Professor Vikramaditya Khanna

July 31, 2007

Written Statement of
Professor Vikramaditya Khanna
University of Michigan Law School
Before the Committee on the Judiciary
United States Senate
Hearing on Evaluating the Propriety and Adequacy of the OxyContin Criminal Settlement
Tuesday, July 31, 2007

1. Introduction

Chairman Leahy, Senator Specter, and distinguished Members of the Committee, thank you very much for inviting me to testify today regarding the prospect of criminal liability for executives who knowingly introduce defective and dangerous products into the market.

My remarks will primarily focus on the circumstances in which it might be desirable to impose criminal sanctions on executives as a supplement (or complement) to monetary sanctions on corporations. I will also discuss methods for ameliorating concerns raised by the possibility of such sanctions. My discussion is largely based on the analyses in a series of my articles over the last decade or so on liability regimes for corporate wrongdoing.1

My overall conclusions are that a case can, in theory, be made for the imposition of criminal liability on executives for knowingly introducing defective products into the market which, due to the defect, can cause death and serious injury. However, before that course is chosen we should first exhaust other options for deterring corporate wrongdoing (e.g., increasing civil liability for corporations, increasing civil liability for executives). If these civil options have indeed been exhausted then a case can be made for criminal sanctions. Moreover, if we are proceeding with criminal liability then we must carefully draft and apply it to assuage concerns that the proposed criminal liability may lead to even worse results than those it was designed to address.

2. Corporate Wrongdoing & Corporate Liability Regimes: When are Criminal
Sanctions on Executives Desirable?2

Imposing civil monetary sanctions on corporations is an important tool for deterring corporate wrongdoing. However, it is only one tool, amongst many, that we have at our disposal. Others include civil monetary sanctions on executives (or employees);3 criminal sanctions on corporations; criminal sanctions on executives; sanctions of various kinds on associated third parties; and combinations of these regimes. In light of all these options a natural question is which liability tool is best to use and when?

As a general matter, I would recommend relying on civil monetary sanctions against corporations as the primary means of deterring corporate wrongdoing. This is for a number of reasons. First, if we relied solely on sanctions against employees there may be many employees who would be judgment proof - that is, not have enough assets to pay for the harm caused. Such employees would not have appropriate incentives to exercise care in their activities to avoid harm.4 This would lead to riskier products and more of these types of products being made and sold.5 Imposing corporate liability addresses these problems to the degree that the combined assets of the corporation and employee can pay for the harm caused (i.e., that the corporation and employee together are less likely to be judgment proof than the employee alone).

Second, corporate liability can often deter as effectively as direct monetary liability against employees. This is because monetary sanctions on the corporation will motivate shareholders to prevent or deter employees from engaging in harmful acts. This may manifest itself in closer monitoring of employees and modifications to employment contracts providing employees with appropriate incentives. This, in effect, deputizes the corporation to monitor employees, whereas direct liability requires the government (or private litigants) to monitor employees' activities. This suggests an advantage of corporate liability. The corporation may be a better monitor of its employees' behavior than the government or private litigants because the corporation is closer to its employees and is probably already monitoring them to some extent (e.g., to ensure they are performing their primary tasks and for promotion purposes).

Third, corporate liability may address risk bearing costs better than executive liability. The reason is that executives are more risk averse than shareholders (or the corporation). The risk of liability can be diversified better by shareholders (who can invest in many corporations to spread their risk) rather than executives who cannot work at nearly as many corporations to spread their risk. The greater risk aversion of executives could reveal itself in attempts to over-comply with the law. Executives may exercise too much caution in approving projects, may take too long to make decisions and may become reluctant to take a position as an executive if liability risks are very large. Corporate liability helps to reduce this by placing liability on shareholders, who are better able to diversify risk and thereby are less risk averse.6

Although this provides a number of reasons for preferring corporate liability, it also implicitly suggests when direct liability on employees or executives would be a desirable supplement. For example, if the corporation itself might be judgment proof with respect to the harm caused then it will be under-deterred. The corporation would not have appropriate incentives to monitor its employees. In addition to this, if the maximum sanctions the corporation can impose on employees (e.g., denial of salary, future pay and pensions) are not sufficient to obtain the desired level of deterrence then the corporation's ability to effectively deter its employees is hampered. In these situations direct liability on executives may be desirable.7

For example, we might consider imposing additional civil monetary sanctions or criminal sanctions (e.g., prison) on executives. Generally, we prefer to first rely on civil sanctions before criminal ones. The reason for this is that criminal sanctions are more costly than civil sanctions (e.g., criminal sanctions have the costs of maintaining prisons). Consequently, if we can obtain the desired level of deterrence with the cheaper civil sanctions then we should prefer to rely on them first. For example, we might increase civil penalties on executives or deny them insurance coverage for certain wrongs.

However, sometimes even these higher civil sanctions on executives would not obtain the desired level of deterrence. For example, if the harm caused exceeds the combined assets of the corporation and the executive then one might need to consider imposing criminal sanctions. This seems most likely for activities that cause very high levels of harm (e.g., a sizeable risk of death or serious injury to a number of people from defective products). For such cases the assets of the corporation and executives may prove to be insufficient to pay for the harm and a case can be made for the imposition of criminal sanctions on executives.8

However, this presumes that the deterrence potential of civil sanctions on corporations and executives has been exhausted. Only after then would criminal sanctions be worth considering. However, assuming for the moment that the civil alternatives have been exhausted, and that there is still a need for greater deterrence, then we need to examine how criminal sanctions might be drafted and used.

3. Costs of Criminal Liability and Potential Ways to Reduce Those Costs9

Criminal sanctions are a powerful tool. However, this power comes at considerable cost. There are, of course, the direct costs of operating and maintaining prisons, but there are important indirect costs as well. One that merits particular attention is how the threat of criminal sanctions may lead executives to exercise too much caution and potentially lead to even more wrongdoing.

Because criminal sanctions are severe the risk of bearing them is likely to motivate risk averse executives to become too cautious and spend too much time and effort on monitoring employees. What is more troubling, however, is if executives become too scared to take on positions of importance at firms that face a greater risk of criminal liability (e.g., pharmaceutical firms whose products can cause the kinds of serious consequences being considered at this hearing). It is not difficult to imagine good people who would refuse to become executives of such firms due to the fear of facing criminal liability if things went wrong.10 If this happened then these positions would be taken by people who are perhaps not so careful and more tolerant of high risk activities. However, having more risk-tolerant people in charge of corporations that produce highly dangerous products does not seem like a way to reduce harm or wrongdoing. Indeed, it may lead to more harm or wrongdoing depending on the circumstances.

There may, however, be ways to address this concern in some measure. If we could avoid imposing liability on executives who attempted to do a good job or those who tried their best to prevent the harm, then the risk of scaring away good people would be reduced. One way to do this would be to premise criminal liability on a showing that the executive knew of the large risks associated with the product and its defects.

Although, in theory, this seems a desirable solution, in practice it may flounder. If there is legal error in applying the knowledge requirement (or perceived error in applying it or uncertainty amongst executives about what it means) then there will still be concerns about frightening away (or "chilling") good people from taking positions of importance in these firms. Thus, if one is considering criminal sanctions on executives it would seem critical to lay out in considerable detail, and without too much room for error, how and when executives would be held liable. A high mental state requirement might help in this direction with clear examples of criminal behavior provided either by legislation or enforcement agency policy.

If this can be achieved then these concerns are somewhat reduced. However, other concerns would be generated. For example, if we premise criminal sanctions on how much executives knew (or how much information they had) then some executives may find that knowing very little is a good way to avoid criminal liability. Indeed, if executives do not bother to gather information then they cannot be found to have "knowledge" and thereby they could avoid liability. If this were common practice then there is a potential for more harmful products to be introduced in the market. Of course, one could try to make "willful blindness" a potential mental state that would trigger liability, but this adds yet more uncertainty in the process (e.g., when is "willful blindness" met) and is likely to push in the direction of scaring executives away. Thus, it appears that however we draft the mental state requirement we will face a trade off between executives being unwilling to serve on corporate boards for fear of liability or allowing too many executives to escape liability by not gathering information about product risk.

There is no simple solution to this problem and perhaps the best one can hope for is to try to minimize the costs associated with scaring away executives and providing executives an escape route. However, I will attempt to provide a more hopeful outcome in the next few paragraphs.

One method of addressing the concern of executives' not gathering information is to make corporations strictly liable for the harm caused or make gathering information about product risk a matter that reduces or eliminates corporate liability. Under either alternative (strict liability or reducing liability), the corporation has an incentive to set up a system of gathering information about product risk. Under strict liability this is because corporations will bear the costs of harm and will want to reduce those costs by avoiding the sale of dangerous products. This means they will need a system in place to ferret out which products are indeed dangerous. Under the reducing or eliminating corporate sanctions alternative the firm would probably have a system for gathering information about product risk because its presence would help to reduce the liability the firm would face.11

Once such a system is set up it becomes more difficult for executives to avoid being aware of information or gathering information. As this information is gathered, distributed and used in making decisions it will become more difficult to avoid. Surely, some executives will find ways to avoid gathering information, but that number will be less than when corporations do not set up such a system.

Thus, these options, although not perfect, could address some concerns of imposing criminal sanctions on executives. 12 Nonetheless, the foregoing discussion highlights the kinds of trade offs that would need to be considered if we were to follow the path of increasing criminal sanctions on executives.

4. Other Concerns.

Before concluding, I want to offer a few thoughts on coordinating sanctions and the continuing importance of prosecutorial discretion. First, coordinating the sanctions imposed on executives and corporations in civil and criminal proceedings is something that needs to be explored in greater depth to ensure that we are obtaining the appropriate level of deterrence. Absent coordination we may in some cases over-deter and in other cases underdeter the relevant actors.13 I will not address methods for achieving coordination in the context of today's testimony, but will be happy to provide my thoughts on it should that be of interest.

Second, prosecutorial discretion remains important even with well drafted legislation. The prosecutor's decisions to charge, indict and pursue a prosecution are all important in determining the likely deterrent effect of the law. Greater guidance on how this discretion may be exercised would be helpful in assessing the likely benefits of a law imposing criminal sanctions on executives. For example, it may be helpful to have clear enforcement policy (or legislative) guidelines suggesting that criminal sanctions on executives should be pursued when alternative civil sanctions cannot pay for the harm caused and when proof of mental state is fairly clear.

5. Concluding Remarks

Corporate wrongdoing can lead to grievous harms. However, the kinds of liability regimes we put in place to deter or prevent such wrongdoing requires careful thought to avoid creating a situation where the cure is worse than the disease. In my testimony I have tried to highlight when criminal sanctions for executives may prove desirable. This is generally when the harm caused is so large that civil sanctions on corporations and executives prove to be insufficient in terms of deterring harmful behavior. If so then criminal sanctions on executives are worth considering. Defective products that cause serious injury or death to many people appear to be this kind of large harm, but before adopting criminal sanctions we would need to first exhaust the deterrent effects of civil sanctions.

If it appeared that criminal sanctions were indeed warranted, then considerable care is needed in drafting them to ensure that we do not scare away good people from becoming executives and also that we provide adequate incentives for corporations to gather information about product risk. Keeping these two features in mind one might be inclined to have a high mental state requirement (e.g., knowledge or more) with clear examples of what is meant by it, accompanied by either strict liability or some kind of liability reduction for corporations that have appropriate information gathering systems in place.

1 See V.S. Khanna, Corporate Criminal Liability: What Purpose Does It Serve? 109 HARV. L. REV. 1477 (1996) [hereinafter Corporate Criminal Liability]; V.S. Khanna, Is The Notion Of Corporate Fault A Faulty Notion?: The Case of Corporate Mens Rea, 79 B.U.L. REV. 355 (1999)[hereinafter Corporate Fault]; V.S. Khanna, Corporate Liability Standards: When Should Corporations Be Held Criminally Liable? 37 AM. CRIM. L. REV. 1239 (2001)[hereinafter When Corporations]; Vikramaditya S. Khanna, Should the Behavior of Top Management Matter?, 91 GEO. L.J. 1215 (2003)[hereinafter Top Management]; Vikramaditya S. Khanna, Politics and Corporate Crime Legislation, 27 REGULATION 30 (2004)[hereinafter Politics]; Vikramaditya S. Khanna, Corporate Defendants and the Protections of Criminal Procedure: An Economic Analysis, DISCUSSION PAPER NO. 2004-015, JOHN M. OLIN CENTER FOR LAW & ECONOMICS, UNIVERSITY OF MICHIGAN LAW SCHOOL, 2004[hereinafter Corporate Defendants]. I will be focusing on deterrence based arguments for liability rather than other types of arguments that may be used to support liability.

2 The discussion in this section relies on Khanna, Corporate Criminal Liability, supra note 1; Reinier H. Kraakman, Corporate Liability Strategies and the Costs of Legal Controls, 93 YALE L.J. 857 (1984); A. Mitchell Polinsky & Steven Shavell, Should Employees Be Subject to Fines and Imprisonment Given the Existence of Corporate Liability?, 13 INT'L REV. L. & ECON. 239 (1993).

3 I will use the terms employees and executives interchangeably for purposes of this testimony.

4 A numerical example would be as follows. Assume exercising care costs employees $20 in time and effort, the harm avoided is $100, and the employee's assets are only $5. In this scenario, the employee would be inclined not to spend $20 in time and effort when the most he or she can lose is $5. Society would want care exercised for $20 because it avoids $100 of harm.

5 If only employees are liable and employees are judgment proof then the firm can avoid including the full cost of the harm into the price of its products and too much of the product will be sold. See Steven Shavell, Strict Liability Versus Negligence, 9 J. LEGAL STUD. 1 (1980).

6 Corporations may also over-comply (e.g., when there are very large penalties under uncertain legal standards). See Richard Craswell & John E. Calfee, Deterrence and Uncertain Legal Standards, 2 J.L. ECON. & ORGANIZATION 279 (1986). But that point may come later for corporations than executives given that executives are generally thought to be more risk averse than corporations.

7 After all, corporations cannot send their employees to jail without state intervention and they cannot tap employees' savings that easily.

8 I do not discuss the potential for increasing criminal sanctions on corporations because I do not think that will generally be very helpful. See Khanna, Corporate Criminal Liability, supra note 1; Khanna, Politics, supra note 1.

9 The discussion in this section relies on Khanna, Corporate Fault, supra note 1; Khanna, When Corporations, supra note 1; Polinsky & Shavell, supra note 2; Jennifer Arlen, The Potentially Perverse Effects of Corporate Criminal Liability, 23 J. LEGAL STUD. 833 (1994); Jennifer Arlen & Reinier H. Kraakman, Controlling Corporate Misconduct: An Analysis of Corporate Liability Regimes, 72 N.Y.U. L. Rev. 687 (1997); Steven Shavell, Liability and the Incentive to Obtain-Information About Risk, 21 J. LEGAL STUD. 259 (1992).

10 This is especially true of executives who have excellent reputations. Those reputations would have taken time to build and can be seriously injured with one corporation's harmful activities.

11 I do not discuss whether strict liability or the reducing or eliminating corporate sanctions option is better in my testimony today, but it is a matter worthy of much fuller discussion.

12 Other methods of ameliorating these concerns could be considered (e.g., requiring executives to certify certain matters), but I do not discuss those here.

13 This point also relates more generally to examining the kinds of sanctions that seem appropriate in the context of corporate wrongdoing. For general discussion see Khanna Top Management, supra note 1; Vikramaditya Khanna & Timothy L. Dickinson, The Corporate Monitor: The New Corporate Czar?, 105 MICH. L. REV. 1713 (2007).