Judiciary
Homepage
Mr. Chairman and Members of the Subcommittee:
My name is Thomas D. Rowe, Jr., and I am a Professor of Law at Duke University. The subjects I teach include Civil Procedure, Complex Civil Litigation, Federal Courts, Judicial Remedies, and Constitutional Law. I have done research and published articles on various forms of attorney fee shifting rules; I have also worked for groups that have considered these issues, such as the Federal Courts Study Committee and the American Law Institute's study of Enterprise Responsibility for Personal Injury. I now serve as an appointed member of the United States Judicial Conference's Advisory Committee on Civil Rules, which considers proposed changes in the Federal Rules of Civil Procedure; in appearing here today, however, I speak only as an individual.
In this statement I cover four main areas: 1) briefly, the
issue of constitutionality of a loser-pays attorney fee
liability rule targeted on diversity cases; 2) some general
points about the desirability of loser-pays and other attorney
fee liability rules, including their fairness, the incentives
they seem likely to create for the pursuit and settlement of
claims, and experience with some uses of the loser-pays rule
in American jurisdictions; 3) specific observations on H.R.
lO's use of the loser-pays rule for
Rowe statement--loser-pays fee shifting February 6,
1995
cases filed in the federal courts' diversity jurisdiction; and
4) possible alternative approaches or adjustments to the H.R.
10 version of the loser-pays rule.
I. Constitutionality
In my judgment, in adopting a loser-pays rule applicable only to cases filed in the diversity jurisdiction, Congress would be within its powers to legislate concerning jurisdiction and procedure in the federal courts. The Supreme Court's leading statement on the scope of Congressional power in this area, as with other federal legislative powers, speaks in broad terms:
[T]he constitutional provision for a federal court system (augmented by the Necessary and Proper Clause) carries with it congressional power to make rules governing the practice and procedure in those courts, which in turn includes a power to regulate matters which, though falling within the uncertain area between substance and procedure, are rationally capable of classification as either.
Hanna v. Plumer, 380 U.S. 460, 472 (1965). The Court has repeatedly and recently reaffirmed this broad "arguably procedural" definition of the scope of Congress's power over federal court practice and procedure, citing Hanna's formulation about matters that are "rationally capable of classification as either" substance or procedure in decisions upholding statutes and rules governing proceedings in federal courts. See, e.q., Budinich v. Becton Dickinson & Co., 486 U.S. 196, 198 (1988); Burlington Northern R.R. v. Woods, 480 U.S. 1, 5 (1987).
Rowe statement--loser-pays fee shifting February 6,
1995
Whatever one thinks of the desirability of loser-pays fee
shifting (and I have substantial doubts, which I will explain)
it does not seem to fall beyond the "uncertain area between
substance and procedure," and it is "rationally capable of
classification" as procedural--especially considering the
proponents' concern for frivolous litigation. I would expect
constitutional challenges, based on arguments that Congress is
stepping over the line into pure substance in cases otherwise
governed by state law, if H.R. lO's loser-pays rule were
adopted. Those arguments do not strike me as frivolous, and
indeed they deserve some weight as policy reasons against the
H.R. 10 proposal; but I would give long odds that such
arguments would not persuade anything close to a majority of
the Supreme Court that the measure was unconstitutional.
Loser-pays attorney fee liability is at least arguably
procedural, and given the breadth of its authority Congress
need not adopt identical procedures to govern both diversity
actions and other cases within federal court jurisdiction.
II. General Arguments For and Against the Loser-Pays Rule
Most of the industrialized democracies follow some form of
loser-pays rule for attorney fees in civil litigation,
although the loser's liability is often only for part
(sometimes a fairly small part) of the winner's attorney fees.
The two main exceptions are the United States
and--interestingly, given its reputation for non
litigiousness--Japan. In both the European and Commonwealth
Rowe statement--loser-pays fee shifting February 6,
1995
nations with the loser-pays rule, and here with the American
rule that each side is responsible for its own attorney fees
win or lose, the prevailing rule mostly seems to have taken
root long ago, without a great deal of thought being given
until recently to policy questions of what attorney fee
liability rule should govern and why. It thus makes sense to
ask what main kinds of rationales should guide the choice of
such a rule.
Without exhausting all the plausible types of arguments, let me suggest that three kinds deserve to be prominent: justice, incentives, and administrability. Many people start with a gut feeling that loser-pays is fair; the winner should not suffer for having had to spend to establish the rightness of a claim or defense. Moreover, when plaintiffs win damage awards, recovering fees--rather than having to pay their lawyers out of the recovery --brings them closer to getting the full compensation that the law means to give. But even losers often have fairness arguments against having to pay both sides' fees, because many cases are close enough that it is reasonable for both sides to take them to trial. With such variable influences as jury sympathy and lawyers' skills, we cannot assume that everyone should have been able to foresee the result and the eventual loser should have known better. As Justice Cardozo once said, "I have seen enough of the judicial process to know its imperfections. I would not lay too heavy a burden on the unsuccessful litigant."
Rowe statement--loser-pays fee shifting February 6,
1995
Another way to put it is that loser-pays, at least in pure
form, amounts to strict rather than fault liability: whether
or not we think you were unreasonable about insisting on your
right to trial, if you lose you suffer not only the judgment
on the merits against you but must pay what it cost both sides
to get the verdict. The loser-pays rule works as an especially
harsh form of strict liability when costs are likely to be
highest and the loser's conduct most reasonable--in closely
contested cases. (The sponsors of H.R. 10 share a suspicion of
strict liability in another context, as section 103 of the
bill would mostly limit the product liability of product
sellers to cases involving fault or express warranty.)
Further, as ethical codes now stand and as H.R. 10 is now
drafted, loser-pays fee liability would fall on the client and
not on the lawyer. Some clients such as poor plaintiffs may be
judgment-proof, making the loser-pays rule ineffective. With
middle class claimants who have plausible claims that are
short of sure things, the rule may be all too effective, as
such people may be highly reluctant to risk even a fairly
small chance of not just going away empty-handed but ending up
in the hole. Thus loser pays fee liability may be easier to
defend in some contexts where both sides will usually be well
financed, such as commercial contract litigation; Arizona has
the loser-pays rule for such cases.
To turn from fairness to incentive considerations, it helps to
distinguish between the likely effects of attorney fee
liability
Rowe statement--loser-pays fee shifting February 6,
1995
rules on, first, whether people pursue claims in the first
place and, second, whether and how cases settle or go to trial
once a claim is pursued. Empirical research on the actual
effects of fee liability rules is difficult and scarce, but
legal academics and economists have enthusiastically churned
out a great deal of theory. To give a brief summary, some of
the main effects on pursuit of claims expected from loser-pays
fee shifting are: 1) discouragement of "nuisance"
claims--although it is not clear how widespread the
nuisance-claim problem is, because contingent-fee plaintiffs'
lawyers usually have a strong incentive not to take on likely
losers; 2) encouragement of strong small claims, which are not
worth pursuing under the American rule--although this effect
of loser-pays seems of limited relevance under H.R. 10,
because it applies only to cases involving claims for over
$50,000 to begin with; and 3) discouragement of reasonably
strong but not clearly winning claims of people with modest
means, who are likely to be "risk averse" and strongly
deterred by a significant threat of down-side liability.
The likely effects of attorney fee liability rules on parties'
decisions to settle or go to trial are hotly debated. Perhaps
the point that deserves most emphasis is that it is essential
to consider the impact of a possible fee shift on both sides'
settlement incentives, which can make predictions complex.
Just because a defendant with a weak defense and facing fee
liability will likely be eager to settle, and for more than
under the American
Rowe statement--loser-pays fee shifting February 6,
1995
rule, does not mean that settlement is automatically more
likely. The probability of winning not just damages but fees
means that the plaintiff with a strong claim can hold out for
more, and the parties may be at least as far apart as under
the American rule. A useful if rough generalization may be
that loser-pays may strengthen the bargaining position of
those who feel able to bargain hard, either because they are
well financed or because their case is very strong. That
effect can make settlement less likely in a case involving
well-financed parties who disagree about the probable outcome
already, or more likely if the threat of an adverse fee shift
plays on the risk aversion by weakening the bargaining
position of a middle-class plaintiff with less than a sure
winner.
The final type of factor I mentioned for choosing a fee award
rule was administrability. Whatever its defects, the American
rule has the virtue of saving courts from having to spend time
ruling on fee award amounts. Applying the common "lodestar"
formula--a reasonable hourly rate times the hours reasonably
expended on successful claims--often leads to complex and
costly second rounds well after the merits are concluded.
Better ways may exist, but they have yet to find widespread
acceptance here. Percentages of recovery are simpler and
correspond to present contingent fee practice; yet they could
be excessive with very large recoveries, inadequate in cases
of obstinate resistance to a small
Rowe statement--loser-pays fee shifting February 6,
1995
claim, and beside the point for cases involving non monetary
relief or defense victories.
Issues of fee award administration bring me to brief mention of two recent state experiences with loser-pays fee shifting. Florida had it for medical malpractice cases from 1980 to 1985, at the behest of doctors who saw it as protection against frivolous claims. The medical organizations quickly came to support its repeal when many fee awards proved uncollectible--except for those against losing doctors, which were sometimes startlingly large. One law review article reported that the law was regarded as "ineffective to prevent suits or encourage settlements.lll A leading empirical study of the Florida experience concluded that it had probably had some net deterrent effect on claims, but had also led to fewer settlements and more expensive trials in cases that did get pursued. 2
Alaska has long been the sole general exception to the Ameri- can rule, with fee liability established by statute and award schedules written into state court rules. The state has kept its loser-pays system, but with considerable grumbling and a recent major revision that reduced the percentages of winners' fees for which losers can be held liable. A survey that preceded the rule
1 F. Townsend Hawkes, The Second Reformation: Florida's Medical Malpractice Law, 13 Fla. St. U. L. Rev. 747, 766 n.94 (1985).
2See Edward A. Snyder & James W. Hughes, The English Rule for Allocating Leqal Costs: Evidence Confronts Theory, 6 J.L., Econ., Org. 345 (1990).
Rowe statement--loser-pays fee shifting February 6,
1995
revision found wide divergences of opinion among lawyers not
only about the desirability of the basic rule but over its
impact on such key points as pursuit of medium-strength
claims.3
III. Special Features of H.R. 10
Not all loser-pays rules are identical. The foregoing summary of major issues concerning attorney fee liability has spoken in general terms, saving for now any consideration of the particulars of H.R. 10's loser-pays proposal for diversity cases. Three features or effects of the provision seem worth special emphasis: 1) the limit on fee liability to the amount of one's own fees, with provision for reckoning a proxy amount when losing plaintiffs had contingent fee arrangements with their lawyers and hence incurred no fee; 2) its applicability to cases filed in the diversity jurisdiction, but not to cases filed in state court and removed by defendants on the basis of diversity; and 3) its phrasing that would probably keep Rule 68 offers of settlement from affecting fee as opposed to cost liability. This last feature would make entitlement to a fee award turn on the determination of liability on the merits, leaving the amount of damages recovered irrelevant except to the court's discretionary reduction or denial of a fee award.
3See Kevin M. Kordziel, Rule 82 Revisited: Attorney Fee Shiftinq in Alaska, 10 Alaska L. Rev. 429 (1993).
Rowe statement--loser-pays fee shifting February 6,
1995
First, subsection 2 of the loser-pays provision, proposed 28
U.S.C. 1332(e)(2), would limit the loser's fee liability to
the amount of the loser's own fees. This cap seems a
reasonable accommodation to avoid what could be very harsh
effects if, say, a plaintiff of modest means opposed a large
company that invested heavily in expensive defense lawyers.
The limit does, of course, dilute the deterrent impact of the
loser-pays rule; in effect, for amounts over the plaintiff's
likely fee, the cap reinstates the American rule. It thus
remains possible for a plaintiff to bargain for a nuisance
settlement by pointing to the unrecoverable fees a defendant
may have to incur. This result may be the least of the
available evils, but it does reduce just how much can probably
be claimed for the bill's likely reduction of nuisance liti-
gation.
Second, the core provision in subsection 1 makes the loser pays rule applicable when a federal district court "exercises jurisdiction in a civil action commenced under this section"--28 U.S.C. 1332, the statute establishing jurisdiction for state citizen diversity, alienage, and foreign-state-as-plaintiff cases.
(The Subcommittee might wish to cut out the foreign-state-as-
plaintiff cases brought under 1332(a)(4), because the present
bill seems to create an arbitrary distinction between those
cases and foreign-state-as-defendant actions under 28 U.S.C.
1330.) From the text of subsection 1, it appears that the
loser-pays rule is to govern when plaintiffs file diversity
cases, but not when
Rowe statement--loser-pays fee shifting February 6,
1995
they sue in the courts of a state that follows the American
rule-whether or not the defendant removes to federal court.
This approach does make some sense and may be the best
available; making plaintiffs strictly liable for defendants'
fees when they filed state law cases in the courts of a state
that follows the American rule, but the defendant removed to
federal court, could be extremely harsh.
This difference in fee rule depending on filing in state or federal court creates a unique choice for plaintiffs' counsel. Most loser-pays rules are hard to escape; an English plaintiff cannot avoid the rule by filing in Wales. H.R. 10's loser-pays rule for diversity cases, though, would let plaintiffs pick the applicable fee liability rule, by choosing whether to file a case involving diverse parties and a claim over $50,000 in federal or state court. The proposal in H.R. 10 seems to proceed from a concern for excess mainly on the plaintiffs' side; it implements a plank in the Contract with America that refers to "excessive legal claims" and "frivolous lawsuits," while there are such things as frivolous defenses as well--which the Contract does not mention. Yet the effect seems to be a plaintiff's dream; as a plaintiff's lawyer, I would feel about this loser-pays rule like Brer Rabbit about the briar patch.
The calculations for plaintiffs and their lawyers seem
straightforward: File very strong claims in federal court and
lock in the prcbability of a favorable fee shift. File
"iffier"
Rowe statement--loser-pays fee shifting February 6,
1995
but still plausible claims in state court and be protected
against the chance of having to pay the defendant's fees, even
if the defendant removes to federal court. The result would be
not to "stop . . . frivolous lawsuits" but, when they do get
brought and pursued, to divert them to state court--with no
guarantee they would not end back up in federal court. To the
extent that cases filed in state court did not get removed,
the federal courts would probably see a diversity docket
somewhat richer in claims that were strong on liability. That
would not necessarily mean, though, a higher settlement rate
because plaintiffs entitled to a fee shift could hold out for
good settlements--and also because major disagreement could
still exist on amounts of damages.
This choose-your-rule-when-you-choose-your-court feature has some side effects. A usually fairly minor although unfortunate one could be races to the courthouse: if both sides to a prospective suit have claims against each other and see their chances of success differently, they could scramble for the advantage of picking the fee rule by filing first. A second implication is that H.R. 10's loser-pays rule would not afford a clean experiment in the workings and effects of such rules, because the selection possibilities would skew the samples of cases that were and were not under the rule. An experiment it would be, but with a form of loser-pays so unusual as to make it hard to derive answers to some of the most important questions about such rules.
Rowe statement--loser-pays fee shifting February 6,
1995
Third, the bill as drafted speaks of an award of "an attor-
ney's fee," without using the talismanic words "as part of the
costs" which appear in some other federal attorney fee award
statutes such as the Civil Rights Attorney's Fees Awards Act,
42 U.S.C. 1988. Under Marek v. Chesny, 473 U.S. 1, 7-11
(1985), this omission makes it appear that Federal Rule of
Civil Procedure 68 on offers of judgment will not affect fee
liability under the proposed 28 U.S.C. 1332(e) as drafted. A
defendant's Rule 68 offers would thus not stop the clock on
the fee entitlement of a plaintiff who rejected but did not
improve on the defense's offer, although the court might
exercise its discretion under subsection 3 to reduce or deny
a fee award in such a case.
This inapplicability of Rule 68 to H.R. 10's loser-pays fee shifting would mean that fee liability under subsection 1 would remain tied to the result on liability on the merits, unaffected by the outcome on damages. The distinction may sound technical, but the practical effects could be worrisome. Some cases are strong on both liability and damages, while others are weak on one or the other or both. The filtering incentives created by H.R. 10, with plaintiffs at least sure of no fee shift against them if they win on liability, would attract plaintiffs to file in federal court cases that are strong on liability--whether they are strong or weak on damages.
No one is likely to object if the rule works as a magnet for
cases that are strong on liability and damages, but the
paradigm
Rowe statement--loser-pays fee Slitting February 6,
1995
of a case that is strong on liability and weak on damages is
a whiplash claim. Plaintiff's counsel with a rear-end
collision case that is a sure thing on liability, while
questionable on substantial damages but able to support a good
faith initial claim for over $50,000, could see nothing to
lose as far as fee liability was concerned by filing in
federal court. Neither state court with the American rule, nor
federal court with loser-pays, would threaten an adverse shift
when liability is certain; but federal court would offer the
chance of a favorable shift that is unavailable in the state
forum.
IV. Alternatives
A principal problem with loser-pays fee shifting as a response to perceived problems of "excessive legal claims, frivolous lawsuits, and overzealous lawyers" is that it threatens the frivolous and the non frivolous alike. In most forms it also threatens the clients, not the lawyers. We do have problems with frivolous claims--and frivolous defenses--and we need to deal with them. My first preference is for approaches that aim more closely at t e perceived abuses and abusers. That is precisely what the framers of the Federal Rules of Civil Procedure have been trying to do with the forms of Rule 11 adopted in 1983 and 1993.
I was not a member of the Advisory Committee on Civil Rules
when it proposed what became the 1993 amendments, with their
controversial limits on the sanctions process as it had stood
under
Rowe statement--loser-pays fee shifting February 6,
1995
the 1993 version. All I will say now is that the 1993 changes
were the product of especially careful consideration, that
they aimed at cutting back on severe "satellite litigation"
problems that had arisen under the 1983 version, and that the
amendments drew perhaps surprisingly broad if not unanimous
support from the bar. They still permit sanctions against
attorneys and law firms, which H.R. 10's loser-pays provision
does not. The 1993 changes apply to all federal court civil
litigation, which the diversity proposal in H.R. 10 does not.
I urge you to give these important and still new amendments a
chance to work before changing them or adopting other and more
drastic measures.
If some step beyond letting new Rule 11 prove its worth (or lack thereof) is felt necessary, rather than loser-pays you could consider an enhanced form of present Rule 68 on offers of judgment. The rule now lets defendants make formal settlement offers; if the plaintiff rejects the offer and does not win a more favorable judgment, the rule makes the plaintiff liable for the defendant's post-offer "costs"--which do not usually include attorney fees. Largely because of this limited effect, the rule has seen relatively little use. Thoughtful proposals, however, have been made to make it more effective, most notably in recent years by Judge William Schwarzer, Director of the Federal Judicial Center.4
4William W Schwarzer, Fee-Shiftinq Offers of Judqment--An Approach to Reducing the Cost of Litiqation, 76 Judicature 147 (1992).
Rowe statement--loser-pays fee shifting February 6,
1995
Briefly, the Schwarzer proposal would let either plaintiffs or
defendants make Rule 68 offers. Rejecting an offer and not
improving on it at trial would make the rejecting party
liable, subject to certain limits, for the offeror's
post-offer attorney fees. Thus a defendant facing a very weak
claim could make a low, early offer with teeth--take this
limited amount and go away, or you may be on the hook for some
of the fees I have to pay to fight you. The plaintiff with a
strong claim could make a parallel, effective threat to an
obstinate defendant. Such a rule could smoke out serious
settlement offers and might get some cases settled earlier,
although its net effects on settlement are complex and
uncertain. It would also get away from H.R. 10's problem of
tying fee liability to outcomes on liability on the merits,
shifting it instead to damaqe results, and would apply to most
federal civil litigation and not just originally filed
diversity cases.
I should not oversell the idea of an enhanced Rule 68. While I have been on the Advisory Committee on Civil Rules we have considered the Schwarzer proposal and seen many complexities, and have also thought it might be inappropriate for us to propose because of possible effects on substantive rights. But I do regard the rapiers of Rule 68, and amended Rule 11, as more promising avenues for dealing with litigation excesses than the broadsword of H.R. 10's loser-pays rule. I urge you to think seriously about these alternatives before you go ahead with the proposal before you.
Curriculum Vitae
Thomas D. Rowe, Jr. Home:
Duke University School of Law 712D Constitution Dr
.
Durham, N.C. 27708-0360 Durham, N.C. 27705-2801
919: 613-7099 919: 383-6775
Education
Undergraduate:
B.A. sum ma cum laude, Yale University, 1964 Major in political science and economics; "honors with exceptional distinction" in major
Graduate:
M.Phil. in general and comparative literature, Oxford University, 1967
Legal:
J.D. magna cum laude, Harvard Law School, 1970 Harvard Law Review, 1968-70; Supreme Court and Note Editor, 1969-70
Employment
1970-1971: Law clerk to Associate Justice Potter Stewart, Supreme Court of the United States
1971-1973: Assistant Counsel, U.S. Senate Subcommittee on Admin- istrative Practice and Procedure
1973-1975: Associate at Miller, Cassidy, Larroca & Lewin, Wash- ington, D.C. (civil and criminal litigation)
1975-1979: Associate Professor, Duke University School of Law
1979-date: Professor, Duke University School of Law
1979-1980: Visiting Professor, Georgetown University Law Center
1980-1981: Visiting Fellow, United States Department of Justice
1981-1984: Associate Dean for Research, Duke University School of Law
Fall 1985: Visiting Professor, University of Michigan Law School
Jan.-Aug. 1991: Attorney, Munger, Tones & Olson, Los Angeles
Fall 1991: Visiting Professor, University of Virginia Law School
Honors and Awards
Phi Beta Kappa; graduated first in Yale class
Rhodes Scholar, 1964-67
Duke [student] Bar Association Distinguished Teaching Award, 1985
Thomas D. Rowe, Jr
. Principal publications
Articles:
Abolishing Diversity Jurisdiction: Positive Side Effects and Potential for Further Reforms, 92 HARV. L. REV. 963-1012 (1979)
The Emerqinq Threshold Approach to State Action Determinations: Trying to Make Sense of Flaqq Brothers, Inc. v. Brooks, 69 GEORGETOWN LAW JOURNAL 745-71 (1981)
The Leqal Theory of Attorney Fee Shiftinq: A Critical Overview, 1982 DUKE LAW JOURNAL 651-80
Predictinq the Effects of Attorney Fee Shiftinq, 47:1 LAW AND CON- TEMPORARY PROBLEMS 139-71 (Winter 1984)
Beyond Diversity: Federal Multiparty, Multi forum Jurisdiction, 135 UNIVERSITY OF PENNSYLVANIA LAW REVIEW 7-58 (1986) (with Sibley) The Supreme Court on Attorney Fee Awards, 1985 and 1986 Terms: Economics, Ethics, and Ex Ante Analysis, 1 GEORGETOWN JOURNAL OF LEGAL ETHICS 621-39 (1988)
Empirical Research on Offers of Settlement: A Preliminary Report, 51:4 LAW AND CONTEMPORARY PROBLEMS 13-39 (Autumn 1988) (with Vidmar)
American Law Institute Study on Paths to a "Better Way": Litiqation, Alternatives, and Accommodation--Backqround Paper, 1989 DUKE LAW JOURNAL 824-902
Jurisdictional and Transfer Proposals for Complex Litiqation, 10 REVIEW OF LITIGATION 325-400 (1991)
No Final Victories: The Incompleteness of Equity's Triumph in Federal Public Law, 56:3 LAW AND CONTEMPORARY PROBLEMS 105-21 (Summer 1993)
Book:
CONSTITUTIONAL THEORY: ARGUMENTS AND PERSPECTIVES (with Gerhardt) (Michie Co. 1993)
Subiects taught
Civil Procedure Complex Civil Litigation
Constitutional Law Constitutional Theory
Federal Courts Remedies
Orqanizational affiliations
Member, American Law Institute, since 1978
Chair, Association of American Law Schools Section on Civil Pro- cedure, 1989
Member, Federal Advisory Committee on Civil Rules, 1993 Chair, Fourth Circuit Advisory Committee on Rules and Procedures,1994