November 18, 1999

Mr. Jonathan G. Katz
Secretary
Securities and Exchange Commission
450 Fifth Street, N.W.
Washington D.C. 20549

Re: Release No. 34-41987
File No. S7-22-99

Dear Mr. Katz:

Cell Pathways, Inc. appreciates this opportunity to comment upon the necessity, appropriateness and effectiveness of the proposals contained in the Release in the context of trying to prevent the type of financial fraud described in the Release.

The comments of this letter may reflect the concerns of the hundreds of small publicly traded companies which do not have product sales or earnings. These comments may also reflect the considered thoughts of large, established companies which have both a need, and a thorough understanding, of techniques to prevent financial fraud.

This letter supports some of the proposals, opposes others. Where opposition is voiced, the opposition is not to the practice in question but to the mandating of the practice. Universal mandate of good practice in the absence of compelling need tends to stifle, to restrain and to increase costs - none of which represents the wise use of government power. The several securities marketplaces are currently experimenting with various measures; experience gained from this experimentation will be important.

The Commenting Company

Cell Pathways was founded ten years ago to develop pharmaceutical agents to prevent cancer. It may be the only public company founded for this purpose. Its programs have since expanded to include cancer therapy, as well. It has no sales, no earnings. Some day we hope it will. It has about 60 employees. It formed an audit committee of outside directors about two years before becoming a public company.

The Premises and Purposes of the Release

The Release discusses a number of laudatory practices. In a number of instances, the Release fails to make the bridge from laudatory practice to the need for mandatory universal imposition of the practice. This disconnect is not unusual in the ongoing cycle of scandal/reaction/reform proposal. Indeed, the comment process is designed to examine the premises, the purposes and the connective bridges to purposeful and effective regulation.

The Release joins the broad front of current activism seeking to "promote quality financial reporting." We endorse and support that effort. We disagree that certain laudatory practices should be made mandatory. The discussion starts with the premises.

Non-Helpful Premises

Premises which over-promise will disappoint. The premises of the Release, as borrowed from expert bodies or stated as original propositions in the "Executive Summary" and "Background," over-promise, over-state. Such overstatement does not contribute to an understanding of the problem to be addressed. Without unduly extending this comment letter with a detailed critique of the opening sections of the Release, here are a few illustrative observations.

The Release endorses the statement that "quality financial reporting can only result" from a host of effective interrelationships among reviewers and second-guessers. This trashes individual ability, integrity and responsibility. This abandons the QA/QC ethic of "getting it right the first time." This undermines the very building blocks upon which the system must rely in order to be effective - the daily, principled attention to detail, the lack of which no system of reviewers can adequately remedy.

The Release characterizes the audit committee essentially as a communicator among various acting parties and suggests the need to improve public disclosure "relating to the functioning" of audit committees. Veterans of audit committees do not see themselves in such an Oprah Winfrey role. They understand that theirs is a much more direct, substantive and serious task.

The Release relegates the CEO and CFO to a quasi-criminal class of persons whose principal attributes are that they have no credibility and must be monitored. Only "effective oversight" can "preserve ... integrity" and "assure ... credibility." This is highly reminiscent of the wry humor formerly directed at the many Russian guards at the Russian War Memorial in West Berlin: which guards were guarding the Memorial and which guards were guarding the guards (so they would not defect)?

Helpful Premise

While the overstatements and over-promises of the "Executive Summary" and "Background" do not contribute to an understanding of the problem to be addressed, these sections otherwise highlight and underscore the problem - the manipulation of earnings to meet market expectations. With refinements and acceleration of the communication process, analysts relentlessly predict quarterly corporate performance to the penny. Registrants fail to meet these expectations at their peril. The snake, the apple and the temptation are all there.

So, how do we prevent fraudulent manipulation of earnings in this environment? One wag has suggested outlawing specific earnings projections by analysts. There would thus be no target to which to manage earnings. While we do not support this proposal, it does ask questions about the environment.

Under such a hypothetical ban, analysts would be free to comment on all of the factors affecting industries and individual companies. They could make comparisons and recommendations of the usual variety. They simply could not yell "fire" in a crowded theatre (issue specific dollar and cent earnings projections). Analysts' First Amendment rights would be intact; they simply could not engage in the manipulatory practice of issuing specific dollar-and-cent earnings projections, for the well reasoned considerations which would be articulated in the hypothetical promulgating release.

Such a ban on specific earnings projections should not be necessary. We favor, instead, the free flow of ideas and information. We share the optimistic view that, even in the face of unrelenting earnings projections, mature managements can - and do - act in a responsible fashion.

An aspect of managed earnings which the Release might have discussed a bit more is that, in some industries and in many companies, earnings must be managed in order not to be misleading. For businesses characterized by long-term contracts whose performance stretches out over five or ten or more years, quarterly reporting of financing results is inherently misleading; even annual reporting is on too short a cycle to reflect true financial performance. Accounting principle has tried to accommodate this by the use of estimates. Viewed from a different perspective, these estimates manage earnings. And they must - or the resulting financial statements will be false and misleading.

This is also true, mutatis mutandis, of conglomerates. The truth of the conglomerate is not the literal reflection of blips in each piece of the enterprise as of arbitrary (to the business) monthly, quarterly or annual dates; rather, it is the enduring effect of the many transaction/financial cycles at intervals which may not coincide with the neatness of calendar periods. Again, accounting principle comes forward with an estimating process. Again, the estimating process must be seen, with other glasses, as managing earnings.

And, in the gray area where applicable accounting principle may permit the exercise of judgment, the question is often asked as to whether the stockholders and the marketplace are better served by (a) the jagged ups and downs of quarterly earnings which reflect period accounting entries literally without taking into consideration the relevant transaction/financial cycle(s), or (b) quarterly earnings which more nearly reflect the longer term trend of the transaction/financial cycle(s) which the financial statements are ultimately trying to reflect (assuming, of course, no distortion of either aggregate earnings/losses or trend data over the relevant financial cycle)?

As the Release indirectly recognizes, there is legitimate or necessary management of earnings through application of the estimating process mandated by accounting principle. And, in the words of the Release, there is "inappropriate" earnings management. It was not incumbent upon the Release to provide a full treatise as to the distinctions between "appropriate" and "inappropriate" "earnings management." On the other hand, appreciation of the distinctions is helpful in evaluating whether one or more of the proposals in the Release would be necessary or appropriate or effective in preventing fraudulent earnings management.

The Key Premise, Purpose and Question

Implicit in the use of federal government securities regulation to stop inappropriate earnings management is the intent to protect stockholders from swings in stock price affected by such fraud. Initially, the fraud is perceived as contributing to a higher (or less low) stock price on the basis of rosier earnings reports. When the fraud inevitably unravels, the stock price drops noticeably. One group of buyers and sellers is favored by the fraud, and another group of buyers and sellers is disfavored.

If the fraud of inappropriate earnings management could be prevented by government regulation, then earlier rosier earnings reports would be prevented, the stock would not rise so high (or fail to fall so low) and there would be no later unraveling of the fraud to drop the stock price later. Under this fairer scenario, different groups of buyers and sellers are favored and disfavored when compared with the fraud scenario.

The key questions then ask: How much of our resources should be devoted to preventing fraud? How much to providing remedy for the fraud? And how much to punishing the fraud? And what are the relative cost, effect and benefit of each component of each approach? What are the differences in outcomes? Where are the burdens shifted? Are these shifts equitable and efficient?

The response of the Blue Ribbon Committee, as quoted in the Release, begs these key questions:

"Improving oversight of the financial reporting process necessarily involves the imposition of certain burdens and costs on public companies. Despite these costs, the Committee believes that a more transparent and reliable financial reporting process ultimately results in a more efficient allocation of and lower cost of capital. To the extent that instances of outright fraud, as well as other practices that result in lower quality financial reporting, are reduced with improved oversight, the benefits clearly justify these expenditures of resources."

The cost of preventing one fraud should be ...? Of two frauds ...? The effect of the deterrent step is ...? The benefit of deterring the fraud relates to the cost in the following way...? The overall effect on buyers and sellers in the securities markets is ...? The overall benefit to society is ... ?

Begging the question does not support increased government regulation. It does, however, serve as a reminder of the old saw about being careful where one seeks advice. Lawyers are prone to suggesting lawsuits as a solution; surgeons, surgery; etc. Perfection in financial accounting is a desirable goal. However, it should not be pursued at the cost of encumbering securities markets with non-productive or counter-productive processes.

Enough discussion of premises and purposes. This letter will now turn to discussion of whether the individual proposals of the Release address the problem of inappropriately managed earnings in a necessary, appropriate and effective way and whether the substance of the proposals should become mandatory universal practice for all reporting companies.

Auditor Review of Interim Financial Statements on Forms 10-Q and 10QSB.

At the outset, it should be noted that this proposal, if enacted, would represent merit regulation. This would not be unprecedented; audits are already required for registration and on an annual basis. Any proposed merit regulation must meet a heavy burden of demonstrating need, effectiveness and proportionality.

Would quarterly auditor reviews diminish the incidence of improperly managed earnings? Possibly. Would quarterly audits cut down on improperly managed earnings? Possibly. But neither is calibrated as an anti-fraud process and neither would necessarily be effective.

We are trying to combat fraud, which, to a large extent, is beyond the effective scope of either a review or an audit. As the Commission's recent increase in enforcement actions is undoubtedly documenting, earnings management is a subtle, sophisticated process. Only amateurs would attempt to manage earnings with the type of adjustment which would be examined by a review or an audit.

We can throw time, money, inside professionals and outside experts at additional reviews and audits. We can feel that increased auditor participation on a quarterly basis may breed a more effective continuing relationship which will lead somehow to financial statements which look a bit more like what the auditors would have created themselves had they been acting in the role of accountant.

But we are not going to catch sophisticated fraud this way. Fraud is combated not with customary reviews and audits but, in the first instance, with Treadway's "Tone at the Top." And the "Tone at the Top" should be abetted by the internal audit process and a supportive audit committee. Anti-fraud processes cycle from the top of the company to the bottom and back again. Normal audits and reviews are not anti-fraud processes. Adding additional outside reviews and audits to the present regulatory scheme will burden the honest, not deter the crooked, and breed cynicism rather than confidence in our financial reporting system.

In short, the link has not been made between the cost and the supposed benefit, between the proposed non-fraud quarterly reviews and the effective detection of earnings fraud. The case for governmental regulation mandating quarterly review lacks requisite support.

The large auditing firms are currently experimenting with the practice of quarterly review. Let that continue. The auditors understand that they are not to become so familiar that they become the accountant of first impression, rather than the auditor. They also understand the cost factor to their clients. They also understand the risk of displacing those functions which they are supposed to be auditing.

For small companies without sales and earnings, mandated quarterly reviews would be entirely out of proportion. There are no earnings to manage. There are not even sales to manage. There is precious little money to develop the business, let alone pay for an extra set of Russian guards.

Large, established companies may welcome the practice of quarterly auditor reviews at certain phases of their business cycle or certain phases of their corporate development. At other phases, they will see waste of resources - expenditure of time and money with no compensating return. It is a highly individual matter, depending upon the nature and needs of the enterprise.

The notion of mandated quarterly auditor reviews is currently embraced with the same fervor with which "one share - one vote" was embraced in the 1980's as the cure for abusive anti-takeover tactics. A brief comparison is in order.

What is wrong with "one share - one vote"? Nothing. It continues as predominant corporate practice. But, as this Commission, and also the Proxmire Committee of the United States Senate, came to understand during hearings in the late 1980s, "one share - one vote" should never be made a mandatory practice. It is not a particularly appropriate measure to prevent abusive defensive tactics. Moreover, it would mandate uniform capital structure - an unwarranted, drastic and unintended result of a superficially attractive idea - and would have precluded public ownership of a variety of leading businesses, ranging from Hershey Foods Corporation (owning trust precluded from diluting its vote) to the great newspapers of the country (which had limitations on stock voting for fear of having editorial policy dictated by outside stock voting interests) to entrepreneurial enterprises (who saw success only in continuing control for limited periods). Accordingly, the simplistic "one share - one vote" notion was supplanted by serious and proper inquiry into more focused techniques for preventing fraudulent disenfranchisement of stockholders.

What is wrong with interim auditor reviews? Nothing. It is a growing and experimental practice. But the lack of demonstrated need, the lack of demonstrated effect, and the law of unintended consequences all counsel against mandating the practice through government regulation of the securities markets. It isn't going to detect or deter the sophisticated process of earnings management. It will simply become embedded as one more expense of publicly held corporations, large and small, foreign and domestic, seeking to tap the capital markets of this country.

The cost for small companies, in particular, is unwarranted. Small companies should not be spending their money this way. Many have neither the sales nor the earnings to generate the problems intended to be reviewed by the process.

The experience of the large companies should prove instructive. Are the auditing firms capable of taking on the added volume of work? Do they have enough quality people to benefit their clients? Do they have the time? Do mature corporations find this interaction productive and helpful? In which respects? With what drawbacks? The current experimentation with large corporations will be instructive as to both the worth of the practice and the ability of the auditing profession to deploy quality resources to implement the process. And perhaps this process will yield data with respect to effectiveness in sniffing out "inappropriate" earnings management.

Additional considerations counsel against mandated auditor review of earnings prior to public release. And against mandated filing of earnings prior to public release. Forcing all earnings activity into a prior review straitjacket would confirm the CEO and CFO as members of a quasi-criminal class not worthy of belief. It would jam all earnings information into narrow windows of time where the investing pubic will be overwhelmed and unable to discern and digest. It would exalt process above any demonstrable improvement in outcomes.

In short, whether under the viewpoint of the Regulatory Flexibility Act or the dictates of regulatory prudence, the Commission should stay its hand. Evolution in the private sector is underway. Let this experimentation continue. We are not wise enough to know that another layer of mandated auditor review would represent necessary, appropriate or effective regulatory response to the type of fraud sought to be discouraged.

With respect to the small company dilemma highlighted by the COSO Report, it does not seem intuitively productive to pile added regulation on to small companies because of their thin staffs or perceived incompetence. Let the other disciplines of the real world take care of those problems. Honorable, smart people can run small companies well. Let's not discourage small enterprise by encumbering it with a perfection-seeking- process which is out of proportion to its activity and its capacity to sustain.

Proposed Audit Committee Report

The substance of proposed SK 306 (a)(1) and (b) should be moved to Item 7(e) of Schedule 14A. In fact, it is already part of Item 7 (e).

The proposed requirement of SK (a) (1), which is already largely embraced by Item 7 (e), might be expanded to require disclosure of whether the audit committee reviewed the audited financial statements prior to their public release and/or prior to their acceptance by the company.

Traditional good corporate practice has asked the audit committee to make a recommendation to the board of directors as to whether or not to accept the proposed audited financial statements. The board then has the decision as to whether to accept the financial statements or send them back to management and the auditors for further work. Hence the value of the disclosure suggested in the prior paragraph.

Placing the disclosure of proposed new SK 306 (a)(1) in Item 7(e) of Schedule 14A permits integration into the current proxy statement disclosure of what the committee does and obviates the need for a "report." And (b), names of committee members, is already in the current disclosure fabric.

The remainder of proposed SK 306 (a)(subparts 2, 3 and 4) causes more mischief than it contributes to stockholder well being. The many reasons for this are, we assume, covered extensively on the comment letters of others. They include the fact that the corporation and the board are already liable; why add a "me-too" statement from the audit committee? If the audit committee's report creates reliability, does this create the inference that corporate utterances not so validated are not to be taken as so reliable, and are not to be visited with the same degree of liability?

To an extent, part of the matter enumerated in the proposed disclosure of SK 306 (a) (2) (3) and (4) is already embraced in Item 7 (e) of Schedule 14A. Each registrant must determine what the functions of its audit committee are and disclose them under Item 7 (e). However, requiring disclosure of whether the committee received certain documents or had certain discussions or entertains certain views will provide little of interest to the stockholder and will primarily help four classes of persons: those professionals whose considerable services would be needed to perform, shepherd and monitor this activity; class action lawyers looking for evidence to surmount the pleading requirements under the Private Securities Litigation Reform Act of 1995; regulators looking for a problem or looking for an easier way to attribute a cause to a problem; and the media/entertainment industry that thrives on the prospect of a possible irregularity. Business should not be put in the position of having to explain steps internal to its processes. The taste of sausage is not enhanced by watching it made.

Thus, the concept of a "report" by the committee is unnecessary and unwise.

The Release suggests that "public confidence" in the "reliability" of financial statements "depends on" investors' perceptions. This is a good thought, but it does not aptly apply to the proposals of the Release. Perception does, in fact, become a reality. But we want the substance of the financial statements to be the reality, not the rain dances which accompany their formation. Liturgy is important. But inviting stockholders to ceremony is neither necessary nor productive. Ceremony and liturgy must not displace substance - in either the mind of the investor or the pocket of the investor.

The Release credibly concedes that "the benefits of improved disclosure about the audit committee's communications are not readily quantifiable." Reliance is then placed on the belief that the proposed added disclosures would result in increased investor confidence and, thereby, increased market efficiency. We are concerned that elevating perceptions above reality is not a good idea. We believe that the case is not made for the full extent of the proposed disclosure and that the disclosure amendments to be made should be as discussed above.

Audit Committee Charter Issues

The Release is wisely cautious in its approach. The probable answer to the question of whether there should be specific disclosure as to the existence and substance of an audit committee charter is to defer and to jawbone. Item 7 (e) of Schedule 14A already requires disclosure of the functions of the committee. It does not ask whether these functions are set forth in a by-law or in a board resolution or in a charter or are simply a matter of practice. Each year each registrant is confronted with the task of appropriately disclosing the committee's function as of that time. This works well. Let it continue. And the adopting release may, if appropriate, note that, in not adopting the charter-disclosure proposal, the Commission will be looking more closely at disclosure under Item 7 (e) to determine if certain important generic functions of the committee are being adequately reflected in the disclosure.

We must not fall into the trap of elevating form over substance. There was a time when we all subscribed to the view that nominating committees and boards with majorities of outside directors would cure all corporate ills. Experience has taught us otherwise. Outside boards with all the most advanced procedures have presided over corporate disasters; and inside boards have managed highly successful enterprises. Many of us still favor outside boards with all appropriate committees. But we do not favor government regulation enshrining these procedures and elevating their importance out of proportion to their contribution. Committee charters may have their merits, but their importance should not be exaggerated.

Independence of Audit Committee Members

It is far preferable that audit committee members be independent. Disclosure should be made of the absence of independence. However this disclosure is fashioned, it should not convey the aura of assurance of prevention of fraud. The disclosure should speak to the constitution of the committee, not to its effectiveness in preventing fraud.

Safe Harbor; Increase in Litigation

It is naïve to think that opening up to public scrutiny the process of review of financial statements will not lead to increased litigation. Discussion of possible defenses is weak palliative. The audit committee becomes under the securities laws the equivalent of the unprotected Good Samaritan of our tort law. The Company and the stockholders pay the bills. The interests of society are not enhanced by either increased litigation or prolonged litigation.

The Pursuit of Perfection

In the early days of the movement in support of consumer protection, Ralph Nader was maligned in satirical cartoons of the "safe" bicycle and the "safe" horse - encumbered in each case with sufficient safety devices to defeat utility and dissuade use. The essential truth of the cartoons was obvious: perfection has its costs; life has risks; we should never lose sight of what we are trying to achieve; balance and proportion are essential. These principles have been embodied in tempered legislation and regulations dealing with consumer product safety.

The experience with consumer product safety is instructive on another front. Safe design is not the only weapon; safe design need not be pursued at all costs. Remedy and punishment are also useful weapons. The Private Securities Litigation Reform Act of 1995 enacted important reforms to curb the abuse of certain types of class actions. But it does not protect corporations or individuals from the consequences of the type of fraud sought to be addressed in the Release. Nor does it inhibit the Commission in its enforcement activity. The threat of both private and agency litigation remains a significant behavior modifier to the private sector. Perhaps the Commission should petition the Congress to be permitted to retain the fees it collects so that it might appropriately augment its enforcement resources.

On the one hand, we do not have to rely upon, or insist upon, the perfection of internal corporate processes. On the other hand, we do not have to fear that un-audited or un-reviewed interim financial statements constitute a class of unsafe-at-any-speed lethal weapons which will achieve the mass destruction or meltdown of a naive marketplace. In the event of misbehavior, there is always remedy. In the immortal words of that little figurine so often seen on lawyers' desks, we can always "Sue the bastards!"

From a more global perspective, perhaps our forefathers got it right in 1933 and 1934 when they fixed liabilities on corporations and on directors. They did not try to sort out internal processes or parse out more finely who should do what or how. Corporations and their boards step into the pubic limelight at their peril. This provides them with significant motivation to sort out their own internal affairs in ways which will forestall the legal, financial, social and political consequences of not doing so.

Concluding Observation

Financial fraud happens. It shouldn't. Much group activity has focused on this recently. One should join in the concern. One should not join in the fad and the fashion.

The Foreign Corrupt Practices Act of 1977 provides an interesting example of the consequences of following fad and fashion. Post-Watergate concern for public corruption had been expressed at every responsible level of society during the early 1970's. By 1977, the Congress got involved - with mixed results. On the anti-bribery front, the Congress joined the fanfare and passed statutory provisions which appeared to outlaw for the first time a wide variety of bribery practices. The fact was that these practices had already been outlawed under other federal statutes, but there had not been much enforcement. The newly worded anti-bribery provisions added no legislative substance but did elevate public awareness. Was this a constructive legislative exercise?

On the other hand, the FCPA added new books-and-records-keeping provisions which provided the Commission with a significant new tool with which to combat fraud. A useful addition. And, by most accounts, the Commission has used this power wisely and well.

In the present climate, it is important that the Commission be seen as serious about combating financial fraud, particularly "inappropriate" earnings management. On the other hand, the Commission should stay its hand from imposing by way of governmental regulation mandatory practices which have not been demonstrated to be necessary, appropriate and effective in combating the types of fraud which are the source of the concern.

We appreciate the opportunity to share these views with you.

Respectfully submitted,

Richard H. Troy
Senior Vice President - Corporate Development,
General Counsel and Secretary