company characterizes and interprets underlying facts are "simply not the stuff of which Section 11 or Rule 10(b)(5) liability is made." And finally, the court refused to find that either the defendants' failure to predict certain events — such as the regulatory agencies' decisions to deny rate increases — or their faulty prognostications about those events could give rise to liability under the securities laws. With respect to the procedural posture of the defendants' motions, the court had this to say:

From the district court's order, plaintiffs now bring this appeal. As expected, plaintiffs dispute each of the three bases upon which the district court rested its decision that summary judgment was appropriate. In addition, plaintiffs — characterizing defendants' motions below as motions to dismiss — protest the district court's decision to treat the documents as motions for summary judgment without first giving plaintiffs notice and the opportunity to respond. Defendants, on the other hand, make arguments in support of the district court's judgment; one of those arguments also turns on the procedural posture of the case. Although we ultimately conclude that we are unable in large part to review the substance of the district court's decision, we think — because they are potentially dispositive of this appeal — that the respective procedural arguments which the parties make can and should be addressed. We therefore turn to those procedural arguments first.

II.

A. The Procedural Posture of the Case

1. MOTIONS FOR SUMMARY JUDGMENT OR MOTIONS TO DISMISS?

Our first question is whether the defendants' motions below, as they related to the grounds upon which that court dismissed plaintiffs' suit, were Rule 12(b)(6) motions to dismiss for failure to state a claim or Rule 56 motions for summary judgment. The answer to this question is rendered more difficult than usual because of the peculiar nature of the defendants' motions. For in this case, the defendants' motions require us, on the one hand, to refer to evidence outside the complaint — normally the sign of a summary judgment motion — and, on the other hand, to assume that the allegations in the complaint are true — normally the sign of a motion to dismiss. We explain.

The thrust of defendants' argument was that, as a matter of law, defendants did not misrepresent or omit any material information from the documents it filed with the SEC and disseminated to its shareholders. In support of their argument, defendants quoted, from the documents plaintiffs' complaint focused on, portions which plaintiffs did not quote in their complaint; they included, as the substance of their Statement of Uncontested Facts, verified copies of parts of those documents; and they invited the district court to read the document portions and to determine whether those portions included and accurately represented all the facts that Middle South was required to disclose. The district court therefore, in order to address defendants' argument, was compelled to look beyond the plaintiffs' pleadings to these document exerpts. But looking only to the document sections that defendants provided the district court — as that court realized — did not permit it to evaluate defendants' argument. That is because the thrust of plaintiffs' complaint was that because the defendants knew particular facts and were aware of particular circumstances at particular times, they were required by the securities laws to disclose those facts and circumstances in a particular manner, in particular documents, at particular times. Before determining whether the documents adequately disclosed what Middle South was required to disclose, therefore, the district court had to first determine what Middle South was required to disclose. And because, in this case, the defendants failed to provide summary judgment evidence other than portions of the documents themselves, the trial court was required to look elsewhere for those facts. Where it turned, and where we also turned for the facts contained in this opinion, was to plaintiffs' complaint. The district court, as it acknowledged in its order, "accepted as true for purposes of [defendants' motion]" the allegations of the complaint.

Despite the apparent hybrid nature of defendants' motions — as interpreted by the district court3 — we conclude that it is appropriate to treat the motions, for purposes of our review, as ones for summary judgment. To the extent that the motions' success depends upon a review of the disclosure documents themselves, our conclusion is dictated by the language of Federal Rule of Civil Procedure 12(b):

As the Second Circuit noted, "[a]lthough this provision is not applicable to motions under other rules, it is mandatory with respect to motions pursuant to Rule 12(b)(6)," Goldman v. Belden, 754 F.2d 1059, 1066 (2d Cir.1985); our cases recognize the truth of this statement, see, e.g., Partridge v. Two Unknown Police Officers, 791 F.2d 1182, 1189 (5th Cir.1986); Hooten v. Jenne, 786 F.2d 692, 695 (5th Cir.1986); Underwood v. Hunter, 604 F.2d 367, 369 (5th Cir.1979). We have explained, however, that the defendants' claim of full and adequate disclosure cannot be evaluated except against the background of facts which existed at the time each individual disclosure was made. Since defendants' minimal statement of undisputed facts did not even attempt to set forth the needed background information,4 the success of defendants' motions also depends upon our acceptance of the factual allegations in plaintiffs' complaint as proper summary judgment evidence.

The Federal Rules themselves recognize that pleadings may, under the proper circumstances, be considered in deciding a motion for summary judgment. See Fed.R.Civ.P. 56(c). To date, however, our cases have only recognized the propriety of using verified pleadings as summary judgment proof — and even then, only to the extent that the pleadings "meet rule 56(e)'s requirement that summary judgment affidavits `shall be made on personal knowledge, shall set forth such facts as would be admissible in evidence, and shall show affirmatively that the affiant is competent to testify to the matters stated therein.'" Lodge Hall Music, Inc. v. Waco Wrangler Club, Inc., 831 F.2d 77, 80 (5th Cir.1987) (citing cases) (quoting Fed.R.Civ.P. 56(e)). There is no question, since most of the allegations in plaintiffs' complaint are explicitly based on "information and belief" and not personal knowledge, that the complaint's allegations do not meet Rule 56(e)'s stringent standards. In those cases where we required pleadings to comply with Rule 56, however, we did so because the pleadings were being used to support summary judgment in favor of the party who drafted them. See id. In this case, in contrast, defendants are attempting to use the complaint's allegations against plaintiffs; in essence, defendants are asserting that even if all of plaintiffs' factual allegations are true, Middle South correctly disclosed all the facts that plaintiffs complain were omitted or misrepresented. For example, in their memorandum in support of their Motion to Dismiss, defendants asserted that "plaintiffs' allegations of misstatements are refuted by many of the [Middle South] disclosures included in plaintiffs' Complaint" and that "[Middle South's] disclosure was legally sufficient in each of the areas identified by plaintiffs." Other examples run throughout the memorandum. Because of the position defendants have chosen to assert, we see no reason why, for purposes of resolution of the motion for summary judgment only, we cannot treat the factual allegations of the complaint as admissions or stipulations. As did the district court, therefore, we review defendants' motion as a motion for summary judgment supported by the factual allegations in plaintiffs' complaint and the document excerpts which made up defendants' Statement of Uncontested Facts. With this understanding, we turn to the procedural argument plaintiffs claim requires us to reverse the district court's judgment.

2. DID PLAINTIFFS HAVE NOTICE OF THE NATURE OF DEFENDANTS' MOTIONS?

Plaintiffs complain that the district court was wrong to treat the defendants' motions as motions for summary judgment without first giving plaintiffs notice and the opportunity to respond. In support of their argument, plaintiffs point to the fact that in their Memorandum in Opposition, plaintiffs specifically took the position that defendants' motions to dismiss could not properly be converted into motions for summary judgment and asked the district court, should it disagree, for notice and leave to submit summary judgment evidence. The Federal Rules of Civil Procedure and the case law of this circuit, plaintiffs argue, support their right both to make this request and to receive a response from the district court. We cannot agree.

It is true, of course, that whenever a motion to dismiss is converted into a motion for summary judgment,5 the non-movant is entitled to the "procedural safeguards of Rule 56." Capital Films Corp. v. Charles Fries Prods., Inc., 628 F.2d 387, 391 n. 1 (5th Cir.1980). One of these safeguards is that "[t]he motion shall be served at least ten days before the time fixed for the hearing." Fed.R.Civ.P. 56(c). But this does not mean, as plaintiffs contend, that they were entitled to notice that the court would, as opposed to could, treat the motion as one for summary judgment; we specifically rejected that argument in Clark v. Tarrant County, Texas, when we said:

798 F.2d 736, 746 (5th Cir.1986). The proper question, therefore, is whether the plaintiffs had ten days' notice after the court accepted for consideration matters outside the pleadings. In Clark, the plaintiffs were put on notice when the district court "accepted for consideration," at an oral hearing on several Rule 12(b) motions, testimony relating to the issues raised in the 12(b)(6) motion before it. 798 F.2d at 745-46. Here, the parties agreed that the court would consider the motions on submission, without an oral hearing. When no oral hearing is held, we have explained, the notice language of Rule 56(c) contemplates "10 day[s] advance notice to the adverse party that the matter will be heard and taken under advisement as of a certain day." Kibort v. Hampton, 538 F.2d 90, 91 (5th Cir.1976); accord Barker v. Norman, 651 F.2d 1107, 1119 n. 14 (5th Cir. Unit A July 1981).

The record in this case shows that on May 5, 1986, the court filed an order setting July 16, 1986 as the submission date for defendants' motions. Thereafter, the submission date was reset two times — on June 20, 1986 submission was reset for August 13, 1986, and on July 11, 1986 submission was reset for August 27, 1986. On both occasions, the court's impetus for resetting the date was a joint request for continuance filed by all parties. The Middle South defendants' motion was filed — with its supporting exhibits — on April 18, 1986; when the court on July 11, 1986 set the final submission date, therefore, it clearly had the extra-pleading material before it for review. Consequently, from July 11, 1986, the plaintiffs were on notice that the district court could, when it began reviewing the motions on August 27, 1986, consider matters outside the pleadings. Since the plaintiffs had over one month from the date of notice until the submission date and, in fact, had nearly five more months until the court ruled on the motions after it began reviewing them, plaintiffs had considerably more notice than the Federal Rules require.

Our conclusion that plaintiffs had adequate notice is not changed by the fact that plaintiffs specifically asked the district court to inform them if and when the court decided to treat the motions as motions for summary judgment. Plaintiffs requested this information, as their Memorandum in Opposition demonstrates, because they believed that without further opportunity for discovery, summary judgment would be premature. Once informed by the court that it would treat the motions as requesting summary judgment, plaintiffs suggested, plaintiffs would submit affidavits demonstrating their need for additional time in which to conduct discovery. However, when non-pleading materials are filed with a motion to dismiss, we have explained, see supra note 3, a district court has complete discretion under the Federal Rules of Civil Procedure to either accept the exhibits submitted or not, as it sees fit — and the Federal Rules impose on a court exercising its discretion in favor of accepting the non-pleading evidence no notice requirement beyond that which we have already considered. Plaintiffs, therefore, had no "right" under the Federal Rules of Civil Procedure to the information they asked the court to give them. Nor did the plaintiffs have a compelling argument to support their "need" for the information in this case. In fact, the plaintiffs offered the district court no reason at all why they could not have proffered affidavits supporting their desire for further discovery at the time they filed the Memorandum in Opposition; certainly the Federal Rules did not preclude them from doing so. See Fed.R.Civ.P. 56(f). Given that plaintiffs have shown neither a right to, nor a need for, the particular notice they requested of the district court, we see no reason for imposing on the court in this case a notice requirement more stringent than the Federal Rules of Civil Procedure already impose. Because we conclude that plaintiffs were on notice that the district court could treat defendants' motions as motions for summary judgment, we reject plaintiffs' procedural argument6 and turn our attention to defendants'.

3. ARE DEFENDANTS ENTITLED TO JUDGMENT BECAUSE PLAINTIFFS' RESPONSE BELOW TO DEFENDANTS' MOTIONS WAS INAPPROPRIATE?

Defendants urge — for the first time on appeal — that we must uphold the judgment for a reason which is strictly procedural and does not require a substantive review of defendants' disclosures. According to defendants, the Federal Rules of Civil Procedure, plus cases from the Supreme Court and this circuit, give a plaintiff faced with a motion for summary judgment two choices. The plaintiff can either (1) file an affidavit or respond, as otherwise provided by Rule 56, in a manner which sets forth specific facts showing the existence of a genuine issue for trial, or (2) file a Rule 56(f) affidavit requesting time for further discovery. Also according to the defendants, however, when it came to responding to defendants' motions in this case, plaintiffs took a calculated risk: they chose to treat defendants' motions solely as motions to dismiss and, therefore, made neither of the permitted responses. It was a risk, the defendants explain, because without documentary evidence from plaintiffs to raise an issue of material fact, no issue of material fact existed. And as long as none existed, the district court was right to grant summary judgment. We are not persuaded, however, because the argument misperceives the burden on a party attempting to resist summary judgment. To understand why, we first look to the Middle South defendants' Motion to Dismiss and plaintiffs' response to the motion.

In the motion itself, the Middle South defendants tried to negate an essential element of plaintiffs' claims by persuading the district court that Middle South's disclosures were, as a matter of law, not misleading; to do this, they set out to counter every claim of nondisclosure with evidence of disclosure, every claim of incomplete information with evidence of full and complete information. The "evidence" on which the defendants depended to demonstrate their entitlement to judgment consisted of the allegations of plaintiffs' complaint (to be taken as true) and the document excerpts which formed the substance of the Lupberger affidavit. In responding to this attack by the Middle South defendants on the validity of plaintiffs' claims, plaintiffs tried to persuade that the tactic the defendants used failed. But plaintiffs attached no affidavits or summary judgment evidence of their own to support their position. Instead, plaintiffs pointed to the factual allegations of their complaint, the disclosure excerpts in their complaint, and those disclosure excerpts from the Lupberger affidavit which the Middle South defendants believed entitled them to summary judgment. When the factual allegations are compared with the excerpts, and when the excerpts are compared with each other, plaintiffs argued, fact questions about the adequacy of Middle South's disclosures clearly appear. This is especially true, plaintiffs asserted, with respect to several "key" allegations of misrepresentation and omission which plaintiffs made in their complaint but which the Middle South defendants, in their motion, ignored. Because the evidence before the court raised fact questions which only a trial could resolve, plaintiffs concluded, the district court should not enter judgment for defendants.

To dispute the Middle South defendants' asserted right to judgment, therefore, plaintiffs pointed to fact questions which they believed the defendants' summary judgment evidence raised. According to the defendants, this response to the Middle South defendants' motion was not one of those which either the Federal Rules of Civil Procedure or case law allow. Implicit in the defendants' argument is the notion that a party faced with a motion for summary judgment supported by summary judgment evidence must proffer its own competing summary judgment evidence to survive the motion. Support for this position, the defendants seem to think, is found in a string of Fifth Circuit cases and the Supreme Court's recent decision in Celotex v. Catrett, 477 U.S. 317, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986).

In Celotex, of course, the Supreme Court did not specifically examine how a nonmovant properly responds to a motion for summary judgment; instead, the Court's focus was on how a party moving for summary judgment discharges its burden of "`show[ing] initially the absence of a genuine issue concerning any material fact,'" Adickes v. S.H. Kress & Co., 398 U.S. 144, 159, 90 S.Ct. 1598, 1609, 26 L.Ed.2d 142 (1970), quoted in Celotex, 477 U.S. at 325, 106 S.Ct. at 2552. See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250 n. 4, 106 S.Ct. 2505, 2511 n. 4, 91 L.Ed.2d 202 (1986). On that question, the Court rejected the notion that when a defending party moves for summary judgment, that party can meet its initial burden only by producing evidence which negates the existence of some material element of plaintiff's claim. Instead, the Court said, when the issue on which summary judgment is sought is not an issue for which the moving party will bear the burden of proof at trial, the moving party may, in lieu of proffering such evidence, simply point out that "the pleadings, depositions, answers to interrogatories, and admissions on file" contain no proof "concerning an essential element of the nonmoving party's case." Celotex, 477 U.S. at 325, 106 S.Ct. at 2552; accord Fontenot v. Upjohn Co., 780 F.2d 1190, 1195 (5th Cir.1986). In the process of resolving this basic issue concerning the movant's burden, however, the Court touched lightly on the nonmovant's responsive burden. According to the Court, once a party moving for summary judgment successfully discharges its initial burden, the nonmoving party must "`go beyond the pleadings and by [its] own affidavits, or by the depositions, answers to interrogatories, and admissions on file' designate `specific facts showing that there is a genuine issue for trial.'" Celotex, 477 U.S. at 324, 106 S.Ct. at 2553 (quoting Fed.R.Civ.P. 56(c), (e)). As the Court explained further, "Rule 56(e) permits a proper summary judgment motion to be opposed by any of the kinds of evidentiary materials listed in Rule 56(c), except the mere pleadings themselves, and it is from this list that one would normally expect the nonmoving party to make the showing to which we have referred." Id.

We disagree with the defendants that this language, which recognizes the shifting burdens of summary judgment, also requires the nonmovant to respond to a motion for summary judgment by proffering its own evidence. For one thing, the language the Court quotes from Rule 56 explicitly requires only that the evidence to which the nonmovant points be "on file" — not, as the defendants would have it, that the evidence be put on file by the nonmovant as part of its response. A more important consideration, however, is that finding summary judgment to be proper simply because the nonmovant failed to put forth its own evidence could result in summary judgments being granted in favor of movants who have not met the initial burden Rule 56 imposes on them — that is, showing the absence of a genuine issue concerning any material fact. For example, as Celotex recognizes, Rule 56 permits a movant to discharge this initial burden by "pointing out to the district court" that the record contains no evidence of an essential element of the nonmovant's case. Celotex, 477 U.S. at 325, 106 S.Ct. at 2554. Just because a movant makes that claim, however, does not mean that the claim is true. And if it is not — if the movant wrongly characterized the information in the record — the movant has not shown the absence of a genuine issue which would entitle it to summary judgment. The non-movant should be able to inform the court that summary judgment would be improper without having to offer more and different proof of the element for which defendants said there was no proof. Instead, the non-movant should simply be able to "show" the district court, by pointing to evidence in the record, that the movant has failed to sustain its initial burden. Accord Celotex, 477 U.S. at 332, 106 S.Ct. at 2557 (Brennan, J. dissenting) ("the nonmoving party may defeat a motion for summary judgment that asserts that the nonmoving party has no evidence by calling the court's attention to supporting evidence already in the record that was overlooked or ignored by the moving party") (not disagreeing with, but further explaining, the majority's analysis); see Slaughter v. Allstate Ins. Co., 803 F.2d 857, 860 (5th Cir.1986) (when plaintiffs failed to file any evidentiary material in opposition to defendant's motion for summary judgment, checking to see whether the plaintiffs pointed "either in the record or in their brief" to any genuine dispute of a material fact). Similarly, where the movant attaches summary judgment evidence which supposedly demonstrates that, as a matter of law, the movant must prevail on an element of the nonmovant's claim, the nonmovant should be able to answer the motion by arguing that the evidence raises a fact issue, which must be resolved at trial, concerning that element. Any other rule would exalt form over substance and, in the process, transform the filing of a motion for summary judgment into a mere mechanical process for the movant. The Celotex Court did not ignore the fact that the process of moving for summary judgment also imposes a substantive burden on the party requesting it; therefore, neither may we.

The Fifth Circuit cases to which the defendants point us do not affect our resolution of this question. Each case stands for a rule of summary judgment law that is beyond dispute: when a nonmovant is faced with a motion for summary judgment "made and supported" as provided by Rule 56, the nonmovant cannot survive the motion by resting on the mere allegations of its pleadings. See, e.g., Slaughter, 803 F.2d at 860; Fontenot, 780 F.2d at 1195-96; John Hancock Mut. Life Ins. Co. v. Johnson, 736 F.2d 315, 316 (5th Cir.1984); Daniels v. All Steel Equip., Inc., 590 F.2d 111, 113-14 (5th Cir.1979); see also Fed.R.Civ.P. 56(e). In these cases, this general rule meant that summary judgment was appropriate because the nonmovant, although faced with a summary judgment motion, chose not to respond to the motion at all. Here, of course, the plaintiffs tried — in a 102 page response — to defeat the defendants' motions by pointing to fact issues in the defendants' summary judgment evidence instead of proffering evidence of their own. This difference demonstrates why these cases hold no sway here. Not one of the cases holds — or even implies — that unless a nonmovant files its own evidentiary materials, the nonmovant is "resting upon the allegations of its pleadings." Nor would such an implication be proper; if a nonmovant points to record evidence which the movant used to support its claim to summary judgment, the nonmovant has necessarily gone beyond the mere allegations of its pleadings.7 Therefore, we conclude that the plaintiffs took an acceptable course below when they responded to the Middle South defendants' Motion to Dismiss by arguing that defendants' evidence created a fact issue as to the adequacy of Middle South's disclosures; consequently, the defendants are not, as the defendants assert, entitled to an affirmance simply because the plaintiffs did not support their response with their own evidence. Having resolved this procedural issue, we turn from our consideration of procedure to a review of the substance of the district court's decision.

B. The Adequacy of Plaintiffs' Allegations of Misrepresentation and Omission

1. DOES THE WAY IN WHICH INFORMATION IS DISCLOSED MATTER?

We begin our review with the district court's attack on the basic theory upon which part of plaintiffs' complaint is premised — that how information is disclosed, as distinguished from what information is disclosed, can lead to liability under both section 11 of the 1933 Act and Rule 10b-5, promulgated pursuant to section 10(b) of the 1934 Act.8 To demonstrate why it rejected the theory, the district court focused on a specific complaint by plaintiffs: that various of Middle South's disclosure documents emphasized data which showed an increase in the system's peak demand over previous years but glossed over the fact that — despite the increase in peak demand — the system's overall demand for electricity declined.9 The district court dismissed plaintiffs' dissatisfaction over the presentation of these facts by holding, unequivocally, that "[e]mphasizing and glossing do not create liability." However, by refusing to even consider plaintiffs' complaints about the form of Middle South's disclosures, the court erred.

Contrary to the district court's conclusion, courts interpreting the securities laws have long recognized that reviewing the context in which a disclosure appears is an essential part of determining the disclosure's adequacy. The Second Circuit has specifically recognized the importance of such a review in determining whether information is "misleading" to investors within the meaning of section 11 of the 1933 Act. In Greenapple v. Detroit Edison Co., 618 F.2d 198 (2d Cir.1980), the question was whether the "gloss" Detroit Edison placed upon its use of AFUDC in a prospectus rendered the prospectus misleading. The court began its analysis by noting that the "factual nature of the disclosure" was not at issue; instead, the focus was on "the propriety of the explanatory and collateral references" to AFUDC which the prospectus contained. Id. at 205. Foreclosing the argument that a lack of dispute over either the truth or presence of information means that the securities laws are not implicated, the court explained that "notwithstanding the broad discretion which issuers have in assembling and organizing their data, where the method of presentation obscures or distorts the significance of material facts, a violation of Section 11 will be found." Id. Later in the opinion, the court eloquently explained why consideration of the way in which material is portrayed in disclosure documents is so important:

Id. at 210 (citations omitted).

Just as the Second Circuit found context to be instrumental in determining whether a statement is "misleading" within the meaning of section 11 of the 1933 Act, we have recognized the part manner of disclosure plays in determining whether statements are "misleading" within the meaning of Rule 10b-5. In Smallwood v. Pearl Brewing Co., shareholders of Pearl Brewing Co. challenged under section 10(b) of the 1934 Act and Rule 10b-5 the adequacy of disclosures in a proxy statement informing Pearl's shareholders of the terms of a proposed merger between Pearl and Southdown, Inc., 489 F.2d 579 (5th Cir.), cert. denied, 419 U.S. 873, 95 S.Ct. 134, 42 L.Ed.2d 113 (1974). Specifically, the shareholders complained that the proxy materials were misleading because of how they disclosed a particular provision of the merger agreement — a provision which gave Pearl the power to waive an underwriting commitment requirement the agreement also contained. At trial, the jury found that the proxy statement did not fail to disclose adequately Pearl's power to waive the underwriting commitment. On appeal, we were asked to overturn the jury's finding as clearly erroneous. In analyzing the sufficiency of Pearl's disclosure, we recognized that

Id. at 602. Translating this general standard into a workable method of evaluating claims questioning the manner of disclosure, we held that "[a]dequacy of disclosure is a function of position, emphasis, and the reasonable anticipation that certain future events will occur." Id. at 605.10 Moreover, we found the way in which information is disclosed to be so important that we refused to "accept the premise that prior disclosure in one communication will automatically excuse omissions in another." Id. As we explained:

Id. at 605-06.

As these cases11 make clear, emphasis and gloss can, in the right circumstances, create liability under both section 11 of the 1933 Act and Rule 10b-5. The district court erred, therefore, if it dismissed — just because they appeared to focus on form instead of substance — plaintiffs' allegations which questioned the way in which Middle South presented specific information. The district court should have judged these allegations just as it judged those of plaintiffs' allegations which questioned the content of Middle South's documents.

2. ARE PREDICTIONS ACTIONABLE?

For similar reasons, we find fault with the district court's conclusion concerning another broad category of plaintiffs' allegations. In granting summary judgment, the district court refused to consider those of plaintiffs' allegations which "relate[d] to predictions or the failure to make them." Just as it viewed complaints about the manner in which information is disclosed as not actionable under the securities laws, the court concluded that predictions, whether made or not, are "simply not the stuff" from which a securities claim can be fashioned. We find the district court's generalization to be overbroad and, therefore, wrong. Neither section 11 nor Rule 10b-5 exempts predictions, as a category, from its reach; instead, both the section and the rule explicitly offer protection against a certain kind of statement: one that is misleading because it either fails to state a material fact or states a material fact falsely. Therefore, support for the district court's generalization — to the extent it exists — must come from case law. The case law relating to predictions, however, quickly demonstrates two things: first, that the district court was clearly wrong not to differentiate between predictions which a company makes in its disclosure documents and predictions which a company does not make; and second, that the matter of whether and under what circumstances predictions are actionable is today a complicated and evolving area of securities law.

Courts in the past have consistently recognized that a defendant does not place itself beyond the reach of the securities laws merely by disclosing information that is predictive in nature. For example, when necessary, courts have readily conceded that predictions may be regarded as "facts" within the meaning of the antifraud provisions of the securities laws. See Marx v. Computer Sciences Corp., 507 F.2d 485, 489 (9th Cir.1974); Abrams v. Oppenheimer Gov't Sec., Inc., 589 F.Supp. 4, 9 (N.D.Ill.1983); Eichen v. E.F. Hutton & Co., 402 F.Supp. 823, 829 (S.D.Cal.1975). Most often, whether liability is imposed depends on whether the predictive statement was "false" when it was made. The answer to this inquiry, however, does not turn on whether the prediction in fact proved to be wrong; instead, falsity is determined by examining the nature of the prediction — with the emphasis on whether the prediction suggested reliability, bespoke caution, was made in good faith, or had a sound factual or historical basis.12 See, e.g., Luce v. Edelstein, 802 F.2d 49, 56 (2d Cir.1986); Eisenberg v. Gagnon, 766 F.2d 770, 775 (3d Cir.), cert. denied, 474 U.S. 946, 106 S.Ct. 342, 88 L.Ed.2d 290 (1985); Goldman v. Belden, 754 F.2d 1059, 1068-69 (2d Cir.1985); First Virginia Bankshares v. Benson, 559 F.2d 1307, 1314 (5th Cir.1977); Polin v. Conductron Corp., 552 F.2d 797, 804-07 (8th Cir.), cert. denied, 434 U.S. 857, 98 S.Ct. 178, 54 L.Ed.2d 129 (1977); Marx, 507 F.2d at 490; G & M, Inc. v. Newbern, 488 F.2d 742, 745-56 (9th Cir.1973); Alfaro v. E.F. Hutton & Co., 606 F.Supp. 1100, 1104 (E.D.Penn.1985). Because a clear body of law exists which recognizes those circumstances where liability can be imposed under the securities laws for disclosed predictions, the district court obviously erred in concluding that disclosed predictions are never actionable.

Undisclosed predictions, however, present a different problem. Until the early 1970's, the SEC prohibited companies from including most predictive information in the documents they filed with the SEC. See Disclosure of Projections of Future Economic Performance, Securities Act Release No. 5362, [1972-1973 Transfer Binder] Fed.Sec.L.Rep. (CCH) ¶ 79,211 at 82,666 (Feb. 2, 1973) [hereinafter Securities Act Release No. 5362] ("It has been the Commission's long-standing policy generally not to permit projections to be included in prospectuses and reports filed with the Commission."); see also Adoption of Amendments to Proxy Rules, Exchange Act Release No. 5276, Fed.Sec.L.Rep. (CCH) ¶ 76,380 (Jan. 17, 1956) (listing "predictions as to specific future market values, earnings [and] dividends" as examples of what may be misleading in proxy statements). Courts generally recognized this policy and followed the SEC's lead by holding that the failure to disclose predictive information was not actionable under the securities laws. See, e.g., Rodman v. Grant Foundation, 608 F.2d 64, 72 (2d Cir.1979) ("Full factual disclosure need not be embellished with speculative financial predictions."); Marsh v. Armada Corp., 533 F.2d 978, 987 (6th Cir.1976), cert. denied, 430 U.S. 954, 97 S.Ct. 1598, 51 L.Ed.2d 803 (1977); Arber v. Essex Wire Corp., 490 F.2d 414, 421 (6th Cir.), cert. denied, 419 U.S. 830, 95 S.Ct. 53, 42 L.Ed.2d 56 (1974). In 1973, however, the SEC determined that "changes in its present policies with regard to the use of projections would assist in the protection of investors and would be in the public interest." Securities Act Release No. 5362 at 82,667. As the SEC explained: "Persons invest with the future in mind and the market value of a security reflects the judgments of investors about the future economic performance of the issuer. Thus projections are sought by all investors, whether institutional or individual." Id. In 1978, after taking a series of steps to implement this change of policy, the SEC issued a statement encouraging, although not requiring, disclosures of management projections both in filings with the SEC and in general. See Guides for Disclosure of Projections for Future Economic Performance, Securities Act Release No. 5992, [1978 Transfer Binder] Fed.Sec.L.Rep. (CCH) ¶ 81,756 (Nov. 7, 1978). The SEC's policy of encouraging the disclosure of projections is presently set forth in, among other places, its instructions for complying with Regulation S-K, see 17 C.F.R. ¶ 229.10(b) (1987), and certain safe harbor rules which it adopted in 1979, see supra note 12.13

Because of the SEC's change in policy, courts have begun to re-examine the position that a company's decision not to disclose predictive information is always immune from scrutiny under the securities laws. As a result of these re-examinations, courts have adopted a variety of approaches to dealing with the problem of undisclosed predictions. These approaches range from continuing a prior practice of never requiring that predictive information be disclosed to formulating new standards which recognize that sometimes disclosure will be required; the approaches vary significantly among the circuits and sometimes within a circuit depending on the type of predictive information at issue.14 See Walker v. Action Indus., 802 F.2d 703, 707-09 (4th Cir.1986), cert. denied, ___ U.S. ___, 107 S.Ct. 952, 93 L.Ed.2d 1000 (1987) (reporting the Third, Sixth, Seventh, and Ninth Circuits' approaches to disclosing financial projections and "other soft information such as asset appraisals," and deciding on an approach for the Fourth Circuit when financial projections are involved); see also J. Kerr, A Walk Through the Circuits: The Duty to Disclose Soft Information, 46 Md.L.Rev. 1071, 1076-109 (1987) (reporting on the approaches of the Second, Third, Fourth, Sixth, Seventh, and Ninth Circuits); B. Hiler, The SEC and the Courts' Approach to Disclosure of Earnings Projections, Asset Appraisals, and Other Soft Information: Old Problems, Changing Views, 46 Md.L.Rev. 1114, 1157-166 (1987) (analyzing the approaches adopted by the Third, Fourth, Sixth, and Seventh Circuits and proposing an approaches articulated by the Second Circuit prior to the SEC's change of policy).

We make no attempt today either to examine in detail the several approaches which the various circuits have endorsed or to determine when, or even if, predictive information must be disclosed in our circuit. Instead, we use these decisions from other circuits to make a point: that in each case, regardless of the particular approach the court finally followed, each court made its decision by recognizing the statutory policy interests both in requiring and not requiring disclosure, by focusing on the nature of the particular predictive information whose lack of disclosure was questioned, and by determining what the specific facts surrounding that predictive information indicated about the information's importance, reliability, or impact on the potential investor. What is required on remand, therefore, is the same kind of careful, sensitive inquiry in determining whether those predictions plaintiffs challenge that Middle South did make and those predictions plaintiffs challenge that Middle South did not make are actionable under either section 11 of the 1933 Act or Rule 10b-5. Such an inquiry is all the more appropriate in this case because of the central role that predictions have played in the history of Middle South.15

3. WERE MIDDLE SOUTH'S DISCLOSURES ADEQUATE AS A MATTER OF LAW?

The district court insisted that the way in which a company characterizes or interprets underlying facts cannot subject the company to liability under the securities laws. The district court also insisted that a company cannot be liable under the securities laws for either making predictions or failing to make predictions. Despite its insistence on these points, the district court, according to its opinion, still performed (albeit in pectore) a substantive evaluation of Middle South's disclosures to determine their adequacy.

On appeal, plaintiffs argue that in making this substantive review, the district court engaged in conduct which is contrary to accepted summary judgment procedure: the court decided inferences, ambiguities, and disputed interpretations in defendants' favor and, in so doing, improperly resolved genuine issues of material fact about the disclosures' adequacy. The Middle South defendants argue in favor of the court's judgment by, as they did below, quoting extensively from the various documents they filed with the SEC from 1981 to 1985; they argue from the excerpts, as they did below, that Middle South accurately and adequately disclosed each of the "negative investment factors" which the defendants see as the basis of plaintiffs' Complaint. We begin our review of the court's evaluation with a look at the legal standards which the court used in conducting its evaluation.

The district court granted summary judgment because "no reasonable fact-finder could conclude that the information published by defendants was misleading to such an extent as to make them liable to purchasers of [Middle South] stock." This conclusion demonstrates that the district court both knew and applied the proper standard for summary judgment. As the Supreme Court recently explained, "[the summary judgment standard] mirrors the standard for a directed verdict under Federal Rule of Civil Procedure 50(a), which is that the trial judge must direct a verdict if, under the governing law, there can be but one reasonable conclusion as to the verdict." Anderson, 477 U.S. at 250, 106 S.Ct. at 2511. In this case, of course, the governing law comes from the federal securities statutes — specifically, section 11 of the 1933 Act and Rule 10b-5, promulgated pursuant to section 10(b) of the 1934 Act. The district court indirectly explained its understanding of the governing law, putting aside for the moment discussions of prediction and characterization, when it found that "a reasonable potential investor who would have read any of these documents as a whole would have been fairly informed about the negative investment factors which plaintiffs contend were misstated or omitted by defendants."

Working backward from the district court's conclusion, we have, we think, come up with the standard it applied for determining generally the adequacy of a disclosure: whether the information disclosed would have been misleading, on those points about which the information's adequacy is questioned, to a reasonable potential investor who read the information as a whole. We agree with the court that this standard is correct. See Durning v. First Boston Corp., 815 F.2d 1265, 1268 (9th Cir.), cert. denied, ___ U.S. ___, 108 S.Ct. 330, 98 L.Ed.2d 358 (1987); Admiralty Fund v. Hugh Johnson & Co., 677 F.2d 1301, 1306 (9th Cir.1982); cf. Associated Builders, 505 F.2d at 98 (holding that district court was correct to grant defendant's motion to dismiss where a reasonable fact-finder could not find that a prospectus that plaintiff appended to the complaint was misleading). The objective nature of this test, of course, comes from the definition of materiality the Supreme Court adopted in TSC Industries, Inc. v. Northway, Inc., 426 U.S. 438, 449, 96 S.Ct. 2126, 2132, 48 L.Ed.2d 757 (1976): whether "there is a substantial likelihood that a reasonable shareholder would consider it important in deciding how to vote."16 As the Ninth Circuit explained, when the question is whether a disclosure was adequate, "[t]he application of the reasonable investor test is appropriate because adequacy of disclosure, like materiality, only can be judged against an objective standard of investor behavior." Durning, 815 F.2d at 1268 n. 4. In TSC Industries, the Court also admonished that its standard of materiality contemplates

426 U.S. at 449, 96 S.Ct. at 2132. Thus, the test for determining the adequacy of disclosure also focuses on the disclosure as a whole.

Generally, like materiality, determining whether information has been adequately disclosed is a mixed question of fact and law and, therefore, is a question for a jury. Durning, 815 F.2d at 1268. The reason is that "`[t]he determination requires delicate assessments of the inferences a "reasonable shareholder" would draw from a given set of facts and the significance of those inferences to him, ... assessments [which] are particularly ones for the trier of fact.'" Id. (quoting TSC Industries, 426 U.S. at 450, 96 S.Ct. at 2132-33). Consequently, we will only remove the question from the jury if the disclosure is so obvious that reasonable minds cannot differ. Id. The district court, as we explained, found that this difficult standard was met for all "negative investment factors" upon which plaintiff's complaint focused. It is that conclusion which, at long last, we are called upon to review.

As we see it, a court concludes that a disclosure is adequate as a matter of law by engaging in a three-part analysis. First, the court determines, based upon the facts of a particular case, the scope of a particular defendant's disclosure-duty under the securities laws. The court makes this determination by responding to four questions, the answers to which depend on the particular facts before the court: what information had to be disclosed, where did the information have to be disclosed, when did the information have to be disclosed, and how did the information have to be disclosed? During the second part of the analysis, the focus shifts from the disclosure which should have been made to the disclosure which was made, and the court's questions reflect the change: what information did the defendant disclose, where was the information disclosed, when was the information disclosed, and how was the information disclosed? And, as the last step, the court compares the disclosure the defendant should have made against the disclosure it did make to see if the defendant's disclosure was "as a matter of law" adequate under the securities statutes. While the steps of the analysis can be just this easily set out, performing the analysis can be a significantly more imposing task. As the number of disclosures challenged and the number of documents in which the disclosures appear increase, as the facts which define the scope of the defendant's duty to disclose become more complicated, and as the information sought itself becomes more or less obviously important, more or less obviously certain, more or less obviously present, the task of deciding whether a defendant's disclosures are adequate as a matter of law can take on almost herculean proportions.

In this case our review is complicated by more than just the fact that we are faced with a challenge to over forty different disclosures, involving at least eighteen different documents and based on a complex but sketchy factual scenario which unfolded over a fifteen year time period. In this case, our review is also difficult because we know the district court's ultimate conclusion — that Middle South's disclosures were adequate as a matter of law — but have not been favored with any detailed analysis supporting that conclusion.

In its opinion, the district court concluded that a reasonable potential investor who would have read any of Middle South's documents as a whole would have been fairly informed about the negative investment factors which form the basis of plaintiffs' suit. As to what information Middle South had to disclose, therefore, the court apparently decided it was "the negative investment factors" which form the basis of plaintiffs' complaint. As to how the information had to be portrayed, the court apparently decided that Middle South's disclosure documents had to "fairly inform" the investor. And as to when and where the information had to be included, the court apparently decided that it was in "any of Middle South's documents as a whole." In theory, these findings, which we can adduce from the district court's ultimate conclusion, should together define the scope of the duty against which the court compared Middle South's disclosures; in practice, however, they only leave us with questions which make us unable to review the court's conclusion at all. For example, as even a quick look at plaintiffs' complaint will confirm, plaintiffs' allegations do not divide unequivocally into a discrete list of "negative investment factors;" instead, the allegations are interdependent, susceptible to interpretation, and therefore capable of being characterized in more than one way.17 To review the district court's conclusion that these factors were adequately disclosed, we would first have to guess at what the court found these factors to be. This would be especially difficult given that although the court claimed to be looking for all of the negative factors on which plaintiffs' complaint focused, the court also expressly, and mistakenly, found that all of plaintiffs' allegations which "relate[d] to predictions or the failure to make them" were not actionable. The obvious question, of course, is whether the district court's list of "negative investment factors" included or excluded those factors which have about them an element of prediction. The court did not say.

Another of the court's findings comes into play when we consider the court's decision that Middle South's documents "fairly informed" the potential investor of these factors. Earlier we explained that the district court mistakenly found that the way in which a defendant portrays factual information cannot subject that defendant to liability under the securities laws. The next obvious question, therefore, is whether the district court considered or ignored, when it determined that the reasonable potential investor was "fairly informed," the way in which the information was portrayed to that investor. Again, we could only guess at an answer. Finally, we cannot tell how the district court decided that the proper information was disclosed "when" and "where" it had to be disclosed. Although the court purported to evaluate the disclosures from the point of view of one "who would have read any of these documents as a whole," the court clearly did not at any time have the "whole" of any document before it; instead, as we have explained, the court had only the excerpts included in the complaint and the few pages of each document that the Middle South defendants attached to the Lupberger affidavit. Did the court mean, therefore, that a reasonable potential investor who read all eighteen documents generated over a period of three years would, at the end of that three year period, have been fairly informed about the negative investment factors? If that is not what the court meant (and, if the court performed a proper analysis, it should not be), how did the court determine which documents had to contain what information at what point in time? Again, to review the court's finding, we would have to guess.18

Because we do not know how the district court arrived at its conclusion that Middle South's disclosures were adequate as a matter of law, we are seriously handicapped in reviewing that conclusion; any attempt to do so would necessarily result in our conducting a separate and independent evaluation of all of Middle South's disclosures. Of course, since we are reviewing a summary judgment, we have the power to make such an independent evaluation. See Reid v. State Farm Mut. Auto Ins. Co., 784 F.2d 577, 578 (5th Cir.1986); Turbo Trucking Co. v. Those Underwriters at Lloyd's London, 776 F.2d 527, 529 (5th Cir.1985). We decline, however, to exercise that power today.

We are fundamentally a court of review, not of first analysis, and we have in part performed our fundamental role: in order to respond to certain arguments raised by the parties, we pieced together the procedural puzzle of this case, and we laid to rest several basic misconceptions which, at least in part, led the district court to find that Middle South's disclosures were adequate as a matter of law. For us to go further — that is, for us to measure Middle South's disclosures for the first time by the proper standards — would, we find, require an enormous additional commitment of time and resources.

As we have explained, plaintiffs challenged the adequacy of over forty different disclosures contained in twelve different documents filed over a period of several years. Some disclosures were challenged because they failed to give the potential investor information which was, to some extent, predictive in nature; to evaluate those disclosures, therefore, we would have to first determine whether and under what circumstances Middle South had a duty to disclose predictions — a complicated and difficult inquiry in and of itself. The resolution to that initial inquiry, moreover, would be only a small part of our task. We would then have to engage in the three-part analysis discussed earlier: we would have to determine, for each disclosure challenged, the scope of Middle South's disclosure duty and the extent of Middle South's actual disclosure, and decide whether any discrepancy existed between what should have been disclosed and what was disclosed. This inquiry is also complicated. One reason is that, as we have explained, the test for determining the adequacy of a disclosure focuses on the disclosure as a whole; it is only if a fact is inadequately disclosed in the "total mix" of information available to the potential investor that liability arises. The lens through which we would have to determine the adequacy of any one disclosure, therefore, would necessarily be fixed to the point at which the disclosure was made. Fixed at that point in time, we would have to expand the lens to include all the information which was available to the investor before that point and contract the lens to keep out all the information which came after. In essence, therefore, we would have to judge each disclosure independently of each other disclosure — and that would mean determining over forty different times what Middle South's duty was and what information Middle South had, up to that time, disclosed.

Although we sometimes act beyond our traditional role as a reviewing court and decide issues in the first instance, we cannot justify doing so here. The expense — measured by the drain on our judicial resources which evaluating this raw, complex lawsuit would take — is too great. We have therefore determined to vacate the district court's judgment and to remand the case to the district court for further development. On remand, however, the district court has more than one option. It can remain on the same procedural course and reevaluate whether, with the benefit of certain corrections we have made, the defendants are entitled to judgment as a matter of law. If the court elects to follow this path and once again to enter summary judgment for the defendants, it should provide a detailed analysis supporting that result so that a reviewing court can properly perform its function. Alternatively, the district court can decline to rule on the defendants' summary judgment motions until further development has taken place and the issues in the case have been narrowed through the available pre-trial techniques.

III.

For the foregoing reasons, the judgment of the district court is VACATED, and this case is REMANDED to that court for further proceedings consistent with this opinion. Costs to be borne by defendants.