F ordham Law School

FLASH: The Fordham Law Archive o f Scholarship and History

F a cu lty Scholarship

2000

A b u s iv e B re a c h o f C o n tra c t, T h e

N ich o las J. Joh n son

F o r d h a m U n iv e r s ity S c h o o l o f L a w , [email protected]

F o llo w this and add ition al w o rk s at: h ttp ://ir.law n et.fo rd h am .ed u /facu lty_ sch o larsh ip

o P art o f th e C o n tra cts C o m m o n s

R e c o m m e n d e d C ita tio n
Nicholas J.Johnson, Boundaries of Extracompensatory Relief fo r Abusive Breach of Contract, The , 33 Conn. L. Rev. 181 (2000-2001)
Available at: http://ir.lawnet.fordham.edu/faculty_scholarship/433

T h is A rtic le is brought to y o u for free and open access b y F L A S H : T h e Fo rd h am L a w A rch iv e o f S ch olarship and H istory. It has been accepted for

in clu sio n in F a c u lty S ch olarship b y an authorized adm inistrator o f F L A S H : Th e Fo rd h am L a w A rc h iv e o f S ch olarship and H istory. F o r more

inform ation, please contact tm eln ick@ law .fordham .edu .

 T h e B o u n d a rie s o f E x tra c o m p e n sa to ry R e lie f fo r
A b u s iv e B re a c h o f C o n tra c t

Nicholas J. J ohnson*

I. Introduction
The idea of extracompensatory damages for abusive breach of contract
presents a fundamental conflict.*1 Contract doctrine aims to facilitate ex­
changes. Extracompensatory damages are disincentives. These aims are
essentially irreconcilable. And traditionally the goal of facilitating ex­
changes has trumped any interest in punishing bad conduct. But there is a
lingering sense that sometimes a proportionate response to bad conduct
surrounding breach requires more than the traditional measure of damages.
At the edges of contract doctrine, two notable experiments manifest the
sense that some breaches demand more than compensatory damages. One,
the failed California experiment with bad faith breach, permitted the plain­
tiff to collect punitive damages for defendant’s “bad faith” denial of the
existence of a contract.2 Criticism of the bad faith breach was rife.3 Lower

Professor of Law, Fordham University School of Law. J.D., Harvard Law School, J934. I
would like to thank Joe PeriUo, Jill Fisch, Steve Thel, Chantal Thomas and Susan Block-Liebfor their
comments on drafts of this Article. Thanks to Ellen Johnsonfor research and editing work on this
Article.
1. I use “abusive breach” here to capture the various articulations of bad behavior surrounding
breach; e.g., bad faith breach, tortious breach, and willful, opportunistic breach. I use “ex­
tracompensatory” damages to connote not only punitive damages, but expanded (tort measure) conse­
quential damages of the type that some propose as a better choice than punitive damages. See, e.g.,
Thomas A. Diamond &Howard Foss, Proposed Standardsfor Evaluating When the Covenant of Good
Faith andFairDealing Has Been Violated: A FrameworkforResolving the Mystery, 47 HASTINGSLJ.
585,585-86(1996).
2. Seaman’s Direct Buying Serv., Inc. v. Standard Oil Co., 686 P2d 1158, 1167 (CaL 1934),
overruled by Freeman &Mills, Inc. v. Belcher Oil Co., 900 R2d 669 (CaL 1995).
Courts and commentators seem to have used the terms “tortious breach" and “bad faith breach”
interchangeably to describe the decision to award punitive damages for violation of the implied cove­
nant of good faith. See James H. Cook, Comment, Seaman’s Direct Buying Service, Inc. v. Standard
Oil Car Tortious Breach ofthe Covenant ofGoodFaith and FairDealing in a Noninsurance Commer­
cial Contract Case, 71 IOWAL. Rev. 893,894 n.12 (1986).
3. See, e.g., C. Delos Putz, Jr. &Nona Klippen, Commercial Bad Faith: Attorney Fees—Not Tort
Liability—Is the Remedyfor “Stonewalling", 21 U.S.F. L. REV. 419,499 (1987).

181
182 CONNECTICUTLAWREVIEW [Vol. 33:181

courts had difficulty drawing even rough boundaries around the concept4
and split over many questions.5 The California Supreme Court ended the
confusion in a 1995 decision scuttling the bad faith breach.67
Another experiment, allowing the award of punitive damages against
insurance companies for bad conduct breaches, is an enduring exception to
the general bar on extracompensatory damages in contract law? The
model case is a denial of coverage on some specious pretext.8
With the failure of the California experiment, we are tempted by basic
tenets of contract doctrine to dismiss the insurance cases as singular aber­
rations in a doctrine that properly limits recovery for breach to what we
loosely call compensatory damages.9 It is easy to understand the tradi­
tional response to breach of contract as a necessary element of our core
commitment to making the cost of contracting—viz., the price of breach—
certain and predictable.10 This predictable exit price limits the risk of mak­
ing promises and encourages trading.11 Any escalation of damages for bad
conduct or otherwise would raise the cost of promise-breaking and thus
discourage exchanges. Because it is difficult to harmonize ex­
tracompensatory damages with this formula, courts typically refuse to
award punitive damages for breach of contract and narrowly limit conse­
quential damages to those that were easily contemplated when the deal was

4. See Freeman & Mills, Inc., 900 P.2d at 676 (Cal. 1995) (commenting on the confusion sur­
rounding bad faith breach).
5. See Robert L. Rancourt, Jr., Note, Freeman &Mills, Inc. v. Belcher Oil Co..* Yes, The Seaman’s
Tort is Dead, 27 PAC. L.J. 1405,1406-07 (1996) (illustrating the disagreement as to whether to expand
the concept of bad faith breach to include bad faith assertion of a defense to breach and whether bad
faith breach was grounded on the implied covenant of good faith or an expanded view of tort liability).
6. Freeman &Mills, Inc., 900 P.2d at 669.
7. California generated one of the first such decisions. See Comunalc v. Traders & Gen. Ins. Co,
328 P.2d 198 (Cal. 1958). See also Susan D. Gresham, Comment, "Bad Faith Breach": A New and
Growing Concernfor Financial Institutions, 42 Vand. L. Rev. 891, 892 & n.5 (1989) (tracking deci­
sions from the jurisdictions that followed the California courts in allowing tort recovery for bad con­
duct breaches of insurance contracts).
8. For a representative description of the insurance cases, see Roger C. Henderson, The Tort of
Bad Faith in First-Party Insurance Transactions After Two Decades, 37 ARIZ. L. REV. 1153 (describ­
ing liability for punitive damages as being triggered by outright refusals to pay, delays,
misinterpretation of records or policies for the purpose of defeating coverage, using threats to force
unfair settlement, falsely accusing the insured of wrongdoing, etc.). See also Douglas R. Richmond, An
Overview of Insurance Bad Faith Law and Litigation, 25 SejONHALLL. REV. 74, 74-76 (1994) (ex­
amining insurance bad faith law and litigation in both the third-party and first-party contexts); Chris
Michael Kallianos, Comment, Bad Faith Refusal to Pay First-Party Insurance Claims: A Growing
Recognition of Extra-Contract Damages, 64 N.C. L. REV. 1421 (1986) (describing factors that should
be present in insurance and non-insurance tortious breach cases before extracompensatory damages are
assessed).
9. See Alan O. Sykes, "Bad Faith" Breach of Contract by First-Party Insurers, 25 J. LEGAL
Stud. 405,409 (1996).
10. See, e.g., Fbley v. Interactive Data Corp., 765 P.2d 373,389 (Cal. 1988).
11. See Applied Equip. Corp. v. Litton Saudi Arabia Ltd., 869 P.2d 454,460 (Cal. 1994) (limiting
damages to those reasonably foreseeable at the time of contracting encourages exchanges by allowing
parties to accurately gauge the risk of contracting at the outset).
2000] R EL IE F FOR A B U SIV E BRE A C H O F CO NTRACT 183

struck.12 Compensatory damages, the argument goes, are as close to opti­

mal as we are likely to get.13 This tempts us to dismiss the insurance cases
as aberrations and to view the California experiment as confirmation that
extracompensatory damages are not compatible with contract doctrine.
But have we been too quick to dismiss these experiments with ex­
tracompensatory damages? Are the insurance cases really so specialized
that they offer no general lessons about responding to abusive breach?
Was the California experiment so flawed that nothing useful can be sal­
vaged from it?
This Article shows that a unified analysis of the two efforts yields an
insight that delineates our capacity to respond to abusive breach. Consid­
ered together, they show that rules punishing abusive breach with ex­
tracompensatory relief are indeed viable so long as their impact is limited
to subcategories that are conceptually and practically severable from the
general pool of transactions. Severability is the core requirement of any
workable extracompensatory response to abusive breach. If we respect it,
then we can punish subcategories of abusive breach with aggressive disin­
centives. Failure to respect severability will violate the basic aims of con­
tract doctrine by leaking collateral risk into the general pool of transac­
tions, thus discouraging exchanges.
The severability insight yields several lessons that enrich our view of
damages for abusive breach. First, a general standard of abusive breach
exhibits the lowest degree of severability and is not viable because it ele­
vates risk in the entire pool of transactions. Second, we achieve the highest
degree of severability (and thus encounter the best opportunities to award
extracompensatory damages without leaking collateral risk) in categories
that key on the type of transaction rather than the manner of the breach.
For example, saying that we will award punitive damages for abusive
breach in insurance contracts (keying initially on the type of transaction
with abuse determined by violation of some intra-category norm) insulates
that risk from the general pool. But saying that we will award punitive
damages for willful, opportunistic or bad faith breach (keying exclusively
on the manner of the breach) elevates risk in all transactions. Third, we
can discern a large category of cases exhibiting a moderate degree of sev­
erability where a viable response to abusive breach depends on the process
of implementation. In that category, the legislative process has crucial

12. See Restatement(Second) of Contracts § 355 cmt. a (1979). Professor Farnsworth con­
tends that the first statement that punitive damages will not be awarded in contract law appeared in a
1909 English case. 3 E. ALLANFARNSWORTH, FARNSWORTHONCONTRACTS§ 12.8, at 192 n.17 (2d
Ed. 1998).
13. See Alan Schwartz, The Myth that Promisees Prefer Supracompensalory Remedies: An Analy­
sis of Contractingfor Damage Measures, 100 YALELJ. 369, 372 (1990); Steven Shavell, Damage
Measuresfor Breach of Contract, 11 BELLJ. ECON. 466,467 (1980). But see William S. Dodge, The
Casefor Punitive Damages In Contracts, 48 DUKELJ. 629,654-55 (1999).
184 CONNECTICUTLAWREVIEW [Vol. 33:181

advantages over the common law process.

Part II will show how severability is essential to mediating the tension
between our commitment to predictability of the cost of contracting and the
uncertainty introduced by extracompensatory damages. Parts III and IV
will consider what is required to achieve severability. Part V shows that
particular types of responses to abusive breach will exhibit differing de­
grees of severability that control the viability of those responses. Part VI
shows that the viability of extracompensatory responses exhibiting moder­
ate degrees of severability depends on the process of implementation— i.e.,
legislation versus common law.
Finally, it is not my aim to engage substantively the large body of eco­
nomic analysis arguing against the award of extracompensatory damages in
contract law or the numerous economic criticisms of punitive damages
generally. My effort here relates to those projects only indirectly, in two
respects: first, in the sense that it is indirectly critical of their predomi­
nately economic focus, to the exclusion of insights that can be drawn di­
rectly from the doctrine; and second, in its assumption throughout that the
decision to punish particular subcategories of abusive breach with ex­
tracompensatory damages is not just an economic one, but also must take
into account social, political, equitable and systemic considerations. How­
ever, except for purposes of illustration in Part VI, I do not make any gen­
eral or particular normative claims favoring extracompensatory damages.14
My aim is only to show that any efforts to impose extracompensatory dam­
ages must operate within the limits that I set out here.

II. A Predictable Exit Price and Severability

A prime tenet of contract doctrine is certainty and predictability of the
cost of contracting.15 Traditional damages rules permit the promisor to
gauge with relative certainty the price of breach—viz., the risk attached to
promise-making. As Professor Farnsworth explains, our system of contract
remedies “is not directed at compulsion ofprom isors to prevent breach. Its
preoccupation is not with the question: how can promisors be made to keep
their promises?”16 The aim instead is to encourage trading through en­

14. This Article will not engage the question of whether awarding extracompensatory damages Is a
good idea in any particular case (with one exception used for purposes of illustration in Part VI). Some
of the commentators cited herein have argued that extracompensatory damages for breach of contract
always are a bad idea from an economic perspective. As suggested below, that is too narrow a per­
spective. The decision is notjust economic, but also must take into account social, political, and equi­
table considerations. However, except for purposes of illustration in Part VI, I do not attempt to make
general or particular normative claims favoring extracompensatory damages. See infra text accompa­
nying notes 66-74 (making the case against the “late-pay scheme” for purposes of illustration).
15. See, e.g.,Foley, 765P.2dat389. See also Dodge, supra note 13, at 695-98.
16. Farnsworth, supra note 12, § 12.1, at 147.
2000] R E U E F FOR A B U SIV E BREAC H O F C O N TRA C T 185

forceable contracts.17 One of the primary ways that contract doctrine en­
courages trading is through a low and predictable cost of promise-making.
Remedies for breach of contract provide obvious incentives to promisors.
Compensatory damages are not literally compensatory.18 Non­
recoverability of litigation expenses, and limitations on consequential dam­
ages, prevent the promisee from recovering fully the benefit of her
bargain.19 But the virtue of our traditional damages rules is that they give
the promisor the ability to gauge with relative certainty the cost of con­
tracting—viz., the price of breach. Promisees, we presume, are in a much
better position than promisors to predict and insure against the collateral
costs of breach.20
Alan Schwartz argues that the predictable expectancy measure of dam­
ages brings the cost of contracting to its optimal level. Supracompensatory
damages, he argues, would turn the contract into a gamble that informed,
risk-averse traders always would choose to avoid.21 The added risk of ex­
tracompensatory damages goes directly to the bottom line cost of the trade,
causing the promisor to charge too much and the promisee to pay too much
for the good or service.22 Some trades will be deterred altogether.
We see this concern explicitly in the evolution of California’s bad faith
breach. As the experiment developed, some appellate courts extended the
concept of bad faith breach to include not only bad faith denial of existence
of a contract, but also bad faith assertion of a defense to breach.2342 Other
courts resisted, emphasizing that such expansion of the bad faith breach
would inject a large dose of collateral risk into the general pool of transac­
tions.
In DuBarry International, Inc. v. Southwest Forest Industries, Inc.™

17. Id. at 756.

18. Sykes, supra note 9, at 408-09.
19. Id
20. Unpredictability of the exit price raises the risk of contracting and discourages trading. If
parties cannot gauge the fiscal risk of entering into a contract ahead of time, then the trade is more like
a gamble. There may be a chance to guess intelligently about the probability of breach, but the price of
breach is pure speculation. See, e.g.. Applied Equip. Corp. v. Litton Saudi Arabia Ltd., 869 P2d 454,
460 (CaL 1994) (explaining that commercial activity is fostered by limiting contract remedies to dam­
ages reasonably foreseeable at the time of contracting).
21. Alan Schwartz argues that generally rational and fully informed parties will prefer merely
compensatory damages. Schwartz, supra note 13, at 372. Schwartz might be right about basic incen­
tives and general preferences of traders, but there are other factors to consider. The decision whetherto
respond to abusive breach with extracompensatory damages requires a social, political, and systemic
evaluation of the conduct and its interaction with the doctrine. An analysis like Schwartz's is only part
of that normative decision. See, e.g., the discussion of late-pay schemes, infra notes 66-74 and
accompanyingtext
22. See supra note 14.
23. Multiplex Ins. Agency, Inc. v. CaL Life Ins. Co., 235 CaL Rptr. 12, 21 (CL App. 1937)
(concluding that bad faith denial of liability under a contract was enough for punitive damages to at­
tach).
24. 282 CaLRptr. 181 (Cl App. 1991).
 186 CONNECTICUTLAWREVIEW [Vol. 33:181

the court refused to expand the concept, emphasizing that such expansion
would mean that “any party attempting to defend a disputed contract claim
. would risk, at the very least, exposure to the imposition of tort damages
and an expensive and time consuming expansion of the litigation into an
inquiry as to the motives and state of mind of the breaching party.”25 This
would blur the line between contract and tort, trench on their purposefully
different measures of damages, and upset the predictability of commercial
transactions.
The theme appears again in H arris v. Atlantic R ichfield Co.26 There,
the court refused to extend the bad faith breach to punish bad faith asser­
tion of defenses, citing the need for predictability of the consequences of
breach and the destabilizing effect of a doctrine that had the potential to
turn every contract breach into a claim for punitive damages.27
Concerns about the loss of predictability substantially influenced the
ultimate decision to close the California experiment. In Freeman & M ills,
Inc. v. Belcher 0/7,28 the California Supreme Court overturned Seam an’s
D irect Buying Service, Inc. v. Standard O il Co.?9 emphasizing that the bad
faith breach violated two essential tenets of contract doctrine: certainty and
predictability of the cost of contracting, and the traditional emphasis on
compensation rather than punishment.30
DuBarry,312 3H arris'2 and Freeman take us part-way toward recognizing
the importance of severability in fashioning a response to abusive breach.
They confirm that a response to abusive breach that exposes virtually any
breach to the risk of extracompensatory damages is irreconcilable with the
central commitment to a predictable exit price.
But in their overall approach these cases fail to discern the full lesson
of severability. Indeed, die severability insight gets lost in the various
other rationales the courts offer to support their conclusions. The opinions

25. DuBany Int’l, Inc. v. Southwest Forest Indus., Inc., 282 Cal. Rptr. 181,192 (Ct App. 1991)
(determining thatSeaman's could only extend to bad faith denial of a contract’s existence, not a party’s
bad faith denial of liability under a contract).
There was at least a possibility that the bad faith breach in its original configuration might hove
operated in harmony with the principle of a predictable exit price. Initially it applied merely to the
arguably small and potentially severable set of transactions where one party contended that a deal hod
been struck and the other unjustifiably denied it However, extension of the idea to bad faith defenses
captured potentially every contested case, leaving traders to gamble on a court’s conception of a bod
faith defense.
26. 17 Cal. Rptr. 2d 649 (Ct App. 1993).
27. Id. at 654.
28. 900 P.2d 669 (Cal. 1995).
29. 686 P.2d 1158 (Cal. 1984).
30. Freeman &Mills, Inc., 900 P.2d at 674-75 (citing Applied Equip. Corp. v. Litton Saudi Arabia
Ltd, 869 P.2d 454,460 (Cal. 1994), thereby reasoning that parties should be able to gauge ahead of
time the fiscal risk of entering into a contract and thus, damages for breach of contract should be lim­
ited to those that can be foreseen at the time the parties enterthe contract).
31. DuBany Int’l, Inc. v. Southwest Forest Indus., Inc., 282 Cal. Rptr. 181,193 (Ct App. 1991).
32. Harris v. Atlantic Ridgefield Co., 17 Cal. Rptr. 2d 649 (Ct App. 1993).
2000] R E U E F FOR A B U SIV E BRE A C H O F C O N TRA C T 187

argue, among other things, that extracompensatory damages require policy

judgments that only the legislature can make,33 and that extracompensatory
damages discourage efficient breach.34 Ultimately, there were so many
objections to the bad faith breach that the severability lesson was obscured.
The full insight is not apparent until we incorporate the experiment
with extracompensatory damages that undeniably has worked. Consider
now the insurance cases. These cases award punitive damages for bad
conduct accompanying breach35 and yet they thrive, seemingly in harmony
with the core tenet of a generally predictable exit price. Here it seems we
have in fact reconciled inexorably conflicting themes.
The explanation is in the character of the cases themselves. The insur­
ance cases create an easily identifiable island of elevated risk that is practi­
cally and conceptually severable from the general pool of transactions
(where a predictable exit price remains essential). This characteristic (sev­
erability) is vital to any extracompensatory response to abusive breach.
The primary evidence of the insurance cases’ severability is the ab­
sence of indications or even concern that the award of extracompensatory
damages in the insurance cases leaks collateral risk into the general pool of
transactions. The possibility is not even raised in cases considering puni­
tive damages for an insurer’s bad conduct36 Compare the concerns raised
by courts in the evolution of the California experiment Recall the
DuBarry court’s admonition that extending the bad faith breach to cases of
bad faith assertion of a defense would upset the certainty of the cost of
contracting on which the commercial world relied.37
As the doctrine has matured, we have seen significant variation and
expansion in what courts are willing to deem bad conduct by insurance
carriers.38 Indeed, some now criticize that courts have permitted punitive
damages in cases that range well outside the boundaries of any conception
of bad conduct.39 But even here, there is no suggestion that this (arguably
unprincipled) intra-category expansion has disrupted general contracting
incentives. Compare DuBarry, where the court worried that a moderate

33. As I will discuss in detail in Pan VI, the prior notice and potential for more precise delineation
of risk that is inherent in the legislative process makes it the preferred mechanism for responding to
certain categories of abusive breach.
34. See Freeman &Mills, Inc., 900 P.2d at 669; see also Dodge, supra note 13, at 698.
35. See, e.g., supra note 8 andaccompanying text
36. Criticism of the insurance cases has been wide ranging. See, e.g., James M. Fischer, Should
Advice of Counsel Constitute a Defensefor InsurerBad Faith?, 72 TEX. L. REV. 1447 (1994); Calhryn
M. little, Fighting Fire with Fire: "Reverse Bad Faith” in First-Party Litigation Involving Arson and
Insurance Fraud, 19 CAMPBELLL. REV. 43 (1996); Ellen Smith Pryor, Comparative Fault and Insur­
ance Bad Faith, 72 TEX. L. Rev. 1505 (1994). But no one seriously contends that there is any danger
of spillover into the general pool.
37. See DuBarry Int’l, Inc. v. Southwest Forest Indus., Inc., 282 Cal. Rptr. 181, 192 (Ct App.
1991).
38. See, e.g., discussion in Sykes, supra note 9, at 412.
39. Mat 431.
188 C O N N E C T IC U T L A W R E V I E W [Vol. 33:181

expansion to encompass a new aspect of bad faith threatened to leak risk

into the general pool.40
Consider also the work of commentators critical of elevated liability
for breaching insurance carriers. Alan Sykes’ thorough-going economic
criticism of the insurance cases is one place we might expect to see con­
cerns raised about collateral risk.41 Sykes criticizes that there are high error
costs (courts, he posits, are not very good at discerning abusive breaches by
insurers) and that the approach over-deters denial of claims, thus artifi­
cially raising the cost of coverage.42 But nothing in Sykes’ analysis sug­
gests that the award of punitive damages in the insurance cases raises the
cost of contracting in the general pool of transactions.43
The insurance cases endure because they only add risk to a discrete
subcategory of easily identifiable transactions. The decisions have pro­
duced such a buffer between these cases and the general pool of transac­
tions that even the arguably unprincipled expansion of punitive damages
beyond initially perceived categories of bad conduct does not add risk to
contracting in the general pool.44

in. Unpacking Severability and the Range of Severable Options

To this point we can understand severability as a description of the
particular characteristics of a demonstrably viable exception to our stan­
dard damages formula. Severability accounts for the enduring vitality of
the insurance cases. The lack of it helps explain the failure of the Califor­
nia experiment. But to transform severability into a general prescription
for abusive breach we must determine what generally is required to achieve
it.
One reason the insurance cases are easily severable is that they arise
from a peculiar, highly regulated commercial subculture.45 This makes
them amenable to an enhanced damages rule governing an easily segre­
gated class of actors. The regulatory structure and barriers to entry ensure

40. See DuBarry Int’l, Inc., 282 Cal. Rptr. at 192.

41. Alan Sykes argues that supracompensatoiy damages are ill-advised in the insurance cases
becausejudges are not very good at identifying true cases of opportunistic breach and thus over-apply
the remedy to the detriment of all involved. Sykes, supra note 9, at 420.
42. See id. at 424-28.
43. See generally id.
44. This is not surprising. We know it certainly is possible to create severable categories of ele­
vated risk in contract law. Witness for example the treatment of usury loans. Rules penalizing these
transactions elevate risk in a variety of ways. In some jurisdictions the usurious lender forfeits both
interest and principle as punishment for lending above the usury rate. See, e.g., In re Estate of Dane,
390 N.Y.S.2d 249 (App. Div. 1976). This tends to confirm that it is possible to apply risk elevating
rules to discrete categories of activity, without introducing uncertainty into the general pool.
45. For a discussion on the nature of the insurance industry, see ERICMILLS HOLMES& MARKS.
Rhodes, Holmes’s Appleman on Insurance, 2d (Eric Mills Holmes ed., 1996) and John Alan
Appleman &jean appleman, Insurance Lawand practice (1981).
2000] R E L IE F F O R A B U SIV E B R E A C H O F C O N TRA C T 189

that general traders will not fall into this class inadvertently. But is an
overarching regulatory structure crucial to achieving severability?
The best indications are that such a structure is not essential. While the
bad faith breach and the insurance cases are the most notable examples,
there are other more obscure experiments with extracompensatory damages
in contract law. These efforts suggest that severability can be achieved
outside the context of a regulated industry. Moreover, they show that sev­
erability has been tacitly respected in the common law for decades.
Historically, punitive damages have been awarded in cases of abusive
breach of contract by public utilities and marriage promisors.46 These ef­
forts are mere relics today, with both sets of cases removed from the ambit
of common law doctrine. It is difficult to discern what, if any, broad theo­
retical influences fueled common law courts to allow the exceptional rem­
edy of punitive damages in these peculiar cases. But it is evident that both
sets of cases were, like the insurance cases, highly severable, keying on the
type of transaction rather than the manner of the breach.47
Other signals supporting the severability theme and confirming that a
background regulatory structure is not a prerequisite appear in the modem
doctrine of several states. At least four states currently allow punitive
damages where the breach of contract occurs in the context of a special
relationship.48 Indeed, one scholar treats the insurance cases as just another
in a list of special relationships where breach can result in ex­
tracompensatory damages.49 The other relationships “include [ ] carrier
and passenger, innkeeper and guest, physician and patient, and attorney

46. See 11 SAMUELWlLUSTON&WALTERRE. JAEGER, A TREATISEONTHELAWOFCONTRACTS

§ 1340 (3d ed. 1968). See Dodge, supra note 13, at 636 no24-25.
47. There is little indication that the judges in die public utility and marriage promisor abusive
breach cases were consciously attempting to reconcile the risk elevating disincentives of punitive
damages with tire risk minimizing incentives oftraditional damages. See Id.
48. WilliamDodge describes these cases in detail. Id. at 640-41.
In California, we also find special cases where extracompensatory damages for breach are
keyed on the characteristics of the transaction. See, e.g., Rogoff v. Grahov,ski, 246 Cal. Rptr. 185,
189-91 (CLApp. 1988) (remanding to determine whether the plaintiffstated a statutory cause of action
under California Public Utilities Code); Commercial Cotton Co. v. United Cal. Bank, 209 Cal. Rptr.
551,554 (Ct App. 1985) (finding that the relationship of bank to depositor was quasi-fiduciary, there­
fore bank could not reasonably claim non-existent legal defenses to its negligent disbursement of the
plaintiffs funds); Wallis v. Superior Court, 207 Cal. Rptr. 123,129-30 (CLApp. 1984) (finding recov­
ery in tort against an employer that breached its agreement to pay its employee severance pay on the
groundthat the parties shared a relationship analogous to that of a disability insurer and insured). It is
suggested thatthe overruling ofSeaman’s Direct BuyingServ. v. StandardOil Co., 686 P2d 1158 (CaL
1984), casts doubt on the vitality of these cases. See Price v. Wells Fargo Bank, 261 Cal. Rptr. 735,
741 (CLApp. 1989) (suggesting thatFoley v. Interactive Data Corp., 765 P2d 373 (Cal. 1988), signals
the endof the “quasi-fiduciary” category of cases).
49. Dodge, supra note 13, at 648. In describing the peculiar and compelling characteristics of the
insurance cases, courts generally emphasize the special relationship between the insurer and the in­
sured.
190 C O N N E C T IC U T L A W R E V I E W [Vol. 33:181

and client.”50
These special category cases, both early and current, suggest that an
overarching regulatory structure is not vital to severability. More impor­
tantly, they reveal its core requirement.

IV . TH E CORE OF SEVERABILITY

There is a lesson in what the special category cases do not attempt.

They do not attempt to sanction abusive breach through a general
standard—e.g., opportunistic breach—that overhangs every transaction.
Risk attached primarily to such generally described bad conduct inevitably
leaks into the general pool. Under such a standard, every trader in every
trade runs some chance of having his conduct deemed opportunistic.
Like the insurance cases, the special category cases key on the type of
transaction rather than the manner of the breach. It is transaction type that
signals elevated risk. Defining the category this way isolates risk so
clearly that traders in the general pool can ignore it. Keying on transaction
type rather than the manner of the breach seems essential to minimizing
collateral risk. Traders in the targeted category have prior notice of ele­
vated risk. Traders operating outside these areas are free to contract and
breach without fear that the manner of their breach will escalate damages.
This maintains predictability of the cost of trading in the general pool of
transactions.

V . D e g r e e s o f Se v e r a b il it y

We know that severability is achievable with rules that key on the

characteristics of the transaction and difficult to achieve through standards
that focus purely on the manner of the breach. But how should we map
the possibilities? Are we facing a constellation of random opportunities to
respond to abusive breach? Or can we impose more structure on the sever­
ability insight?
Let us first establish a fixed point that represents the lowest degree of
severability and the least viable response to abusive breach. It is repre­
sented by the recommendation from William Dodge’s recent economic
critique, The Case fo r Punitive Damages in Contract Law , favoring the
award of punitive damages for abusive breach of contract.51 I raise this
point here not to criticize the proposal substantively (it is a response to the
numerous economic criticisms of punitive damages for breach of contract
and I am sympathetic to much of it). Rather, I use its ultimate prescription
to illustrate the most extreme version of the mistake we make by failing to

50. Id .
51. See ge n era lly id.
2000] R E U E F FO R A B U SIV E BRE A C H O F C O N TRA C T 191

respect severability and how easy that m istake is to make.

U ltim ately, Dodge contends that punitive damages ought to be avail­
able fo r any w illfu l, opportunistic breach.52 In essence. Dodge proposes a
general standard o f abusive breach that allow s us to ask in every case
whether the manner o f the breach warrants the aw ard o f punitive
damages.53 This proposal exhibits an extrem ely lo w degree o f severability.
A s a reference point, it helps us to position the C alifo rn ia experim ent and
the insurance cases, and u ltim ately w ill help us to discern and position a
category o f d iffic u lt cases where severability depends largely on the proc­
ess o f im plem entation ( i e . , legislation versus common law developm ent).
R ecall our discussion o f the C alifo rn ia experim ent T h e nebulous bad
fa ith standard, com bined w ith the m yriad open questions le ft by the court
in S e a m a n ’s , 5* introduced into C alifo rn ia contract doctrine a broad-ranging
notion o f abusive breach. Th e broad trigger “bad faith ”— that is, conduct
that violates accepted notions o f business ethics555
6
— potentially captured a
great deal and gave little guidance about w hat conduct was safe fro m the
ris k o f punitive damages.
This was particularly true after some C alifo rn ia courts expanded the
bad faith breach to include bad fa ith assertion o f a defense to contract li­
a b ility.57 T h at failu re prevented the concept o f bad faith breach from
w orking in harmony w ith the param ount goals o f re lia b ility and predict­
a b ility o f the cost o f contracting. Appended to post breach activity lik e the
assertion o f defenses, it cast a shadow over a ll contested transactions and
threatened to discourage contracting by raising generally the cost o f prom ­
ise breaking.58
Compare the Dodge proposal. B ad faith denial o f the existence o f the
contract and bad fa ith assertion o f a defense are both narrow er, less amor­

52. Id. at 699.

53. Id. at 633-34.
54. See, e.g., Rancourt, supra note 5. at 1426 (showing disagreement over whether it is grounded in
tort or the implied covenant o f good faith and fair dealing; whether it applied just to denial o f contract
or bad faith denial o f liability, whether a special relationship like in the insurance cases was necessary).
55. Seaman’s Direct Buying Serv., Inc. v. Standard O il Co., 686 P.2d 1158,1167 (Cal. 1934). In
the decision overturning Seaman’s, the court in Freeman &Mills, Inc. highlighted this as a particularly
empty standard. Freeman & M ills, Inc. v. Belcher O il Co., 900 P.2d 669,680 (CaL 1995).
56. Responding to the question “under what circumstances w ill a breach o f the implied covenant o f
good faith and fair dealing in a commercial contract give rise to an action in tort,” the court in Sea­
man's asserted that it was unnecessary to decide this broad question. “It is sufficient to recognize that a
party to a contract may incur tort remedies when in addition to breaching the contract, it seeks to shield
itself from liability by denying in bad faith and without probable cause, that the contract exists." Sea­
man’s, 686 P.2d at 1167.
57. See Multiplex Ins. Agency, Inc. v. Cal. Life Ins. Co., 235 CaL Rptr. 3d 12 (C l App. 1937)
(concluding that bad faith denial o f liability under contract was enough for punitive damages to attach).
58. Plaintiffs pleading in good faith were able to state claims o f bad faith breach in a wide range o f
circumstances. See, e.g., Rogoff v. Gabowski, 246 Cal. Rptr. 185 (CL App. 1988); Commercial Cotton
Co. v. United CaL Bank, 209 CaL Rptr. 551 (CL App. 1985).
192 C O N N E C T IC U T L A W R E V IE W [Vol. 33:181

phous standards than Dodge’ s w illfu l, opportunistic breach.59 U nder the

Dodge form ulation no prom isor can guarantee going in that a subsequent
breach w ill not result in an award o f punitive damages.60 A ll traders in a ll
contracts must account fo r this possibility in pricing, and, given the d iffi­
culty o f p ricing the risk, some trades w ill not occur.61 In degrees o f sever­
a b ility, w e fin d the Dodge proposal at the extrem e lo w end, follo w ed by
the C alifo rn ia experim ent. A t the other end o f the spectrum, the insurance
cases and the special category cases exh ib it high severability.
W e have then a suggestion o f a continuum ; a non-severable and non­
viab le general standard o f abusive breach at one end (the Dodge proposal)
and a highly severable subcategory at the other (the insurance cases). They
are distinguished p rim arily by th eir triggers. Th e m istake o f the Dodge
proposal is to key on the manner o f breach— e.g., w illfu l, opportunistic.62
T h e entire universe o f contract breaches is exposed to that designation 63
T h e proposal injects collateral risk into the entire pool. On the other hand,
viab le rules w ill respect severability by keying on the characteristics o f the
transaction ( e . g . , insurance, m arriage and public u tility contracts) thus f il­

59. S ee Dodge, supra note 13, at 652.

60. A t a more general level, this criticism is a fam iliar one. Considering the elusive question of
what comprises a violation o f the implied covenant o f good faith, Professor Summers argued that the
concept was so broad that it was nearly empty o f substance. To give it meaning Summers created
subcategories o f bad conduct that were “excluded” from the range o f good faith performance and fur­
ther defined those categories with descriptive anecdotes. See Robert S. Summers, The G eneral Duty o f
G ood Faith—Its Recognition And C oncepntalization, 67 CORNELL L . REV. 810,818 (1982).
Summers’ attempt to give substance to the idea o f good faith responds to the same problem that
afflicts the Dodge proposal. Like the implied covenant o f good faith, a general standard punishing
w illful, opportunistic breach is hopelessly broad and generates tremendous uncertainty about what
conduct w ill run afoul o f the standard. See also Diamond & Foss, supra note 1, at 590-91.
61. Dodge acknowledges this criticism at the end o f his article and offers a short response. His
rule, he contends, would not inject significant new risk or instability into trades because the promisor
can negotiate for a release from the promisee. Dodge, supra note 13, at 697. However this is two steps
too late. The value o f a predictable exit price under current doctrine is that it drives promisor’s think­
ing before he commits to the deal. It drives his pricing and ultimate willingness to make a promise at
all. The possibility o f negotiating a release after the breach has occurred does nothing to affect this ex
ante calculation. And it would be an odd transaction indeed that incorporated an agreement ex ante that
set damages for promisor’s w illful, opportunistic breach. This is clear when we consider the range o f
language we have used to characterize this type o f breach. See, e.g ., Dold v. Outrigger Hotel, 501 P.2d
368, 372 (Haw. 1972) (allowing the award o f punitive damages “where the contract is breached in a
wanton or reckless manner as to result in a tortious injury”). See a lso Schwartz, supra note 13, at 369.
62. The California experiment made essentially the same mistake. See supra text accompanying
notes 15-44.
63. Traders must not be subjected to enhanced damages without warning. Within the category of
abusive activity, elevated risk is what we seek. But in the general pool our goal must be to maintain the
incentives that traditional contract damages provide. To do this we must give notice and then sever the
category deemed abusive from the general pool. A general common law standard o f abusive breach
prevents this. It invites judges to evolve the standard to include new categories o f activity. Nearly any
trade might provide the case through which a judge applies the general theory to create another cate­
gory o f elevated risk. This possibility injects new risk into the entire pool.
2000] R E U E F FOR A B U SIV E BRE A C H O F C O N TRA C T 193

tering out collateral risk.64

W ith the insurance cases and the Dodge proposal as extremes on a
continuum , another question appears. Is it ju s t a general standard o f abu­
sive breach— one that overhangs every trade— that is scuttled by its failure
to respect severability? Th e original rendition o f the C alifo rn ia experim ent
(w hich w e can fa irly position several steps inside the Dodge proposal) sug­
gests that there are other possibilities and problem s between the extremes.
Consider now a m id-range o f cases where w e key on the manner o f the
breach, not generally, but w ith specificity. Is it possible to sever and pun­
ish particular abusive practices and s till avoid collateral ris k in the general
pool? Th e next P art takes up this question and the problem s it presents.

V I. T h e I m p o r t a n c e of P ro cess at M i d -R a n g e

Im agine a m id-range o f cases consisting o f a peculiar, abusive rendi­

tio n o f a common and typ ically im m aterial type o f breach. H ere, w e can­
not key on transaction type because the conduct is pervasive. Instead, w e
are forced to focus on the manner o f the breach— the very focus that w e
concluded defeats severability in the case o f a general standard o f abusive
breach. B u t these m id-range cases present a d ifferent opportunity. W e
w ould not characterize the violation w ith broad m anner-of-breach designa­
tions lik e w illfu l, opportunistic or bad faith breach. R ather w e w ould key
on peculiar, e xp lic itly designated acts. M ig h t this allo w us to create here
the same type o f e x a n t e signal o f elevated risk that w e achieve by keying
on transaction type? T h e challenge is to create and seal o ff the category
through a sufficiently narrow and detailed description o f prohibited activ­
ity . T ru e, any trader m ight trip this type o f standard. Thus, it poses a
greater danger o f collateral risk than rules keying on transaction type.65
B u t i f the vio latio n is narrow ly and precisely defined, it is possible that w e
s till can signal the risk e x a n t e sufficiently to create only a d e m i n i m u s
level o f collateral ris k in the general pool.
H ow ever, achieving this precision poses a new problem : the uncer­
tainty inherent in the evolution o f a common law ru le prevents the precise

64. The California experiment was problematic because, at best, it hinged on the vague concept o f
bad faith. Guided merely by a m ildly embellished rendition o f bad faith, courts were not guided to
protect any general categories o f breach or any particular transactions.
65. The range o f risk is obviously broader here. For example, for a rule keying on transaction type,
traders w ill be able to determine quite easily that they are not engaged in a contract for provision o f
insurance protection. That single, easy determination is suflicient to insulate them from the elevated
risk attached to the insurance cases. However, under a standard punishing late-pay schemes, as defined
infra in text accompanying notes 66-67, traders are in danger o f making decisions throughout the trans­
action that w ill trigger exposure to punitive damages. A t mid-range, traders still can achieve a good
degree o f certainty that their conduct w ill not be deemed abusive, but they must be aware throughout
the life-span o f the deal that their conduct might trigger a claim for extracompensatory damages.
194 C O N N E C T IC U T L A W R E V IE W [Vol. 33:181

designation o f prohibited conduct that is essential in these m id-range cases

and becomes a significant source o f collateral risk. For these m id-range
cases, the legislative process offers a better chance o f developing an ex­
tracom pensatory response to abusive breach w ith m in im al collateral risk.
F o r a concrete illustration, consider a pervasive form o f buyer’s breach
that I call the late-pay scheme. Professional m oney managers have raised
it to an art form . I t is sim ply the practice o f dragging out paym ent to
vendors beyond the agreed paym ent date, and to the boundaries o f vendors’
tolerance. In its abusive form , it is an intentional, continuing exploitation
o f the rem edial inefficiencies o f the legal system that enriches its most suc­
cessful practitioners b y m illio n s o f unbargained-for dollars per year.6 6
67 In
the European C om m unity, calls to “outlaw ” it appeared in the financial
press resulting in legislation aim ed at stopping it.68
T h e case against the scheme is straightforw ard.69 I t is an attack on the

66. Treasury managers call the practice “payment timing optimization.” The scheme is complex
enough that treasury management consultants do “brisk business showing companies how to increase
cash by stretching vendor payments.” Richard H . Gamble, Taking Advantage o f Suppliers, 7
T reasury & Ris k M g m t ., Jan.-Feb. 1997, at 33.
67. The following passage illustrates typical late-pay practices and the advantages they afford to
corporations.
The C hief Financial Officer o f a $500 m illion East Coast apparel manufacturer and
importer cut $250,000 from his company’s annual interest costs last year by helping himself
to an interest fiee loan. How did he do it? Simple. He essentially borrowed from his ven­
dors by delaying his company’s payment o f their bills. But because he was subtle in his ap­
proach and resisted stretching the b ill prying as much as he could, many o f those vendors
have never even felt his hand in their pockets.
More brazen in the approach is the treasurer o f a $700 m illion maker o f consumer
products who routinely makes vendors wait an additional 11 or 12 days for the money they
are owed. Several o f them have howled. But for this treasurer, it’s worth the noise because
the short-term cash reaped from the delaying tactics has allowed him to reduce some debt
carrying a 7% coupon. That savings on interest payments means an additional $1 million
annually flows to his company’s bottom line.
Id.
68. The Late Payment o f Commercial Debts (Interest) Act, 1998, c 20 (U .K .), 11 June 1998, gives
promisees a statutory right o f interest and requires large firms to report the level o f their late payments.
S ee also Barbara Roche, Why We Must Outlaw Late Payment, T im e s OF LONDON, June 22 1997,
available a t LEXIS , News Library, Ttimes File.
69. There are very few reported cases where plaintiffs have sought damages for the loss caused by
the late-pay scheme. W ith so little litigation over the issue, one might conclude that the market has
adjusted to the practice— that vendors generally adjust prices to account for the costs imposed by the
scheme.
But the idea that it is just factored in to the contract suggests a more level pitying field than has
been proved. For small to medium sized local vendors competing against one another for accounts o f
national corporations, it is false to suggest that the late-pay pattern o f breach is factored in to the con­
tract price. The British Late Payment Act explicitly recognizes the potential imbalance o f bargaining
power where the promisee is a small vendor. S ee a b o Janet Stites, Prospectus: Sm all Technology-
B ased Com panies Constantly F a ce the Problem o f Getting B ig Custom ers to Pay Their B llb on Time,
N .Y . T im es , OcL 5,1998, at C4.
Moreover, as a matter o f process, i f we accept that the market has factored in the pattern o f late-
payment, how are we to respond to litigated cases under the current rules? Are we to ignore the express
language governing payment terms in favor o f defendant’s arguments that im plicit in p lain tiffs sale
2000] R E L IE F FOR A B U SIV E B R E A C H O F C O N TRAC T 195

structure o f contract law . Th e schemer exploits the interface between con­

tract rules and contract remedies and is successful only because o f the costs
the legal system imposes on those seeking re lie f fo r broken promises.
In other contexts, w e have responded aggressively to thw art sim ilar
types o f abuse.70 Exam ples o f this are c iv il abuse o f process and R u le 11
sanctions under the Federal Rules o f C iv il Procedure. Those tools recog­
nize that the open door policy o f our legal system empowers p lain tiffs to
impose unjustified costs on defendants. T h ey restrain this pow er by raising
the threat that the abusive p la in tiff w ill be ^penalized fo r invoking the legal
process m erely to exp lo it its inherent costs.71
In the late-pay scheme, putative defendants exp lo it the same inherent
costs. T h e only real difference is the tim ing. T h e putative late-pay defen­
dant exploits the high costs o f ju stice before litig atio n begins. H e captures
unbargained-for benefits w ith the expectation that transaction costs w ill
m ake it in efficien t fo r anyone to stop him .
There, are three basic renditions o f the late pay scheme. T h e most
clearly abusive (class one) is the scenario where goods or services are ren­
dered in installm ents under a long term contract, and the buyer employs a
prem editated plan o r system72 w ith the e xp licit aim o f skim m ing unbar­
gained-for value fro m die deal. H ere the scheme is m anifesdy intentional
and continuing. T h e class one late-pay schemer aims to m aintain the rela­
tionship in the future even w h ile siphoning unearned value fro m it through

price was compensation for the intentional and continuing late-pay breach that plaintiff knew would
occur? There are too many things wrong with such a view. First, it pushes us to evaluate subjective
intention while ignoring the objective indicators o f intent Second, it suggests that we should counte­
nance abuse o f the remedial structure (exploitations o f its inefficiencies) so long as the targets o f that
exploitation (at least those with sufficient market power to protect themselves) have adapted. Finally, it
counsels against attempts to improve the doctrine.
The argument is also problematic where the target is an organization with layers o f relatively
transient decision-makers. It is harder here for the experiences and expectations o f the point person to
filter up reliably to everyone who w ill make a decision contingent on the observable promises that form
die contract
Moreover, it is just as likely that the targets o f the late-pay scheme realize plainly that there are
no cost effective legal responses available to them. To the degree that damages are viewed as merely
the total amount overdue factored by the pre-judgement rate o f interest, it is understandable that very
few putative plaintiffs have pursued formal legal action where the late-pay scheme is defendant's only
violation.
Again, one is tempted to say that the problem is a small one that gets just about the amount o f
attention from the legal system that it deserves. Indeed, we commonly accept that in the case o f sellers
o f services, and less so for goods, there w ill be a bin o f residual immaterial breach that is not cost
effective to pursue. But this is o f course an incomplete view o f the practice. When we incorporate the
total levels by which practitioners o f the scheme are unjustly enriched through the scheme, it cannot be
dismissed as inconsequential.
70. Our general interest in removing errors and inefficiencies from the structure is a traditional
motivation for legal change that prescribes an aggressive response here.
71. A perfect mechanism would screen out warrantless cases at the outset. Unless we sacrifice the
ideal o f equal access to law, the open door structure seems to be the best available option.
72. See, e.g ., the mechanisms described in Gamble, supra note 66 and accompanying text
196 C O N N E C T IC U T L A W R E V IE W [Vol. 33:181

repetitive breaches. T h e class one scheme is the most easily identifiable

and easiest to police.
M o re com plicated, and m ore d iffic u lt to classify as abusive, is late
paym ent in the repetitive contract (class tw o ), i .e . late paym ent in separate,
s im ilar contracts fo r goods or services delivered over a period o f tim e.73
Th e most d iffic u lt to classify as abusive is the isolated late paym ent (class
three). Isolated late paym ent is a common occurrence that overhangs v ir­
tu a lly every transaction. A ru le that penalized it would vio late our prin ci­
p le aim by m aking the e x it price o f virtu a lly any trade highly uncertain.
A plausible goal is a ru le that penalizes class one late-pay scheme w ith
extracompensatory damages, and ensures that class tw o and class three late
payers are exem pt.74 T h is is viab le as a legislative enterprise. Pursued
through the common law it is m uch harder and maybe im possible.
T h e m ain concern is notice, and it breaks in to tw o parts. I f awarded
through the common la w , extracompensatory re lie f w ill surprise the first
defendant. H e w ill not have anticipated it and w ill not have priced it into
the deal.75 This is a general risk posed by the fact that common law judges
have the pow er to recognize new causes o f action. B u t given the general
bar on extracompensatory damages fo r breach o f contract, this risk is d e
m in im u s ? 6
Th e im portant notice problem occurs after the first case, in the form o f
uncertainty about how fa r the new cause o f action w ill extend. Th e first
com mon law case warns traders that the e x it price o f trades is now inde­
term inately higher. They are uncertain about whether this risk actually w ill
b ite or how costly it w ill be. T h e fact that the rule w ill evolve through the
common law process is i t s e l f a significant source o f risk. O n any day in
any court, a common law ru le penalizing class one late paym ent m ight
expand to other classes or to entirely new categories o f aggressive com­
m ercial activity.77 T h at possibility o f extracompensatory damages attach­

73. One example is the provision o f temporary staff at buyer’s request.

74. W e might decide to lump class two late-pays with either class one or class three. 1 think it is
cleaner to place class two and class three together.
75. This is a general risk all traders face from the general knowledge that the law might change.
76. Paradoxically i f courts acknowledge the importance o f severability, it empowers and perhaps
encourages them to create sub-categories o f elevated risk. This itself, may be enough to increase trad­
ers’ perception o f the risk o f contracting. It is easy to imagine that moving from a regime where the
generally controlling rule is compensatory damages only, to one where extracompensatory damages are
authorized in discrete severable subcategories (even where that discretion is exercised rarely) w ill
change traders’ perceptions o f the cost o f promise-making to at least a small degree. However, that is
not much different from the current state o f affairs. Traders in most jurisdictions operated in the
shadow o f precedent that confirms the possibility o f punitive damages for breach o f contract in amor­
phously defined egregious cases where the court also can push the breach into the category o f tort. See
R esta tem en t (Second ) of Contracts § 355 (1981) (stating that punitive damages are recoverable
if “the conduct constituting the breach is also a tort for which punitive damages are recoverable”).
77. The evolution o f the California experiment is a good example o f this. Consider the message
sent by the conflicting opinions in DuBarry International, Inc. v. Southwest Forest Industries, Inc., 282
2000] R E L IE F F O R A B U SIV E B R E A C H O F C O N T R A C T 197

ing unpredictably to done deals w ithout notice injects unacceptable risk

into the entire pool o f transactions; it prevents us from awarding ex­
tracompensatory damages in harm ony w ife fee prim e tenant o f a predict­
able e x it price.*78
I f w e are to key on fee m anner o f fee breach w ife p articularity suffi­
cient to sever the category from the general pool, w e m ust ensure that the
boundaries hold. T o avoid collateral risk, fee offense m ust be defined w ife
certainty, relia b ly into fee future, w ith adequate notice before the ru le ex­
pands to capture new circumstances.79
Legislation is m ore lik e ly to achieve this fo r reasons inherent in the
process.80 T h e inherent flu id ity o f common law rules invites porous,
evolving boundaries and feus expanding risk— precisely the collateral risk
that w e must avoid. A fte r the firs t common la w case aw arding ex-
tracom pensatoiy damages, given fee uncertainty o f its life-span and vig or,
direct and collateral risk to traders engaged in sim ilar a ctivity rises dra­
m atically; fee a b ility o f traders to assess fee boundaries o f this new ris k is
m in im al u n til the doctrine matures.
Legislation can produce im m ediately som ething m uch closer to a m a­
ture structure fo r levying extracompensatory damages in discrete cases.81
In addition to pre-im plem entation notice that is v ita l to severability, the
legislative process can fix m ore solid boundaries around fee proscribed

Cal. Rptr. 181 (C t App. 1991) (determining that Seam an's could only extend to bad faith denial o f a
contract’s existence, not a party’s bad faith denial o f liability under a contract) end M ultiplex Insur.
Agency, Inc. v. CaL l i f e Insur. C o., 235 Cal. Rptr. 12 (C t App. 1937) (concluding that bad faith denial
o f liability under a contract was enough for punitive damages to attach).
78. Witness the California experiment, which initially keyed on fairly discrete activity—bad faith
denial o f the existence o f the contract— and expanded to include activity—bad faith assertion o f de­
fenses— that introduced uncertainty to a much larger category o f activity. S ee supra text accompanying
notes 14-34.
79. One arguable drawback o f course is that traders w ill be able to come right to the line safely,but
this is the cost o f minimizing collateral risk. A common law rule on the other hand, would not permit
such a close approach. The uncertainty o f what the nextjudge w ill do with a common law rule cautions
the risks averse to give the activity wide berth; that is it over-deters, which also means it leaks collateral
risk.
80. The legislative process might not produce better substantive rules. Indeed they might be worse.
But this merely underscores the fact that at middle-degrees o f severability (Le^ rules keying on peculiar
styles o f breach) process is paramount For example, legislators might be more prone than judges to
make war on “abusive” commercial practices. They might decide to punish a variety o f discrete prac­
tices that they deem abusive breach. So long as they fashion adequate initial boundaries, the pre­
enforcement notice and non-retroactivity o f the process w ill insulate the general pool o f transactions
from risk.
81. In countless categories o f cases a clear legislative statement w ill cause courts to defer to legis­
lative schemes as mature structures h a t permit little room for interpretation. See, for example, the
variety o f cases stemming from Chevron U.SA., Inc. v. N atural R esources D efense Council, In c., 467
U.S. 837 (1984). S ee generally, NORMAN Singer , Sutherland Sta tu to r y CONSTRUCTION (5th ed.
1994). Whether this is purely a function o f judicial philosophy or not, the point remains that the legis­
lative process allows development o f a mature regulatory scheme more quickly than does the common
law process.
198 C O N N E C T IC U T L A W R E V I E W [Vol. 33:181

conduct. There w ill be uncertainty, but dram atically less than would ap­
pear through the slow and unpredictable process o f common law
evolution.82 T h e long lead tim e, pre-enforcem ent notice and non­
retroactivity inherent in legislative rules allow s traders to rely upon an es­
tablished boundary between class one and class tw o/three late pays, w ith ­
out fear that theirs w ill be the case where a common law judge breaks the
seal.83
W e fin d then at m id-range that severability is degraded by the common
law process. Th e capacity o f any category to expand through the common
la w process generates the collateral ris k that w e must avoid. I f w e choose
to aw ard extracompensatory damages at m id-range, w e also must choose
the legislative mechanism.

v n . C o n c l u s io n
A n y response to bad faith breach must occur in harmony w ith the core
com m itm ent to a predictable e x it price. Th e lesson fro m the failure o f the
C alifo rn ia experim ent and the v ita lity o f the insurance cases, is that w e can
pitch an extracompensatory response to abusive breach in harmony w ith
this core com m itm ent so long as that response is lim ited to discrete and
severable categories o f breach.
Extracom pensatory damages rules that key on the type o f transaction
rather than the m anner o f the breach exh ib it the highest degree o f sever­
a b ility and thus the highest degree o f v ia b ility . Broad standards that key
on the m anner o f the breach, generally described, exhibit a low degree o f
severability and the lowest degree o f v ia b ility . Rules that key e xp licitly
and precisely defined types o f breach achieve a moderate degree o f sever­
a b ility and are viab le when pursued through legislation.
W e w ill lik e ly continue to debate what constitutes abusive breach as a
norm ative m atter. B ut in practical terms, w e cannot develop viable re­
sponses to categories o f abusive breach w ithout considering and respecting
severability.

82. A rough example o f the virtues o f legislative process are usury laws. The relatively bright line
boundaries governing usury permit general lending transactions to continue unimpaired by the threat o f
elevated damages. See, e.g .. In re Estate o f Dane, 390 N.Y.S.2d 249 (App. D iv. 1976) (holding the
usurious lender forfeits both interest and principle as punishment for lending above the usury rate).
Contrast the tangled web o f common law rules that attach additional risk to contracts with minors. See
discussion in Pettit v. Liston, 191 P. 660 (Or. 1920).
83. This o f course assumes an adequate job o f legislation to design the boundaries and minimize
ambiguity. It also assumes a strong legislative message about the systemic importance o f the boundary,
sufficient to discourage activist judges.