首页 | 农村土地改革 | 法经济学茶座 | 法经济学论坛
     网站首页≯ 刑法≯
法人的刑事责任

CORPORATE CRIMINAL LIABILITY
John R. Lott, Jr
John M. Olin Visiting Law and Economics Fellow
University of Chicago Law School
© Copyright 1999 John R. Lott, Jr

Abstract
This chapter reviews the economic literature on corporate crime. Economists
have discussed and answered a number of questions about corporate crime in
only few years time span. Perhaps this unusual success is due to the
questions involving corporations being so similar to other crime issues,
previously analysed. However, a number of questions remain largely
unexplored. Why do certain firms commit crimes while others do not? What
are the social costs imposed by the very high corporate criminal penalties?
How does the criminal justice process shape the organisation of firms?
JEL classification: K22, K14, K42
Keywords: Corporate Crime, Firm, US Sentencing Guidelines

1. Introduction

The economic analysis of corporate crime only started gaining momentum
during the last half dozen years. Of course, one notable exception to this rule
is Posner’s textbook on the Economic Analysis of Law (1992), though the
discussion of corporate criminality is relegated to issues such as whether
firms should obey the law or maximise profits. (It is also puzzling to report
how economists still fail in providing a conclusive definition of a firm. For
example, while I am sure that it is not done intentionally, definitions of
organised crime appear to be interchangeable with definition of firms (see,
for example, Dick, 1995)).

The catalyst for this was the debate over and the subsequent adoption in
1991 of the US government’s corporate sentencing guidelines. The research
itself has largely focused on theoretical questions raised by the guidelines
(for example, what should be included in calculating the social cost of
crimes committed by corporations or whether it should be the firm and/or its
managers who are penalised), descriptive empirical evidence on how
corporate penalties have changed over time, and some evidence indicating
what factors predispose firms to violate the law.

2. The Theory
The basic principle behind the economics view of legally imposed sanctions
is to ensure that those who break the law internalise the externalities that
they impose upon others. Several questions arise in this context: (1) how
large is the externality, (2) upon whom should the penalty be imposed, and
(3) should the penalty take the form of criminal or civil sanctions or possibly
just through the loss of reputation.

While the issue of criminal harm is the same for crimes by firms and
individuals (see for example Becker, 1968 or Friedman, 1981), the debate
among economists over what should be included in corporate penalties
largely arises from the sentencing commission’s guidelines, which explicitly
include some notion of the third-party spillover effects from crimes like
financial or consumer fraud (United States Sentencing Commission, Chapter
Eight - Sentencing of Organisations, U.S.S.C. § 8 C5.5, November 1, 1991).
In part this interest arises because fraud cases involve almost half of all
corporate crimes (Alexander and Cohen, 1996a, p. 427). Upward departures
from the guidelines are also required when any corporate crime involves a
‘substantial risk to the integrity or continued existence of a market’ (United
States Sentencing Commission, Chapter Eight - Sentencing of
Organisations, U.S.S.C. Proposed Addition to the Sentencing Guidelines, §
8.12, July, 1988).

Some economists, such as Tom Ulen (1996), defend the theory behind
the guidelines by arguing that criminal penalties are required to ensure that
offending firms internalize the losses imposed on buyers because frauds
cause both consumers and firms to take costly defensive actions. In this
discussion, fraud affects third parties like a tax, raising both the costs of
production by firms and lowering the price that consumers are willing to
pay. While Ulen accepts that there is an optimal amount of fraud, criminal
penalties should account for these third party losses whenever there is more
than the ‘socially optimal amount of fraud taking place’ (p. 358). How one
knows when these frauds are greater than the optimal number is not clear,
though most of this discussion identifies dead-weight losses whenever frauds
effect third-party behaviour. The question of how large the criminal
penalties should be are then a function of those parameters that determine
the dead-weight loss from taxes (for example, the elasticity of demand).

An alternative position argues that third parties should normally be seen
as benefiting from the frauds committed against others (Daniel and Lott,
1995; Lott, 1996). For example, if after a fraud other firms’ customers desire
that they acquire more reputation because the customers learn that the
existing level of reputation produced less quality assurance than they had
previously believed, the revelation of the fraud allows them to better equate
the marginal benefits and costs from buying quality assurance. The
distinction drawn here is whether the fraud changes the underlying
probability with which other frauds will occur or whether it merely provides
other firms’ customers with better information to evaluate their own
situation. Given that few frauds are likely to innovate new methods of
perpetrating the deception that will then be copied by imitators, few frauds
will alter the underlying probability of other frauds occurring.

An interesting debate concerns whether it might be more efficient to
impose penalties on the principals instead of simply holding agents directly
liable for their actions.

Kraakman (1984b) suggests that part of the reason may have to do with
monitoring costs. For example, corporate liability may be efficient if the
government is better at monitoring corporations and these corporations are
good at monitoring agents. Salzburg (1991) and Polinsky and Shavell (1993)
point out that the penalties imposed upon the firm may not always be large
enough to produce the efficient level of deterrence, thus justifying an
additional penalty being imposed upon the agents (while Salzburg merely
provides a verbal argument on this issue, Polinsky and Shavell provide the
formal model behind this reasoning). Salzburg (1991) also advocates
prosecuting corporations because it reduces the prosecutor’s burden of
investigation and conviction and avoids various preceived procedural
problems, though in the end this argument amounts to little more than a
claim that both firms and individuals should be prosecuted because it is
possible to do so.

Arlen (1994) suggests that penalties on principles may have the
unintended effect of inducing principals to expend less effort monitoring,
since their own monitoring of principals may increase the likelihood of
detection. Unless corporations are compensated for its monitoring efforts
through penalty reductions that offset the higher probability of prosecutions
that result from the internal monitoring, the private return to monitoring
will fall below the social return. Davis (1996) raises the issue of how the
possibility that agents’ actions can create liability for a firm’s principals thus
alter agents’ compensation. He asks whether agents actually prefer
government rules that increase their liability. His conclusion is that there are
reasonable assumptions about the agent’s risk neutrality and how the
earning depend on legal and illegal profits, where agents will prefer legal
liability.

However, whatever the answer to these questions, the central question
still remains: should corporate criminality even exist. While there are
philosophical arguments over whether a corporation as compared to
individuals can have intent (Parker, 1996), economists have largely debated
whether criminal sanctions are necessary to ensure that corporations
internalize the externalities that their illegal behaviour creates. Block
(1991), Karpoff and Lott (1993), Fischel and Sykes (1996) and Parker
 (1996), ask whether civil penalties might be preferred. Not only do criminal
proceedings require higher levels of proof than civil proceedings but, for
firms, the ultimate penalty is the same in either case: fines. Civil prosecution
provides other benefits ranging from compensation of victims, law
enforcement efficiencies, and the ability of apparently obtaining the same
stigma produced by criminal prosecution though at a lower cost (Parker,
1996, pp. 387-389).

One question still persists: might the government be better suited to
imposing particularly large penalties on individuals because the government
can use imprisonment while private firms are limited to monetary damages?
While this is true, it misses the central point. Criminal charges can still be
pursued against individual agents even if the case against the firm is
handled civilly (Fischel and Sykes, 1996).

3. Empirical Evidence

Empirical corporate crime studies take several forms. Either they measure
the size of the criminal penalties and how the US Sentencing Guidelines
have altered these penalties, test which characteristics are exhibited by firms
engaging in crime, measure firm profitability from crime, or examine
whether the US Sentencing Guidelines have moved penalties closer towards
efficiency.

Penalties prior to the guidelines appear to have exhibited many patterns
consistent with optimal penalty theory (Cohen, 1996). Not only did firms
creating greater harm face greater sanctions, but individual liability was also
greater when the organisation was unable to compensate for the harm done.
Interestingly, for the crimes with the largest losses, the penalty that the
courts imposed suggested a belief that the probability of detection was higher
for those crimes (Cohen, 1991 and Lott, 1991, p. 440). An easy optimal
penalty explanation appears to exist for this. For example, large oil spills are
virtually certain to be detected, but tiny ones (for example, the discharging
of ballast) are most likely to be unnoticed.

The sentencing guidelines turned much of this pre-existing structure on
its head, with the multiple now instead increasing with the size of the harm.
The guidelines also have dramatically increased criminal penalties. Early
estimates indicated increases of 20, 30 or 40 times over previous penalties,
and changes as high as 50 to 100 fold for a few select crimes (Cohen, 1991).
Others produced estimates of at least twenty-fold (Block, 1991 and Lott,
1991).

With fraud being the largest category of corporate crime and one of the
areas with the biggest increases in penalties under the guidelines, much of
the recent empirical evidence has focused on it. Analysing the Sentencing
Commission’s concerns that the penalties for fraud have been too low,
economists sought to investigate whether traditional emphasis on legally
imposed penalties underestimated the true penalty. To evaluate this, Karpoff
and Lott (1993) measured the extent of reputational penalties for fraudulent
firms prior to the 1991 corporate guidelines. An allegation that a corporation
had defrauded stakeholders or the government corresponds to an
economically and statistically significant loss in the market value of the
common stock. On average, very little of this loss, in the order of 6.5
percent, can be attributed to the accused firm’s expected legal fees and
penalties. It is possible to compute larger portions of the loss that could
reflect higher expected penalties for future frauds and the lost value of the
cheating firm’s profits from committing fraud. But even under extremely
unrealistic assumptions, one third of the loss remains. (Yet, firms are not the
only entities to suffer from reduced earnings when they commit fraud. White
collar criminal face large reductions in legitimate earnings when they are
convicted of fraud (Lott, 1992). The average criminal convicted of fraud
faces almost a 40 percent drop in legitimate earnings when he returns to the
workforce. Similar drops are also observed for embezzlement, larceny and
theft.)

Large reputational penalties are associated with frauds committed
against stakeholders, government agencies, and investors. In contrast, the
reputational loss for frauds involving regulatory violations, in which firms
have not violated an implicit or explicit contract with a stakeholder or
investor, is negligible (see also Block, 1991). The evidence is consistent with
the market penalty for fraud being systematically related to the cost imposed
on parties with whom the firm does business. Because of their large size,
reputational effects play an important role in disciplining firms that commit
fraud. There appear to be very similar reputational penalties when either
criminal or civil cases are filed (Block, 1991).

For fraud, the average decrease in common stock value exceeds the US
Sentencing Commission’s estimates of victim losses by more than 100 times.
The Sentencing Commission reports that Department of Justice prosecutors
believe the probability of convicting a corporate fraud is between 20 and 57
percent, implying an optimal ratio of actual penalty to social costs of
between 1.75 and 5 (Lott, 1996, pp. 374-375). Yet, as mentioned above, the
expected total penalty far exceeds the Commission’s estimated loss borne by
the victims, and this holds even at 1980’s levels of criminal fines. Thus,
unless these numbers are off by a very large magnitude, the past efforts to
substantially increase criminal penalties for corporate fraud were completely
misguided.

Some evidence indicates that managers conceal bad news from the
market (Arlen and Carney, 1992), indirectly supporting a ‘last period
hypothesis’, in which misconduct results from poor performance on the part
of the firm. Yet, other findings suggest no actual or expected deterioration in
firm profitability prior to a fraud (Karpoff and Lott, 1993). A study by
Alexander and Cohen (1996a) examines other aspects of firms’ performance
prior to a fraud, as measured by the firm’s assets, employment and sales.
With the exception of environmental crimes, they find no relationship
between these measures and whether a fraud is going to occurs.

Antitrust penalties have been studied extensively by economists, and up
until recently, when it was displaced by environmental crimes, it involved
the second largest number of cases. Asch and Seneca (1975, 1976) provided
the earliest evidence in this area by investigating whether the prior
performance of firms charged with antitrust violations differed relative to
their Fortune 500 counterparts. Surprisingly, Asch and Seneca found a
significant negative relationship between collusion and several alternative
measures of performance such as growth in sales and return on assets over
the five years before the date of the relevant plea or court finding. Staw and
Szwajkowski (1975) attempt to explain this result as lower profitability
leading to collusion. (However, since these studies rely on actual
government suits, there is always the nagging question of whether the
government can accurately tell which firms if any are engaging in antitrust
violations.)

Game theory has also suggested essentially limitless explanations for
firm behavior. A number of recent game-theoretic models suggest that with
asymmetric information it can be profitable for some firms to acquire a
reputation for toughness in order to discourage later entry. One may view
these models as providing a method of understanding what factors increase
the likelihood that firms commit crimes, though initial empirical evidence
provides little support that these models will be particularly useful. Lott and
Opler (1996) identify institutional arrangements that firms must undertake
to ensure the credibility of any predatory commitments. Simply hiring
managers who value market share or output maximisation is not sufficient if
the manager can be removed whenever it actually becomes necessary to
engage in predation. It is also required that the firms make removing the
manager difficult. In addition, the incumbent manager should be rewarded
for increasing output as opposed to increasing short-term profits. A study of
firms charged with predation failed to yield any support that allegedly
predatory firms are organised as these game-theoretic models imply. If
anything, the reverse seems to be frequently true.

Alleged predators more often are large firms with managerial
compensation schemes relying heavily on short-term profits (Lott and Opler,
1996, pp. 362-363), with this last fact hardly seeming consistent with any
known notion of predation. Just as Bork (1978, pp. 347-364) argues, the
general picture is consistent with the laws being used to harass large
profitable firms.

4. Conclusion

Economists have discussed and answered a number of questions about
corporate crime in only few years time span. Perhaps this unusual success is
due to the questions involving corporations being so similar to other crime
issues, previously analysed. However, a number of questions remain largely
unexplored. Why do certain firms commit crimes while others do not? What
are the social costs imposed by the very high corporate criminal penalties?
How does the criminal justice process shape the organisation of firms? These
are areas that should lend themselves to very interesting future research.

 


中国法律经济学网登载此文出于学术研究之目的,绝不意味着中国法律经济学网赞同其观点或证实其描述。以上内容仅供研究者学习与交流,无意侵犯版权。如有侵犯您的利益,请告知。我们将尽快删除。

加入日期:2007/1/11 9:16:08浏览次数:2384
发表评论
名号:
内容:
验证: 5208
法律经济学网
联系站长: 柯华庆 lawgame@263.net 京ICP备09028584号
北京市昌平区中国政法大学法学院(102249)
本网站由卡卡鱼网提供技术支持 网站总访问量:1996356