Speech by SEC Staff:
This was prepared to introduce a panel on "Penalties and Sanctions for Securities Fraud" at the American Economic Association meetings in Chicago on January 6, 2007. The Securities and Exchange Commission disclaims responsibility for any private publication or statement of any SEC employee or Commissioner. This presentation expresses the author’s views and does not necessarily reflect those of the Commission, the Commissioners, or other members of the staff.
1. Introductory Comments
The subject of “Penalties and Sanctions for Securities Fraud” is attracting considerable attention and controversy in the aftermath of Sarbanes-Oxley and the dramatic increases in monetary sanctions that have occurred in recent years. There has been considerable interest in the use of corporate vs. individual penalties as well as the magnitude of monetary sanctions. At the onset of my remarks I also should emphasize that my remarks reflect my own views, but not those of my colleagues on the staff of the SEC, or the Commissioners. To reinforce this point note that my remarks are not focused upon the current state of securities law, but instead the current conceptual environment in which securities enforcement operates.
A portion of the case for sanctions against corporations reflects the notion that claimants on the cash flows of the corporate entity and especially the firm’s shareholders should have sufficient incentives to protect against fraud being committed during the period for which they have claims on the firm and that the fraud reflects the actions of many individuals in the organization rather than simply that of the few individuals directly involved.1 In many cases it can be difficult to establish individual culpability and indeed, as I discuss later indemnification by the firm of the attorney cost incurred by individual employees can make individual employees even less likely to settle actions by regulators and private litigants and more likely to fight allegations. Of course, the indemnification of monetary sanctions and settlements would have the opposite effect. However, to the extent that the firm’s shareholders were themselves victimized by the fraud, it is not clear that additional sanctions should be imposed upon the firm to be borne by the current shareholders.
2. Indemnification of Executives
An important facet of regulatory penalties and the cost of private lawsuits against executives is that in many cases these amounts are paid by their employer. Much of my presentation will address issues related to indemnification; I will focus on the case when indemnification is part of the manager’s labor contract, whether or not the employee has been terminated. Of course, the firm also can decide to indemnify an employee ex post without an explicit ex ante contract because of the ongoing value of retaining the employee’s services.
Indemnification raises a number of interesting questions about the reaction of regulators to indemnification of the employee (or former employee) by the firm. While my own views about this specific issue are not sharply formed, I do think that it is helpful to view the issue of indemnification, like many others, from the perspective of economic theory. Indeed, this seems particularly appropriate, since the statutory framework itself, under which the SEC sets penalties, permits considerable latitude in that it allows the Commission to consider the best interests of public policy.
From an economist’s perspective I view indemnification from the vantage point of a principal-agent framework in which there are three key decision stages:
1) The contracting stage in which the executive is hired and the terms of employment are determined.
2) The operations stage in which the executive makes operating decisions or actions that can result in financial fraud or wrong doing that affects adversely third parties.
3) The negotiation stage in which the (former?) executive negotiates a settlement with a regulator or private litigants.
Indemnification of executives, including potential payment of attorney’s fees, can be part of an efficient labor contract. This is quite natural in light of (a) the risk aversion of an executive relative to the firm and the capital market as a whole and (b) the potential that regulatory matters or litigation could arise in which the executive would incur potential legal costs or obligations because of either the actions of other firm employees or the complexity of the modern setting of the firm relative to those that arise due to willful misconduct of the individual employee.2 In this sense indemnification is potentially an important feature of an optimal risk-sharing contract, which may help explain its widespread use.
Understanding whether the labor contract is efficient in various situations is an important issue and directly bears upon whether regulators should discourage indemnification in employment contracts, through state laws or self-regulatory organization exchange listing standards.3 Despite the risk-sharing benefit there also is the cost associated with not being able to impose as sharp a set of regulatory incentives at both the operating and negotiating stages in the face of indemnification. For example, at the negotiating stage reduced incentives to settle would lead to greater deadweight losses.
Of course, the regulator can attempt to achieve similar ends ex post by insisting upon non-indemnifiable settlements—either terms that are inherently non-indemnifiable or explicitly attempting to preclude indemnification as part of the settlement. Such tacks will make settlement more difficult, reducing its frequency and distorting the settlement terms in what might appear to be an inefficient fashion. For example, to what extent should regulators structure settlements in non-indemnifiable rather than indemnifiable forms such as “bars” restricting future service as officers or directors of public companies or jail time (which is outside the domain of the securities regulator, but not criminal prosecutors), while monetary sanctions are potentially reimbursable? Of course, restricting the settlement terms would reduce the ability to settle, especially in light of an executive’s ability to fight given the structure of the indemnification. This in turn ties back to the question of whether the regulator should explicitly try to discourage indemnification.4
An interesting related question is whether and to what degree should indemnification and payment of attorneys fees for executives be disclosed by the firm on either an ex ante or ex post basis? In recent years in response to perceived agency problems there has been greater emphasis on the details of the disclosure of executive compensation, as illustrated by the new executive compensation rule adopted by the SEC last summer. It is certainly plausible that indemnification arrangements would be of considerable interest to shareholders. Unlike much of the executive compensation landscape, which focuses upon incentive benefits despite risk-bearing costs, indemnification is arguably motivated by risk-bearing advantages despite the adverse incentive costs.
3. Remarks about Optimal Deterrence
Some of the interesting questions with respect to optimal deterrence relates to its interaction with the efficiency of the ex ante labor contract. An interesting facet of the optimal deterrence literature is that optimal sanctions are larger the lower is the probability of detection—in effect, the harm created by the behavior is grossed up to reflect the probability of detection. Of course, to the degree that the probability of detection is relatively low this can lead to large penalties, potentially far larger than those observed. However, the optimal sanctions can be limited because of the resources of the respondent. This is parallel to the insight in the agency literature that the optimal contract can be influenced by the impact of the risk aversion or limited financial resources of the perpetrator (or more generally, the agent).5 Indeed, it has been suggested sometimes that the threshold to indict firms should be especially high, because a conviction or even an indictment can serve as the death penalty for the firm, even prior to any appeal process. While the context was different because it involved a partnership, the Arthur Anderson demise provides an illustration of the potential severity of the conviction of a large firm; indeed, in that case the Supreme Court overturned the Arthur Anderson conviction, but it was too late for the firm. Yet one more complication in the mix in Anderson was the welfare consequence of the change in the structure of the market for audit services—from five top-tier firms down to four such firms.
One of the main innovations under Sarbanes-Oxley is the payment of “fair funds” sanctions from disgorgement and penalties to fund compensation for the victims of financial fraud rather than being directed to the Treasury. Critics of this have objected that because the fair funds are redistributed to the victims, this can discourage the shareholders from deterrence. On the other hand, given the various free-rider problems intrinsic to corporate governance, questions plausibly arise about whether reliance upon the shareholders for preventing financial fraud is realistic. There also are natural redistributive advantages of directly reimbursing the victims.
4. Concluding Comments
A potentially important overarching issue is the question of how a regulator should evaluate the success of its approach to enforcement. For example, should a successful program lead to a reduction in the extent of financial fraud over time? Can this be measured by enforcement decisions of a regulator or should one look to the actions of other parties, such as the filings or outcomes of class action suits? Even if the sanction regime were optimally designed and working well, it isn’t obvious that this should lead to systematic reductions or especially elimination over time in the extent of undesired outcomes, such as securities fraud. Of course, an increase in sanctions may lead to a reduction in bad behavior,6 as changes in outcomes should reflect changes in regulatory incentives. However, the repeated use of either high sanctions or optimal sanctions does not imply that there should be continuing improvements in the quality of the underlying outcomes.
There are many types of issues that our panelists may wish to address. For example, what do they see as the roles for firm vs. individual sanctions? How are one's perspectives on corporate sanctions (both use and magnitude) potentially affected by various factors such as whether the shareholders of the firm were victims of the fraud in which case they would be injured twice by imposing sanctions on them or alternatively beneficiaries of the fraud, the extent to which the ownership of the company changed during the fraud, the extent to which the firm indemnified the relevant individual perpetrators? What perspectives should guide trying to achieve optimal deterrence? There are a number of interesting features in indemnification provisions. How can the design of these be improved? What is the impact of class action suits on deterrence?7 How can a regulator evaluate the effectiveness of one's approach to sanctions?
I am delighted that we have such a first-rate panel to explore these and related themes. I would like to thank each of the panelists for their willingness to participate. In alphabetic order our first panelist is Allan Kleidon, a highly visible expert witness and consultant, who is Senior Vice President at Cornerstone Research and who previously served as a finance faculty member at Stanford. Our second panelist is Roberta Romano, who is a chaired professor at the Yale Law School and a leading expert on securities law and corporate governance. Christopher Snyder, a senior faculty member at the economics department at Dartmouth, who has done important theoretical research that examines directly the use of corporate sanctions and indemnification,8 is our third panelist. As you can see, our panelists reflect a broad range of experiences.
|Home | Previous Page||