University of Chicago Law School

Chicago Unbound

Journal Articles Faculty Scholarship

1990

Remedies and Incentives in Private and Public Law: A

Comparative Essay
Saul Levmore

William J. Stuntz

Follow this and additional works at: https://chicagounbound.uchicago.edu/journal_articles

Part of the Law Commons

Recommended Citation
Saul Levmore & William J. Stuntz, "Remedies and Incentives in Private and Public Law: A Comparative
Essay," 1990 Wisconsin Law Review 483 (1990).

This Article is brought to you for free and open access by the Faculty Scholarship at Chicago Unbound. It has been
accepted for inclusion in Journal Articles by an authorized administrator of Chicago Unbound. For more
information, please contact [email protected].
 ESSAYS

REMEDIES AND INCENTIVES IN PRIVATE AND

PUBLIC LAW: A COMPARATIVE ESSAY*

SAUL LEvMOREt & WILLIAM J. STUNT#

When civil and common lawyers observe one another's legal sys-
tems from a law-and-economics perspective, they quickly learn the
value of focusing on the properties of legal rules that distinguish various
systems. Thus, Steven Shavell's exploration of the economics of fee-
shifting rules, William Bishop's investigation of the different rules gov-
erning breach of contract, and Mark Ramseyer's inquiry into Japanese
and American water law, all support the conclusion that when com-
peting legal rules survive in different countries it is quite likely that
both rules may be fairly efficient.' Differences between civil and com-
mon law systems are thus interesting (from an economic perspective)
because the differences draw attention to the positive and negative
incentive effects, or other costs and benefits, of inherited rules. In this
Essay, we emphasize such a comparative law-and-economics approach
by drawing on seemingly unrelated areas of law, where civil and com-
mon law rules are generally (but not always) dissimilar, and by arguing
that fundamental and universal incentive problems go a long way
toward explaining both the similarities and the differences among the
various rules. In particular, we choose one theme: the choice between
damages and disgorgement-that is, the question of whether to control
behavior with a rule requiring an antisocial party to pay over damages
or with a rule calling for the restitution of unjust enrichment-and we
show that this choice permeates the law of torts and contracts, the law
governing unreasonable searches for evidence, and the law of govern-
mental takings of private property. With this unifying theme, we try
* This paper was delivered at the 1990 Conference on the Economics of the Civil
Law in Hamburg, Germany.
t Barron F. Black Research Professor of Law, University of Virginia School of Law.
B.A., Columbia, 1973; Ph.D., Yale, 1978; J.D., Yale, 1980.
t Professor of Law, University of Virginia School of Law. B.A., William and Mary,
1980; J.D. Univ. of Virginia, 1984.
1. Shavell, Suit, Settlement, and Trial: A TheoreticalAnalysis under Alternative
Methods for the Allocation ofLegal Costs, I I J. LEGAL STUD. 55 (1982); Bishop, The Choice
of Remedy for Breach of Contract, 14 J. LEGAL STUD. 299 (1985); Ramseyer, Water Law in
Imperial Japan: Public Goods, Private Claims, and Legal Convergence, 18 J. LEGAL STUD.
51 (1989); See also Levmore, Rethinking ComparativeLaw Variety and Uniformity in Ancient
and Modern Tort Law, 61 TUL. L. REV. 235 (1986).

HeinOnline -- 1990 Wis. L. Rev. 483 1990

 WISCONSIN LAW REVIEW

both to provide the basis for positive theories of these areas of law and
to demonstrate the value of comparing legal rules within an incentive-
oriented framework.

2
I. DAMAGES VERSUS DISGORGEMENT IN TORT LAW

A factory owner who should take a precaution costing 100 in order

to avoid causing 101 of damage to other parties may be induced by
the threat of tort damages to take the precaution, because a net savings
is available. If a restitution, or disgorgement, remedy is to offer a similar
incentive, then the law could call for the disgorgement of unjust en-
richment (of 100 in this case) plus some small amount in order to ensure
that the potential wrongdoer is not indifferent between behaving well
and behaving antisocially. Why, then, do we use tort damages rather
than disgorgement to control tortious behavior? It is probably not be-
cause of any savings in administrative costs, for the costs of running
a restitution system are likely to be comparable to those of a tort system.
Where tort liability is negligence-based, for example, it is necessary to
decide whether expected costs of accidents require that certain precau-
tions be taken; but this threshold determination, whether easy or dif-
ficult, is needed as much in restitution as in tort, because the restitution
remedy is triggered by a finding that the defendant's enrichment was
unjust. The critical remaining comparison, therefore, is between meas-
uring damages-the task in torts-and measuring the costs of untaken
precautions-the means of restitution. 3 Both can be difficult to assess.
In medical malpractice cases, for example, damages are notoriously
difficult to quantify, but it would also be difficult to answer the question
of how much a physician gained by allowing a particular injury to occur.
The choice between tort and restitution remedies is, instead, best
understood through deterrence considerations. 4 Imagine that 7's failure
2. The analysis in Part I of this Essay is drawn from Levmore, ProbabilisticRe-
covery, Restitution, and Recurring Wrongs, 19 J. LEGAL STUD. - (forthcoming 1990).
In this Essay we often use the terms disgorgement and restitution interchangeably. The
traditional connotation of restitution is that A returns to B that which A unjustly received
as a result of B's efforts or misfortune. And inasmuch as some of the circumstances discussed
in this Essay concern situations where B has not caused A's enrichment, some commentators
and readers will prefer the label of disgorgement. See DeLong, The Efficiency of a Disgorge-
ment as a Remedy for Breach of Contract, 22 IND. L. REV. 737, 743 n. 19 (1989); Farnsworth,
Your Loss or My Gain? The Dilemma of the Disgorgement Principle in Breach of Contract,
94 YALE L.J. 1339, 1342 (1985); Levmore, ExplainingRestitution, 71 VA. L. REV. 65 (1985).
3. When liability is strict, the comparison is somewhat different, but it leads to a
conclusion much like the one reached in the text's discussion of restitution and fault-based
liability.
4. A restitution-based tort system must also be sensitve to moral hazard and other
incentive problems. Thus, when 7's failure to take a $50 precaution causes $8 in harm to
each of 10 people, it is apparent that each victim can not be allowed to recover 7's full gain
of $50 both because there will be overdeterrence and because of the moral hazard that it will

HeinOnline -- 1990 Wis. L. Rev. 484 1990

1990:483 Remedies and Incentives in Privateand PublicLaw 485

to take a precaution which costs fifty leads to one of three outcomes:

a 50% chance that each of ten people will suffer harm of ten; a 25%
chance that each of five people will suffer harm of four; and a 25%
chance that no injuries will occur at all. A simple restitution rule, which
allows plaintiffs to collect an antisocial defendant's savings from not
taking needed precautions, will obviously underdeter T because some
persons will be injured with only a 75% probability. Plaintiffs will ma-
terialize with a probability of less than one-and then they will collect
nothing more than the wrongdoer, I's, enrichment. If Ts maximum
liability is fifty-the amount of Ts unjust enrichment-then Twill pre-
fer an expected restitution liability of 37.50 (a 0.75 chance of paying
fifty) to spending fifty on the actual precaution-even though the pre-
caution would save fifty-five in expected injuries.
Put differently, legal rules must control the behavior of wrongdoers
who sometimes create risks but not victims. Tort law deters such be-
havior by requiring wrongdoers to pay full damages-which often
greatly exceed the cost of precaution taking-when injuries do occur
and victims come forward. Restitution law, however, extracts only the
wrongdoer's enrichment from not taking a precaution when there is a
victim, and thus underdeters wrongdoers who, essentially, can count
on disgorging only when there are victims, but on pocketing money
otherwise.
The disgorgement remedy could be salvaged by requiring wrong-
doers to pay a multiple of their present undeserved gains in order to
make up, in deterrence terms, for those occasions where no one knew
of, or had standing to complain about, their wrongful behavior and
unjust enrichment. But these "multipliers" would need to be industry-
or even actor-specific in order both to achieve the right level of deter-
rence and to avoid serious moral hazards. Thus, the hypothetical
wrongdoer described in the previous paragraph requires us to multiply
plaintiffs' claims by 1.33, to make up for the 25% chance that no victims
(plaintiffs) will be forthcoming, in order to collect the full amount of
unjust enrichment and, in turn, not to leave T with an incentive to
behave negligently. But a similar defendant whose negligence was 50%
likely to cause 110 in harm and 50% likely to cause no harm must face
a multiplier of two, or he will profit from his wrongdoing. The need
to calculate these multipliers is itself a good reason to prefer tort over
restitution.
Multipliers will also create moral hazards. If, for instance, I's
negligence in not spending fifty is 50% likely to cause eighty in harm,

be quite profitable to be a victim under such a rule. The problem might be solved by letting
each plaintiff recover only one-tenth of the whole enrichment, although it may be necessary
to give an extra reward to the first plaintiff who identifies the unjustly enriched defendant.

HeinOnline -- 1990 Wis. L. Rev. 485 1990

 WISCONSIN LAW REVIEW

25% likely to cause sixty in harm, and 25% likely to cause no harm,
the required multiplier of 1.33 will, once again, either make it profitable
to be a victim (because 1.33 times fifty is more than sixty) or require
the expenditure of administrative costs in order to tax away part of
the recovery (perhaps to finance compensation for other plaintiffs in-
jured at some future point by the very same defendant or to fund
plaintiffs whose own damages exceed the unjust enrichment of their
defendants). 5 In sum, a restitution system could be substituted for the
tort system but, in order to reproduce the deterrent qualities of the tort
system and to guard against moral hazards, a variety of corrective
devices would be necessary. That a restitution system could only take
the place of the tort system either by sacrificing a good deal of efficient
deterrence or by incurring heavy administrative costs should not dis-
tract us from the fact that restitution might be useful in other areas of
law, where multipliers would be unnecessary or where the damage rem-
edy itself has disadvantages. And, as a comparative matter, we might
expect to find (as we do) that no legal system relies on the restitution
remedy to control tortious behavior, but that disgorgement is used in
other areas of law where the relative advantages of the two remedial
tools are different.

II. THE COEXISTENCE OF DAMAGES AND DISGORGEMENT AS

REMEDIES FOR BREACH OF CONTRACT

Under American law (which favors money damages rather than

specific performance in the event of a breach of contract) S, who prom-
ises a widget to B1 for a price of 100, and then sells it to B2, who
appears and is willing to spend 120, had better take into account that
she will owe B1 damages. The rule is quite consistent with the notion
of "efficient breach" because S is encouraged to breach when B2 will
pay more but not less than what B1 agreed to pay. The real question,
of course, is whether it would not be better still to use a specific per-
formance rule, so that S is forced to deliver to BJ-but then B2 can
find B1 and purchase the widget at a price profitable to B.
It is common in the United States for law-and-economics scholars
to say that damages are preferable because B2 can transact with S (who
5. Similarly, under a more radical version of the multiplier strategy, the first victim
to come forward (or any private attorney general) might be awarded damages equal to the
wrongdoer's total enrichment attributable to the negligently imposed risks on similarly sit-
uated parties. But this scheme generates an obvious moral hazard, takes us away from the
restitutionary ideal that the defendant's gain was tied to the plaintiffs loss, and introduces
large multipliers of the sort that the common and civil law have largely avoided. Perhaps
most importantly, it introduces the serious hazard that an insider (employee of defendant)
will collude with a potential victim under a scheme in which the employee behaves negligently
and the "victim" rushes forward to collect a large amount of money.

HeinOnline -- 1990 Wis. L. Rev. 486 1990

1990:483 Remedies and Incentives in Private and PublicLaw 487

often advertises as a seller of widgets) more easily than B2 can locate

and negotiate with B1. Professor Bishop has also compared what hap-
pens if S owes B1 some damages and sells to B2 with the situation
where there is a specific performance rule in place and S contacts B1,
acknowledges the backdrop of the performance rule, but negotiates out
of the first contract.6 In this situation, S and B2 must agree on a price
and S and B1 need to agree on a settlement which, in turn, will reflect
both BI's damages (which S must learn about) and S's profit on the
resale to B2 (which B1 will have reason to learn about). In contrast,
under the damages rule, S and B2 must again agree on a price, but now
B1 has no need to learn about S's profit on the sale to B2. In short,
specific performance involves the cost of extra information and extra
negotiation. Putting the latter cost differently, B1 and S, in a classic
bilateral monopoly, may sometimes miss a bargain, whereas under the
damage rule no bargain is needed between B1 and S, for S simply deals
with B2 and then worries about liability to B1.
We might modify Bishop's superb analysis with the point, in de-
fense of specific performance, that there is the alternative of S's per-
forming, followed by a sale from BI to B2. The cost of this alternative
is the extra transaction, or telephone call if you will (because it is
generally easy for B2 to find S but not B1), but there is no need to
compute Bl's damages.
In sum, one might rationalize the American rule by saying that
damages are preferred because with specific performance there is either
an extra transaction or both extra information and negotiation costs.
It is also noteworthy that with specific performance, contract prices
will be a little lower than with damages, but it is hard to see why buyers
should care; they either pay less and get some damages, or they pay
more and get extra profit (above and beyond damages) from B2.
An unambitious way to explain the use of damages for breach of
contract is to note that specific performance is in the restitution family,
because it is the tool that forces S to disgorge some of the gains enjoyed
by breaching with BI and dealing with B2, and to reason that a legal
system is likely to be doctrinally consistent. If one rule is to be used
consistently in both tort and contract law, it is easy to see that it would
be the rule of damages, because while either damages or restitution
could be the tool to control inefficient breaches of contract, the damage
rule appears significantly superior in tort.
As a comparative matter, however, it is plain that there are many

6. See Bishop, supra note 1; DeLong, supra note 2; Schwartz, The Casefor Specific
Performance, 89 YALE L.J. 271 (1979).
7. Note that there will not be too many surprises with regard to Bl's damages
because of the rule which provides that in contract law damages are normally limited to
foreseeable damages. That rule, of course, helps breachers allocate their mitigation efforts.

HeinOnline -- 1990 Wis. L. Rev. 487 1990

 WISCONSIN LAW REVIEW

legal systems which use specific performance, or a relative of restitution,

in contract, and yet these systems employ the rule of damages rather
than restitution in tort. It is therefore sensible to recognize the link
between restitution and specific performance, but to think of the debate
between damages and specific performance as reflecting a real difference
in judgment as to which rule is preferable in contract law. If one were
inclined to accept the analysis above, which points to the transaction
costs superiority of the damages rule, then the availability of specific
performance in many civil law systems might be understood as reflect-
ing the observation that the superiority of the damages rule is at best
a very close call, because specific performance surely economizes on
court costs. It calls for some supervision, but a damage rule requires
a real determination of money damages. 8 Indeed, in comparative terms,
the most striking thing about this area of law is that in many legal
systems the disappointed promisee can often choose between rem-
edies 9-and this seems quite sensible because such a promisee, like BI,
often knows best the costs of ascertaining her own damages.1"
If the discussion up to this point seems to imply that chaos reigns,
and that theory is useless to lawyers, then it is especially useful to
introduce another theme: the presence or absence of a kind of symmetry
in legal rules. Consider a case in which S contracts to sell goods to B,
S delivers, and B fails to pay at the time specified in the original agree-
ment. Can S repossess the goods? As far as American law is concerned,
there are two doctrinal paths to the conclusion that S can not do so.

8. Note that the use of specific performance in United States law for such unique
goods as land is easily explicable because it is especially easy for B2 to find B), by searching
local land records or local realtors.
E. Allan Farnsworth notes that in planned economies without well-developed markets
for substitute goods, the preference is for specific performance, apparently on valuation
grounds. Farnsworth, Damagesand Specific Relief, 27 AM. J. COMP. L. 247 (1979).
9. G. TREITEL, REMEDIES FOR BREACH OF CONTRACT: A COMPARATIVE ACCOUNT
47-51 (1988).
10. The choice between damages and specific performance is even subtler, or closer,
than implied in the text. It is arguable, after all, that the fact that B2 pays more to S than
B) agreed to pay for a good or service does not prove that it is worth more to B2, because
B1 may simply be a superior bargainer. B) may not have revealed her true preferences to S,
despite the fact that revelation would now lead to greater damages because of the foreseeability
rule, because S would then have held out for a higher price. And the foreseeability rule would
not be satisfied by BI informing S of her high potential damages only after the contract is
made (because S then loses the chance to charge an "insurance" premium).
In response to this argument about the ability of the damage rule to support efficient
breaches, it might be said that at least on average there is reason to think that B2 values the
item more than does BI because B) may in fact have been the worse bargainer. Without
further information, there is no reason to think that one is systematically a poorer bargainer
than the other. B) can always go to B2 and start all over again-or, when hearing from S
about the pending breachs, she can induce S to breach with B2 and "resell" to B).

HeinOnline -- 1990 Wis. L. Rev. 488 1990

1990:483 Remedies and Incentives in Privateand Public Law 489

First, such retrieval would interfere with the relatively careful rules
regarding security interests in personal property, under article 9 of the
Uniform Commercial Code, because other creditors have no means of
learning about the contract between S and B. There is a filing system
and a serious order of priorities among creditors, and S is neither
recorded in that system nor otherwise especially deserving. Second,
and more plainly, section 2-702(2) of the Uniform Commercial Code
gives S rights if S discovers that B received the goods (on credit) while
insolvent, and S demands the goods back on these grounds within ten
days of the delivery to B. (S does better if B has specifically lied about
his solvency to this seller within three months before the delivery.)
Since no other section of the commercial code seems relevant, the
implication is that S otherwise has no repossession right, and S is left
to sue for damages.
A more thematic (and perilous) way to reach the same result is to
reason that this is an area of law where symmetry will be found, because
whatever reason a legal system has for preferring damages over specific
performance when S breaches probably carries over to the case where
B breaches. Forcing B to return the goods, or allowing S to repossess,
is, after all, the closest thing to specific performance. And allowing S
to repossess would certainly amount to a restitution rule, because B
will be unjustly enriched by keeping the goods. The symmetry argument
is thus another way of saying that the preference for damages over
restitution might continue unless there are affirmative reasons to prefer
restitution. This is not to say that such symmetry or consistency is
required as a matter of efficiency theory. The original argument, after
all, was that sellers are more easily located by third parties (like B2
above) than are buyers-and this point argues in favor of damages when
S breaches (to sell to B2)-but this says nothing about the case where
B breaches. " But because it would appear that B2 can indeed find B1
quite readily once the goods are already in Bl's possession, the argu-
ment for continuing the logic of the damages regime, so that there are
symmetrical rules, is quite strong. And as a comparative matter, the
remarkable thing is that in Japan, for instance, where B1 has a choice
of remedies when S breaches in B2's favor, it is the case that when BI
breaches, S can choose to retrieve. 12 There is, in short, a kind of sym-
11. Where breach occurs because of a drop in value of the item in question, it is
likely that the seller is in a better position to resell. But in a case where B has simply no
longer a need for the item or, most generally, where B has run out of funds, so that the
damages rule will run into B's insolvency and the seller very much cares whether a specific
performance or a damages rule is in place, S's comparative advantage is less relevant.
12. MrNPO art. 545 (Japan). In France, rescission is also an option. See F. LAWSON,
A. ANTON, & L. BROWN, AMOS AND WALTON'S INTRODUCTION TO FRENCH LAW 197-99

HeinOnline -- 1990 Wis. L. Rev. 489 1990

 WISCONSIN LAW REVIEW

metry, or commitment in favor of or against the restitution remedy,

in these areas of law.

III. WRONGFULLY OBTAINED EVIDENCE

In the United States, as elsewhere, the most commonly imposed

remedy for a wrongful police search or seizure is exclusion of the il-
legally seized evidence from the suspect's criminal trial. This remedy
can be seen in terms of the restitution, or disgorgement, concept, since
it does not force the police officer or the government to pay for the
harm caused by the illegal conduct, but instead requires the government
to give up its gains-the evidence that was obtained because of the
police misconduct. As with contract law and quite unlike tort law, the
advantages of the disgorgement rule in this context are primarily ad-
ministrative.
There are two such administrative or practical advantages of ex-
clusion, or disgorgement, rather than damages: ease of valuation and
procedural convenience. And each of these advantages is imposing.
The harm caused by a typical illegal search is largely nonmonetary.
Property damage aside, the injury consists of the victim's humiliation
and loss of privacy, and the more diffuse harm to society's sense of
security. Harms such as these cannot be priced by the legal system with
any accuracy. Yet, accurate pricing is essential to a well-functioning
damages system. The actors (police officers) receive no tangible reward
for the marginal legal search or arrest, and are usually free to avoid
acting altogether-that is, to avoid performing the search or making the
arrest-without suffering substantial sanctions. Under these circum-
stances, if damages are imposed for illegal action (and if, as is probably
the case in all legal systems, the standards that determine what is legal
are somewhat vague), there is the serious danger that society will not
only get fewer illegal searches and seizures, but will also get many fewer
legal ones. 13 This activity-level effect is exacerbated if the relevant dam-
ages are overestimated-and the risk of such overestimates is likely to
be high, given that the harms in question are both socially sensitive
and irreducibly subjective.
Disgorgement-that is, exclusion of illegally obtained evidence-
largely avoids this measurement problem. No valuation is necessary
in order to exclude evidence; the only determination required is

(1963). See generally Cooper, The Reclamation Rights of Unpaid and Unsecured Sellers in
International Trade, 1987 COLUM. Bus. L. REv. 17. (reporting limited rights of reclamation,
with no unifying theory but making no attempt to link the right of reclamation to the remedy
available when the seller breaches).
13. See P. SCHUCK, SuING GOVERNMENT: CITIZEN REMEDIES FOR OFFICIAL
WRONGS 59-81 (1983); Mashaw, Civil Liability of Government Officers: Property Rights and
Official Accountability, 42 LAw & CONTEMP. PROBS. 8, 29-33 (1978).

HeinOnline -- 1990 Wis. L. Rev. 490 1990

1990:483 Remedies and Incentives in Privateand Public Law 491

whether the police misconduct caused the discovery of the evidence in

question. Any difficulties encountered in determining causation are
unlikely to be different from those confronted in a regime based on the
remedy of damages, where it will be difficult to decide whether to count
such harms as the discovery of a crime (through an illegal search). In
short, problems of causation are unlikely to influence the choice be-
tween the remedies of disgorgement and damages in this area.
The second major advantage of a disgorgement, or exclusionary,
rule is its potential to economize on court costs. The remedy of damages
requires lawsuits as the means of enforcement, and the costs associated
with litigation are well known. In contrast, exclusionary rules do not
generate additional work for the court system, because these rules come
into play in criminal cases that the state would wish to prosecute re-
gardless of the existence of the exclusionary rule itself. The adminis-
tration of an exclusionary rule requires only a brief hearing within the
framework of an existing criminal case; the damage remedy requires a
full-blown civil suit.
It would therefore appear that, as a theoretical matter, exclusionary
rules are superior to damage rules in enforcing restraints on police
investigation of crime. And this theoretical conclusion seems to trans-
late into a successful positive theory because it is consistent with the
law of several countries. A number of countries have some form of an
exclusionary rule, 4 and in at least one country-the United States-it
is the primary tool for enforcing search and seizure law.'" In contrast,
we know of no legal system that uses damages as the primary deterrent
mechanism in this area. The real choice is not between disgorgement
(exclusion of evidence) and damages, but between exclusion and po-
litical (or cultural or other nonlegal) constraints, a distinction we return
to below.
The advantages of the exclusionary rule, both in terms of valuation
costs and administrative efficiency, arise out of the fact that the rule
involves the suppression of evidence and not the measurement of a
gain or loss. This distinction points to the major limitation of the rule:
it works only when the motivation for the relevant police conduct was
the gathering of evidence. If a police officer wishes to search a suspect's
house or car not to find evidence, but purely to inflict harm on the
suspect (perhaps out of spite or bias), the threat of exclusion will not
14. See, e.g., sources cited infra note 27.
15. Thus, for example, the American exclusionary rule applies to all violations by
police officers of search and seizure or interrogation law, not merely to violations that a court
finds, after the fact, to have been egregious. See, e.g., Arizona v. Hicks, 480 U.S. 321 (1987)
(applying the exclusionary rule in a case where a police officer, lawfully in the defendant's
apartment, turned over the defendant's stereo to copy down the serial number, on the ground
that the officer should have gotten a warrant for the stereo "search"). For a general discussion
of the role the exclusionary rule plays in American criminal procedure, see Stuntz, The
American Exclusionary Rule and Defendants' Changing Rights, 1989 CRIM. L. REV. 117.

HeinOnline -- 1990 Wis. L. Rev. 491 1990

 WISCONSIN LAW REVIEW

deter the officer. Some other enforcement device will be necessary to

deal with such searches undertaken in bad faith. Thus, under American
law, damage remedies are available when a police officer behaves with
something akin to gross negligence or when both the governing law and
the violation are clear.16 In these cases, the nature of the officer's be-
havior (plus the fact that all American police officers will be familiar
with the exclusionary rule) suggests that exclusion was not an effective
deterrent, perhaps because the gains to the officer from the illegal con-
duct were not primarily evidentiary in nature.
This emphasis on the motivation of those who obtain evidence
may also serve to explain a difference between American and German
law. In the Federal Republic of Germany, evidence can be suppressed
even if it is illegally seized by private parties.17 In American law, only
police illegality triggers the exclusion sanction.1 8 The difference may
reflect different judgments about the hard behavioral, or empirical,
question of why private trespassers wrongfully obtain evidence that is
later used in court. If the motivation of the wrongful private "searcher"
is to find evidence for subsequent use in court against the victim of
the search (as, for example, where an ex-husband breaks into an ex-
wife's house and rummages through her financial records in order to
turn incriminating information over to the tax authorities), then ex-
clusion works in the same fashion as it does for police in the United
States. It accurately disgorges from the wrongdoer (it takes away the
pleasure of revenge in seeing the ex-spouse in trouble with the tax
authorities) without a complicated valuation decision, and it is ad-
ministratively easy because it can be implemented in the subsequent
criminal (or civil tax) prosecution. On the other hand, if the private,
wrongful searcher has other, more self-serving (or perhaps less spiteful)
goals, the suppression of the evidence that is obtained will accomplish
neither disgorgement nor deterrence.
Remarkably, the leading German cases appear (at least to the eyes
of outsiders) to fit the first of these fact patterns, with private parties
trying, usually for personal reasons, to injure other private parties in
some civil or criminal proceeding. 19 In one case, for example, an es-
tranged husband hired a spy who monitored the wife's movements and
16. See Anderson v. Creighton, 483 U.S. 635 (1987).
17. This is true both in civil cases, NJW 1848 (W. Ger. Fed. (Sup.) Ct. 1970), and
in criminal cases, 34 BVerfG 238, 246-47 (1973). These cases are discussed in Baade, Illegally
Obtained Evidence in Criminal and Civil Cases: A ComparativeStudy of a Classic Mismatch
(pt. 2), 52 TEX. L. REV. 621, 622-24 (1974).
18. United States v. Jacobsen, 466 U.S. 109 (1984).
19. See sources cited supra note 17. In one of the cases, the court explicitly stated
that exclusion of the evidence was necessary because otherwise there would be some incentive
for the private trespass-something that would be true only if the trespass were motivated
by evidentiary concerns. See NJW 1849 (W. Ger. Fed. (Sup.) Ct. 1970) discussed in Baade,
supra note 17, at 623.

HeinOnline -- 1990 Wis. L. Rev. 492 1990

1990:483 Remedies and Incentives in Private and Public Law 493

conversations in her home, all for the purpose of gathering evidence

to be used in the couple's divorce proceeding. 20 The best-known Amer-
ican cases, on the other hand, fit the second pattern. The leading Amer-
ican case involved an employee of a private parcel delivery company,
who opened a damaged package, found illegal drugs, and then turned
the package over to the police. 2 1 There is clearly no personal spite
involved in the invasion of privacy, or trespass, in this case, because
the owner of the package and the delivery person were strangers. Con-
sequently, there is no reason to suspect that the searcher had any par-
ticular interest in seeing to the arrest of the owner. Rather, the searcher
was quite plainly motivated by other concerns-chiefly protecting the
company against claims for damage to the package's contents. Where
the motives for search are of this sort, exclusion will likely be an in-
adequate deterrent because it does nothing about the non-evidentiary
reasons for the trespass or illegal search. More generally, the differences
between the American and German rules may have something to do
with the facts of the particular cases that worked their way to important
courts.
In theory, even when the motivation for the wrongful search is
purely evidence-seeking, and motives are not mixed, exclusion alone
might be an inadequate deterrent. As noted earlier, disgorgement alone
leaves the wrongdoer indifferent about the wrongful activity; it forces
disgorgement of the gains but nothing more. Some "premium," or
restitution-plus-a-fine, might therefore be necessary. 22 In the case of
wrongfully obtained evidence, this premium may exist quite naturally
in the form of a skewed distribution of outcomes at the time of the
illegal search. Suppose an officer is deciding whether to search a sus-
pect's house without probable cause and without a warrant (which,
under American law at least, would make the search illegal). Imagine
further that the reason for the search is to look for illegal drugs. If the
officer finds the drugs, the evidence will be suppressed unless the officer
can later show that discovery was inevitable-in other words, that the
search was not the but-for cause of the discovery of the illegal sub-
stance. 23 In practice, such a showing can very rarely be made; in order
to win, the officer would probably have to satisfy the court that other
officers had already sought a warrant to search the house. 24 Yet, the
probabilistic truth is that if the illegal search were not undertaken, there
20. NJW 1848 (W. Ger. Fed. (Sup.) Ct. 1970) discussed in Baade, supra note 17,
at 622-23.
21. Jacobsen, 466 U.S. at 109. The package had been torn by a forklift, and the
employee opened it to see if any damage had been done to the contents. Id. at I 11.
22. See supra text accompanying note 2.
23. Nix v. Williams, 467 U.S. 431 (1984).
24. See, e.g., Segura v. United States, 468 U.S. 796 (1984).

HeinOnline -- 1990 Wis. L. Rev. 493 1990

 WISCONSIN LAW REVIEW

would be a substantial likelihood that the evidence would be obtained

because the suspicious officer can continue to gather more information,
and may eventually satisfy the probable cause standard and then un-
dertake a legal search of the house. If this opportunity-cost story is true
often enough, then suppressing the evidence found in illegal searches
is not the same as simply forcing officers to disgorge what was gained
from wrongful searches. Instead, the exclusionary rule requires that the
wrongful searcher give up the gains from misconduct (the evidence)
and sacrifice the possibility (which exists, but is surely less than 50%)
that the very same evidence would be legally obtained later in the
investigation. In other words, the illegal search may itself lower the
criminal defendant's chances of conviction-and this reduction can be
seen as the premium that the exclusionary rule extracts.
Given the existence of this natural premium, there is no need to
apply the exclusionary rule to evidence that a wrongful searcher stum-
bles upon with no foreseeability. And American law reflects this dis-
tinction with precision, because everything the police find as a result
of an illegal search is excluded, except for evidence discovered in a
manner extremely remote from the misconduct.2 5 The classic example
is the witness who comes forward months after an illegal search, but
who probably would not have done so had the illegal search never
occurred. It seems fair to assume that such rare and unforeseeable
events do not significantly affect police incentives to search; conse-
quently, American law does not exclude the witness' testimony. 26 All
this suggests that a fairly strict exclusionary rule, subject to foreseea-
bility limitations, is a reasonably efficient means of policing the police,
at least insofar as the relevant police conduct is aimed at evidence-
gathering. The real descriptive puzzle for comparative lawyers, then,
is why such a rule is not more common outside of American law. Even
in countries that have an exclusion, or disgorgement, sanction for some
kinds of police misconduct, such as the Federal Republic of Germany
and Great Britain, the sanction tends to be imposed only in cases of
gross misconduct. 27 Given the nature of restitution rules, such a stan-

25. Wong Sun v. United States, 371 U.S. 471 (1963). This rule is modified in one
respect: only the victim of the illegal search (defined as the party whose privacy was invaded)
may invoke the exclusionary rule. Rakas v. Illinois, 439 U.S. 128 (1978). This is problematic
to the extent that the officer may be induced to search illegally in suspect A's house if the
searcher expects to find evidence that will incriminate suspect B. For this problem to exist,
however, the searching officer must be willing to trade off in advance any likelihood of catching
A. Particularly given the premium discussed earlier, this may happen only rarely.
26. United States v. Ceccolini, 435 U.S. 268 (1978).
27. For general English-language discussions of the German exclusionary rule, see
Bradley, The Exclusionary Rule in Germany, 96 HARV. L. REv. 1032 (1983); Pakter, Exclu-
sionary Rules in France, Germany, and Italy, 9 HASTINGS INT'L & COMP. L. REv. 1, 38-48

HeinOnline -- 1990 Wis. L. Rev. 494 1990

1990:483 Remedies and Incentives in Privateand Public Law 495

dard must surely lead to underdeterrence, as it forces police officers to

disgorge only when misconduct was severe. Nor do these legal systems
fill this deterrence gap with a remedy for damages; successful damages
actions against the government or individual police officers are at least
as rare in these countries as in the United States (where the exclusionary
rule is plainly the primary deterrent to police illegality). Why this dif-
ference in various countries' rules, given the apparent advantages of
the exclusionary rule?
The answer lies not in the virtues of other legal remedies, but in
political sanctions, which constitute a remedy that is even less costly
in administrative terms than is disgorgement, or exclusion. The gov-
ernment needs to worry about public approval, since disapproval may
lead to reduced funding for, or increased restrictions on, government
activities. If the police department regularly breaks into voters' homes
without good cause, elected officials will surely spring into action. When
such political accountability exists and functions well, it is much
cheaper to administer than any remedy that relies on litigation for its
enforcement.
Unfortunately, political checks are of uneven quality. This is par-
ticularly so where crime, and therefore police investigation, is concen-
trated in communities that are not very powerful politically. Imagine,
for example, that (1) one ethnic or racial group's members constitute
a solid majority in a given jurisdiction; (2) the same jurisdiction's high-
est-crime neighborhood is overwhelmingly populated by members of
a different ethnic or racial group; and (3) voting is race-based-that is,
biracial political coalitions tend to be unsuccessful. In these circum-
stances, political pressure would not lead to optimal police investiga-
tion, since the costs of such investigation to the minority group would
not be internalized by elected officials, who are accountable, by hy-
pothesis, only to the majority group.
It may be that the exclusionary rule in the United States is broad,
and that search and seizure law is relatively restrictive of police, because
in the United States there is substantial experience with ethnic and
racial tensions of this kind. Other countries, with less demographic
heterogeneity or with less history of majority-minority inequalities,
may be understandably more inclined to use the exclusionary rule only
as a secondary remedy. Exclusion, or disgorgement, may be the most
efficient legal remedy with which to deter illegal evidence-gathering,
but it is, after all, surely less efficient than a system which can rely on
political and cultural checks.
(1985). On the English rule, see Ashworth, Excluding Evidence as Protecting Rights, 1977
CiuM. L. REV. 723; Sanders, Rights, Remedies, and the Police and CriminalEvidence Act,
1988 CRIM. L. REV. 802, 806-08.

HeinOnline -- 1990 Wis. L. Rev. 495 1990

 WISCONSIN LAW REVIEW

IV. DAMAGES AND DISGORGEMENT IN TAKINGS LAW

We turn, finally, to one other area of public law where it is useful

to think about the difference between damages and disgorgement. There
are a number of well-known reasons which support or explain rules
requiring a government to pay when it takes or devalues private prop-
erty. The prominent efficiency-based reasons include: (1) the idea that
uncompensated takings generate an activity-level effect, so that a gov-
ernment which did not compensate for "investment-backed expecta-
tions" it destroyed would find that there would be too little investment
in some activities; (2) the argument that if majorities could freely ex-
ploit the property of minorities there would be inefficient over-invest-
ment in rent-seeking and in coalition formation; and (3) the notion
that it is desirable to push government decision makers to "internalize"
the burdens imposed on others, and a government which must raise
revenues to pay those who are burdened by its actions (or justify such
expenditures to voters) will think about the burdens it imposes.28 On
the other hand, there are efficiency costs to taxes, so that raising money
is not socially costless, and a rule requiring payments for all burdens
would surely lead to an inefficiently low level of government activity.
Virtually all legal systems compromise, as it were, between these ar-
guments, requiring payments for most (but not all) physical invasions
of private property, and for some (but not most) devaluations of prop-
erty, or "regulatory takings."
Here, the contrast between damages and disgorgement is revealing
in several ways. When takings law requires that the government com-
pensate a private property owner, the government must pay the fair
market value of the property without any premium for the value added
because of the government's plans or projects of which the taking is a
part.29 A rule which instead allowed the property owner to extract the
greater of the value of a property to its owner or the value of the property
to the government would reflect the principle of disgorgement, as would
a rule which gave the post-taking value of the property to the private
owner. Instead, the actual rule is best described as allowing recovery
only for damages.
28. R. POSNER, ECONOMIC ANALYSIS OF LAW 51 (3rd ed. 1986) (efficiency of in-
ternalizing cost of land used to construct public buildings).
29. See, e.g., United States v. Virginia Elec. & Power Co., 365 U.S. 624, 636 (1961)
("The value of the easement must be neither enhanced nor diminished by the special need
which the Government had for it"). For the equivalent German rule, see Kimminich, Com-
pensationfor Expropriationof Land andfor "Worsenment" in the FederalRepublic of Ger-
many, in COMPENSATION FOR COMPULSORY PURCHASE: A COMPARATIVE ANALYSIS 206 (J.
Garner ed. 1975) (compensation, but no profit from government action). But see Ehlers,
Compensationfor Compulsory Purchase in Denmark, in id. at 178 (claiming that compen-
sation looks to price at time of surrender so that "expectations increase the compensation
claim").

HeinOnline -- 1990 Wis. L. Rev. 496 1990

1990:483 Remedies and Incentives in Privateand Public Law 497

None of these rules would be devastatingly inefficient. There is the

danger that a generous disgorgement rule, such as one which gave the
property owner the value of the property to the government, would
lead to corrupt politics because property owners would want to have
their property taken by the government in return for generous pay-
ments. But this is the familiar danger associated with many government
procurement plans, and it is one that we normally know must be mon-
itored rather than avoided at all costs.
A better argument against the more generous restitution rules is
that the pre-taking value of a property is generally easy to assess because
one can examine previous sales of this and similar properties. In con-
trast, the government's reservation price and the post-taking value of
the property are hypothetical or largely unknown amounts. In this set-
ting, unlike that in torts,3 0 it is therefore arguable that the administra-
tive costs of the restitution-based system are indeed greater than those
associated with damages. One might also argue that the internalization
function of the damages rule would be poorly carried out by a restitution
rule, which would often leave the government's decision makers in-
different as to whether to pursue a plan because only a very few voters,
the generously compensated property owners, would be made better
31
off.
The damages-disgorgement dichotomy also draws attention to a
very different point about takings law. Many government projects, in-
cluding ones which take and pay for private property, also cast benefits,
or windfalls, on various private parties. A new highway, for example,
is likely to involve: (1) the physical taking of property from owners
who will be compensated; (2) the devaluation of property interests
which are near enough to the highway to suffer from the pollution and
other negative side effects of the road; and (3) the appreciation of prop-
erties which benefit from improved transportation or from better access
to markets. Much as legal systems virtually never compensate (with
damages or with restitution payments) for those interests falling in the
second category, no disgorgement is required from those falling in the
third group. It is possible to imagine either the government or the
owners of property in the second category collecting in this manner;
one might be tempted to think of payments by the government as
representing damages, while payments from the "winners" to the "los-
30. See supra text accompanying notes 4-5.
31. There is still a social benefit to the plan, but its value will be captured through
the disgorgement remedy by those whose property interests are taken. The political impetus
for the project must come from these owners. In contrast, under the familiar damage rule,
a few property owners receive (minimal) compensation for their physically taken interests,
many are burdened without compensation-and are likely to lobby against the project-and
often very many more will receive windfalls. The last group forms the political coalition in
favor of the project.

HeinOnline -- 1990 Wis. L. Rev. 497 1990

 WISCONSIN LAW REVIEW

ers" would reflect a plan of restitution, but to the extent that the gov-
ernment simply funded its liability through taxes on the winners in the
third group, it is convenient to think of either scheme as based on the
restitution, or disgorgement, idea. Such a plan, which through one
means or the other collected from those who were benefited and com-
pensated those who were burdened, would obviously promote the in-
ternalization function of takings law. But inasmuch as modem legal
systems rarely collect or compensate in this manner, such schemes are
mostly of theoretical interest at this point. 32 There is some legal prec-
edent for thinking about the benefits as well as the burdens of govern-
ment activities. In Australia, as elsewhere, there were rules of resump-
tion, under which land grants included a reservation of some percentage
(on the order of six to twenty percent) of land on which the government
could later construct highways (and in some places reservoirs and other
projects) with no requirement for compensation to private property
owners.3 3 Some legal systems simply afforded no compensation for
property taken in order to build roads. 34 It is noteworthy that these
legal rules normally applied in areas with low population densities, and
it is likely that property owners benefited when a highway or reservoir
was built nearby. In other words, the damage or restitution claim that
might have been brought by the landowner whose property was taken
would likely have been offset by a counterclaim for the property owner's
undeserved enrichment (or the absence of net damages). The doctrines
in question might in this way be thought of as rules of thumb meant
to save the administrative costs of actual damage (or enrichment) de-
terminations.
The symmetry that is reflected above in the listing of the second
and third categories of affected property interests is apparent; one can
scarcely imagine extractions of windfalls from the third group (con-
sisting of those who receive windfalls from government projects) with-
out including payments for losses suffered by the second (those who
are burdened, although their property is not physically taken or com-
pletely destroyed). In a transaction-cost-free world, restitution (whether
available directly from the winners in the third group or from the
government itself) would almost surely be a better rule for takings,
because costs and benefits would be incorporated into a government's
calculations. Once the costs of measuring benefits (windfalls) and costs

32. An important strand of the takings literature has favored such transfers. See
WINDFALLS FOR WIPEOUTS: LAND VALUE CAPTURE AND COMPENSATION (D. Hagman & D.
Misczynski 1978).
33. Brown, Compensation for land Acquisition in Australia,in COMPENSATION FOR
COMPULSORY PURCHASE, supra note 29, at 73-74.
34. J. SACKMAN, NICHOLS' THE LAw OF EMINENT DOMAIN § 1.22[l] (1973) (first
statute requiring compensation for road building in colonial America passed in Massachusetts
in 1639).

HeinOnline -- 1990 Wis. L. Rev. 498 1990

1990:483 Remedies and Incentives in Privateand Public Law 499

(burdens) are considered, it is easy to see why takings law has not
experimented much with restitution. In the tort law setting, it is well
known that the damage rule is an excellent means of internalizing cal-
culations of costs and benefits, and we suggested that restitution, or
disgorgement, could come close to doing the same only at great cost.
Here, in takings law, the situation is half reversed, especially because
we need to measure pre-takings values in a post-takings world. The
disgorgement rule, with symmetrical application, is now the great in-
ternalizer, but its implementation costs are also great.

V. CONCLUSION

A good deal of traditional, doctrinal legal scholarship is of little

interest or use beyond the borders of the jurisdiction which generated
the doctrines in question. The opposite is the case for the law-and-
economics literature. Analyses of the incentive effects (and the content)
of civil law rules are of great interest to the common lawyer, for ex-
ample, because the tools of different legal systems are very much the
same. We have tried to show that in various areas of law, where distinct
rules prevail in different countries, it is very much the case that there
are common analytic themes. In particular, we have stressed the use-
fulness of thinking about the relative advantages of remedies based on
one party's losses or another's gains, and we have tried to demonstrate
that a variety of legal topics can be illuminated by exploring this com-
petition between damage and disgorgement remedies.

HeinOnline -- 1990 Wis. L. Rev. 499 1990

HeinOnline -- 1990 Wis. L. Rev. 500 1990