Manhattan Institute for Policy Research.
Subscribe   Subscribe   MI on Facebook Find us on Twitter Find us on Instagram      

Civil Justice Report
No. 1  March 2000

The New American Rule: A First Amendment to the Client’s Bill of Rights

Richard W. Painter
Professor of Law at the University of Illinois

Richard W. Painter, Professor of Law at the University of Illinois, is the author of over a dozen law review articles on securities law and legal ethics, and is the coauthor with Judge John T. Noonan, Jr. of a legal ethics casebook. In 1998 he testified before the United States Senate and House of Representatives on securities class actions.1 

The President says that consumers of medical services are entitled to a “patient’s bill of rights.” We hear nothing from the President, and little from Congress, however, about consumers of legal services. Patients thus need protection from doctors’ excessive fees and poor services, but clients presumably do not need protection from lawyers. This reasoning is no surprise. The President is a lawyer as are many members of Congress. Very few politicians are doctors.

Lawyers’ clients are supposed to be protected by state ethics codes, but these codes do not adequately protect clients from excessive fees, particularly when lawyers work for contingent fees.2  A lawyer’s fee “must be reasonable” (Model Rule 1.5),3  although trial judges almost never initiate review of contingent fees in cases before them,4  and clients rarely challenge a fee as excessive.5  It does not matter whether a lawsuit is an easy win and for a large amount of money. The lawyer is almost always allowed to charge one-third or more, and plaintiffs’ lawyers usually do.6 

At least the client should get what is paid for: a lawyer devoted entirely to the client’s interests and not to his own. Unfortunately, contingent-fee lawyers sometimes have different risk preferences from those of their clients and can profit from litigation decisions that disadvantage clients.7  Lawyers’ financial interests at times encourage them to take risks that clients would avoid, and at other times to settle too soon so they can move on to the next case. It is true that hourly billing creates conflicts of its own8  and that clients are protected from disloyal lawyering by state ethics codes (the same ethics codes that protect clients from excessive fees). Lawyers thus must provide “competent” representation (Model Rule 1.1)9 and must avoid conflicts between their own interests and those of their clients (Model Rule 1.7(b)).10  Lawyers who deploy litigation strategies and recommend settlements that benefit themselves rather than their clients, however, are usually only discovered and punished in the worst of cases.

Not everyone is satisfied with this state of affairs. Professor Lester Brickman accused contingent-fee lawyers of widespread price gouging and conflicts of interest in an article in the UCLA Law Review a decade ago.11  While other scholars have responded with arguments that defend the practices of the contingent-fee bar,12  this author finds it difficult to discern much competition in a market that usually assigns the same risk premium (33%) to a plaintiff’s case, no matter how large the case is and no matter how likely the client is to win. Except for the relatively few cases where the contingent fee drops to around 25% or rises to 40% (which is common if an appeal is involved),13  the only variable seems to be which lawyer is willing to accept which case. Plaintiffs in turn are often psychologically risk averse14  and thus willing to bargain away cheaply their potential upside in order to assure themselves that they will not lose a case entirely by hiring a (supposedly inferior) lawyer who charges a more modest fee. Add to this picture the informational asymmetry between experienced trial lawyers (who know a good case when they see one) and inexperienced plaintiffs (who may not know a skilled and honest lawyer when they see one),15  as well as the burden that comparison shopping imposes on tort victims recovering from an injury or death in the family, and it becomes clear that the lawyer has a decided advantage in whatever competition the market for legal services has to offer.

Proposals for contingent-fee reform have attracted nationwide attention in the past few years. Many reform measures could be implemented at the state level, either by incorporation into state ethics codes or rules of civil procedure, although federal legislation may also be required. Debate over who should implement reform, however, begs a more difficult question: what the reform should be in the first place.

This Civil Justice Report reviews a number of the proposals that have been made so far and discusses the comparative strengths of a new proposal made by Jim Wootton, President of the U.S. Chamber of Commerce Institute for Legal Reform: a “New American Rule” that would require a lawyer to set for each client at the beginning of a representation a limit of any amount (phrased in dollars per hour of legal services) on how high the contingent fee can go and then disclose to the client general information about the fees that the lawyer has charged to other clients.

Earlier Proposals for Contingent-Fee Reform

Almost every legal system outside the United States prohibits contingent fees, and it has repeatedly been suggested that the United States should do so as well. Nonetheless, as Professor Mary Ann Glendon points out, “countries that have outlawed contingent fees generally address their citizens legal needs by regulating legal fees, or by providing broad-based legal assistance, comprehensive social insurance, or some combination thereof.”16  A flat prohibition on contingent fees thus would probably require an alternative way of providing legal services to moderate and low income plaintiffs.17  One option would be regulating hourly rates that lawyers charge to middle-income plaintiffs and requiring pro bono representation of the indigent, but this approach is vulnerable to two objections: that it interferes with contractual freedom and disproportionately burdens lawyers with the costs of access to our legal system. Alternatively, government could increase subsidies for legal services, but legal services organizations have already been criticized for politicized agendas and litigiousness, making the political system reluctant to increase their budgets. The prospect of legal services lawyers giving substantial assistance to individual plaintiffs in tort cases is thus extremely unlikely.

An insurance scheme, similar to the plaintiffs’ insurance that replaces contingent fees in Germany, could be implemented in the United States. American lawyers, however, charge by the hour—unlike German lawyers who generally charge a flat fee according to the matter—and thus would have an incentive to run up their hours and bankrupt the insurer. Insurers would have to restrict lawyers’ fees and/or draw up lists of approved lawyers (as insurers of defendants now do). Furthermore, the insurance industry, and perhaps even the same insurance company, would take on conflicting roles: paying for and controlling the legal representation of plaintiffs and defendants simultaneously. Finally, as with health insurance, there is the question of what to do about the uninsured.18  Replacing contingent fees with litigation insurance might accomplish more than nonmarket solutions such as price controls, mandatory pro bono and expanded state subsidies of legal services, but comprehensive insurance that meets the needs of most plaintiffs is a long way off.

A less comprehensive, and probably less disruptive, approach would be to allow plaintiffs’ lawyers to continue to bear the risks of litigation for clients, but to abolish the contingent fee in favor of a conditional hourly fee that is both higher than the lawyer’s usual hourly rate and payable only if the client is successful. A few countries, such as England and Scotland, are experimenting with conditional fees in order to expand access to legal services, but also cap the amount of such fees, usually at no more than 100% of the lawyer’s hourly rate.19  The problem with the conditional fee, with or without a cap, is that it is not always easy to define “success” in litigation. An award that leaves the plaintiff with nothing after legal fees, for example, is hardly a “success” from the plaintiff’s vantage point, although it might be from the lawyer’s. Furthermore, the conditional fee does not reflect the value of the lawyer’s services to the client—a plaintiff who succeeds in winning a small judgment pays the same fee as a plaintiff who succeeds in winning a large judgment. The conditional fee system can be altered to increase the fee with the size of the judgment, but the more such tinkering is done, the more the fee looks like a contingent fee.

Yet another approach would be to allow contingent fees but cap the fee at no more than a specified percentage of the lawyer’s normal hourly rate (e.g. 200%). The problem here is that defining a “normal” hourly rate is not easy if a lawyer rarely charges by the hour or charges different rates for different cases. Furthermore, the cap, if set too low, might discourage lawyers from accepting cases that are risky (whether because the law or facts are uncertain or because the defendant might be judgment proof by the time a judgment is obtained). Perhaps most important, an externally imposed contingent-fee cap, like a cap on hourly or conditional fees, interferes with both the lawyer’s and client’s freedom of contract. Although the lawyer can avoid the economic impact of the cap simply by doing less work on a case, the client has no opportunity to respond by offering a higher fee in return for the work that she would like to see done.20 

The Brickman-Horowitz-O’Connell Proposal

In 1994, Lester Brickman, Michael Horowitz and Jeffrey O’Connell drafted a proposed rule that allows contingent fees to be charged only against the difference between the amount that a plaintiff actually recovers and a defendant’s early settlement offer. The proposal’s five principal provisions are that : (1) contingency fees may not be charged if the plaintiff accepts a settlement offer that was made before she obtained counsel; (2) the defendant may make an early settlement offer, but no later than 60 days after receipt of a demand for settlement from plaintiff’s counsel, and if such an offer is accepted, plaintiff’s counsel’s fees may not exceed hourly charges capped at 10% of the first $100,000 of the offer and 5% of any greater amount; (3) a demand for settlement from plaintiff’s counsel must include “basic, routinely discoverable information” so the defendant can evaluate the plaintiff’s claims, and the defendant must include similar information with its settlement offer to assist the plaintiff in evaluating the offer; (4), if the plaintiff rejects the defendant’s offer, contingency fees may only be charged on the difference between the plaintiff’s actual recovery and the offer; and (5) if no offer is made within 60 days after the receipt of a demand from plaintiff’s counsel, contingent fees are unaffected by the proposal.21 

This proposal curtails excessive contingent fees in situations in which they are most abusive: cases that settle early for large sums of money. The proposal also allows a plaintiff’s lawyer to earn a contingent fee only on what the lawyer actually accomplishes after a defendant’s settlement offer: the difference between the amount recovered and the offer. The proposal drew enthusiastic early support, including a foreword by Judge John T. Noonan, Jr. and a preface by Derek Bok in its first publication by the Manhattan Institute.

The New American Rule

Jim Wootton, President of the Institute for Legal Reform at the U.S. Chamber of Commerce, has designed an intriguing proposal that may be the best suggestion yet for contingent-fee reform. In essence, his “New American Rule for Contingency Fees” requires a lawyer charging a contingent fee to reach an agreement in writing with each client not independently represented by counsel22 on (i) a contingent percentage-of-net-recovery rate and (ii) a maximum contingent hourly rate (in dollars per hour of legal services).23  At the conclusion of the litigation, the client can choose whether to pay the contingent percentage-of-net-recovery fee or the maximum hourly fee (most clients would presumably choose the lower of the two).24 The maximum hourly rate can be any amount—even $10,000 per hour—but must be decided and disclosed to the client at the time the lawyer is retained, along with basic information about fees the lawyer has charged other clients. The fee actually charged to the client is of course subject to judicial review for reasonableness, just as it is today. The New American Rule would also require that a contingent-fee lawyer who resigns or is discharged before a case is concluded be compensated based on his contribution to the client’s ultimate success (some courts now give the lawyer compensation for the fair value of the hours he worked on the case, without adequately taking into account the client’s ultimate success or lack thereof).25 

Although the New American Rule requires lawyers to set ex-ante an hourly fee alternative that the client can choose ex-post, as a practical matter lawyers can assure themselves a percentage-of-net-recovery fee in all, or at least most, cases by setting a high per-hour rate. The Rule’s purpose is thus not to control the decisions of contracting parties, but instead to require the party having informational advantages in lawyer-client negotiations to make some important disclosures: how high the lawyer’s fees for the case can go in terms of dollars per hour of work, how much the lawyer charges other clients, and how many hours the lawyer has worked on the client’s case. The Rule also would not apply to cases where the client has independent counsel (including in-house counsel) concerning the fee arrangement.

Disclosure under the New American Rule could be made in a variety of ways. Mr. Wootton’s proposal calls for disclosure to the client of (i) a nonbinding estimate of the number of hours that the attorney will spend attempting to settle the case and, if necessary, trying the case (including appeal);26  (ii) the contingent fee and alternative hourly fee for the case as well as any conditions attached thereto (for example, the client’s liability for expenses); (iii) a monthly statement of hours worked on the case and expenses incurred;27 and (after the case is concluded) (iv) a final statement of the hours worked on the case, the total hourly fee, the total alternative contingent fee, and any expenses charged to the plaintiff. The proposal also calls for public disclosure of the range of hourly and contingent fees charged by the attorney. It would be simplest to require this public information to be filed with the state bar association or to be posted on the lawyer’s World Wide Web page, although the lawyer should also be required to provide prospective clients with a hard copy printout. The public disclosure of fee ranges need not be extensive. For example, a lawyer could be required to disclose (i) the high, (ii) the low, (iii) the median and (iv) the median for the top half (the fee falling in the 75th percentile), from the sample of percentage-of-net-recovery fees as well as from the sample of hourly fees quoted by the lawyer to clients in the past year.28  The lawyer should inform clients that these rates depend upon various factors affecting the lawyer’s risk, such as the likelihood of success, the amount of money involved, and the amount of lawyer time a case is likely to require. This should be enough information for a prospective client to see where the client’s case falls on the spectrum of quotes given by a particular lawyer.

The fee ranges disclosed under the New American Rule would be even more useful to clients if lawyers also were to disclose some additional statistics about cases in which they earned a fee in the previous year: their win/loss ratio for litigated cases, the percentage of cases that were settled, the median award obtained, and the median number of lawyer hours spent on a case. None of these additional statistics, however, is necessary for fee range disclosures under New American Rule to accomplish their most important objective: telling clients which lawyers charge all or most clients the same contingent percentage and hourly fees regardless of the type of case involved. Unless all of the cases handled by such lawyers are alike, these lawyers almost certainly expect a higher return from time invested in some cases than from time invested in others, meaning their flat rate is a form of price discrimination and some clients are paying too much.29 

Advantages of the New American Rule

An important advantage of the New American Rule is that the underlying concept is easy to explain: the lawyer cuts his piece of the cake in two separate ways before the case begins, and after the case is over the client decides which of these two pieces goes to the lawyer. If the Rule is put on a ballot in California or another state that uses the referendum, voters will understand it. In an age when legal reform proposals trigger multimillion-dollar advertising campaigns by both sides, as well as a lot of misinformation, the simplicity and clarity of the New American Rule will be critical.

Another advantage of the New American Rule is that, unlike the Brickman-Horowitz-O’Connell Proposal, it does not entangle the contingent-fee issue with settlement offers and discovery disputes. Early settlement and early discovery are laudable goals, but perhaps goals that separate procedural rules should promote without making the lawyer’s fee contingent thereupon. Although the Brickman-Horowitz-O’Connell Proposal is in many ways sound, it could make prospective plaintiffs worry that discovery, and particularly settlement, decisions will be influenced by a lawyer’s desire to maximize his fee (of course this conflict of interest already exists, but many clients are unaware of it). The New American Rule by contrast does not risk embroiling courts in disputes over early discovery motions filed by lawyers who are trying to maximize their fee or minimize the fee of opposing counsel. Because it is thus easy to administer, the Rule should be quite popular with judges as well as voters.

Finally, the New American Rule only minimally interferes with contractual freedom
because it still allows the lawyer to charge what the client will pay (subject to the same ex-post judicial reasonableness determination that exists today). The lawyer’s fee under the Rule only must be phrased in such a way that the client is aware of his maximum liability, in dollars per hour of legal services. It could be argued that the Rule therefore won’t really accomplish much, and it won’t if clients hire lawyers who set exorbitant conditional hourly fees to limit exorbitant contingent-percentage fees. The Rule, however, requires a lawyer to disclose important information about a client’s fee liability exposure in the very cases in which contingent-fee gouging is most prevalent—cases in which the client recovers large amounts of money relative to the amount of work done by the lawyer. It is up to the client to decide what to do with this information.30 

One possible criticism of the New American Rule is that it does not address contingent-fee lawyers’ conflicts of interest. A lawyer who sets a high hourly rate under the Rule is still likely to be influenced by the contingent fee when handling the case and making settlement recommendations.31 Although the Brickman-Horowitz-O’Connell Proposal also does not eradicate conflicts of interest between lawyers and clients, its partial repeal of the contingent fee would reduce perverse incentives created by contingent fees. Nonetheless, lawyers who bill by the hour also have incentives that conflict with those of their clients (to prolong litigation and spend as many hours on a case as possible).32  Because unethical lawyers can always milk excessive profits out of cases, whether they are billing by the hour, on a contingency basis, or with flat fees, stricter ethics rules and more aggressive enforcement by courts are needed to address perverse incentives created by lawyers’ fees. Expecting a fee regulation proposal also to solve the conflict of interest problem is probably expecting too much.

How should the New American Rule be implemented? State ethics rules and rules of civil procedure would be a good place to start. In some states, such as California, it may be best to place the Rule on the ballot for voters to approve. The Federal Rules of Civil Procedure should also probably be amended to make lawyers’ fees subject to the New American Rule when a case is filed in federal court. Finally, state legislatures and Congress have already begun to regulate the practice of law when bar associations and courts are reluctant to enforce or have relaxed existing ethics rules.33  The New American Rule could be an important part of this initiative as well.

The New American Rule in Practice

Although it is impossible to predict with certainty how lawyers and clients would respond to the New American Rule, the possibilities are worth a closer look, beginning with a hypothetical:

Alice suffered a serious and permanent back injury when her car was struck from the rear. The other driver was given a speeding ticket at the scene of the accident for driving 60 miles per hour in a 45 mile per hour zone, and Alice wants to sue the other driver and his insurance company. All of the lawyers Alice has spoken with have told her that she has a good (better than even) chance of recovering at least $300,000 and a very good (almost certain) chance of recovering at least $100,000. Alice can choose between three lawyers who have similar practices and equally good reputations. Each of the lawyers quotes her a risk-weighted fee according to the New American Rule.

Lawyer A: 33% of recovery or $700 per hour, whichever is less.
Percentage Range: 20% low; 33% median; 36% top quarter; 40% high
Hourly Range:  $500 low; $700 median; $800 top quarter; $900 high

Lawyer B: 40% of recovery or $500 per hour, whichever is less.
Percentage Range: 33% low; 40% median; 45% top quarter; 50% high
Hourly Range:  $400 low; $500 median; $550 top quarter; $600 high

Lawyer C: 30% of recovery or $1,000 per hour, whichever is less.
Percentage Range: 20% low; 25% median; 30% top quarter; 35% high
Hourly Range: $700 low; $900 median; $1,200 top quarter; $1,500 high

Alice’s case thus lands at exactly the median percentage and the median hourly rate for Lawyer A and Lawyer B, meaning that her case is probably typical for each lawyer’s caseload with respect to risk, the amount of money involved and the amount of lawyer time required (or at least that divergences from the median for these factors cancel each other out). If Alice’s case were less risky and less time consuming than the typical case in either of these lawyers’ offices, the percentage fee should be lower than that lawyer’s median, unless the amount of money involved were also less than the median, an unattractive feature that would drive the percentage fee back up toward the median.

Lawyer C, on the other hand, presumably believes Alice’s case to be more risky, to involve a lower expected award and/or to be more time intensive than most of the cases in his office, leading him to quote a percentage fee in his top quarter and an hourly rate slightly above his median. Lawyer C might simply have more cases that pose lower risk, involve larger amounts of money and/or are less time consuming than the cases handled by Lawyers A and B. If Lawyer C’s win/loss ratio for the previous year is higher than that for Lawyers A and B, if Lawyer C settles a larger percentage of his cases, if Lawyer C’s cases have a higher median award, and/or if Lawyer C spends less hours on his median case, Lawyer C’s different caseload (particularly if the differences are several and substantial) probably explains the position of Alice’s case in Lawyer C’s range. If so, Alice should not be concerned (some of these differences might even persuade Alice that Lawyer C is a better lawyer). If, on the other hand, the cases handled by Lawyers A, B and C are relatively similar, Lawyer C is probably more skeptical than Lawyers A and B about Alice’s case, in which case Alice might—but would not necessarily—get a worse deal from Lawyer C than from Lawyers A and B.

In shopping for a lawyer, Alice also will probably avoid a lawyer who—unlike Lawyers A, B and C, but like many contingent-fee lawyers today—specifies no range or only a very narrow range of percentage fees for the cases in his office. Unless all of the cases such a lawyer handles are alike, the lawyer is probably engaging in price discrimination (charging according to what a less than fully competitive market will bear instead of according to cost and risk) and thus is probably charging some of his clients too much. The requirement that lawyers publicly disclose a range for percentage and hourly fees is thus one of the most important features of the New American Rule, because it puts clients like Alice on notice that they could be overcharged by lawyers who charge all or most clients the same percentage fee.

Assuming, however, that Alice chooses one of these three lawyers, she can clearly see that Lawyer B is the most likely of the three to charge her an hourly rate in the end, and Lawyer C is most likely to charge her a percentage fee, with Lawyer A somewhere in between. Indeed, Lawyer C’s high hourly fee ranges send a clear message that he prefers percentage fees. If Alice chooses Lawyer C, she will thus pay a percentage fee unless she wins a very large amount of money relative to the amount of time Lawyer C works on the case.

Assuming that each of the three lawyers works on the case for 100 hours before a judgment or settlement, their respective fees are as shown in Table 1, below.

Table 1.

Alice therefore is best off with Lawyer C, and worst off with Lawyer B, if she recovers $100,000, but in the last three scenarios ($200,000, $300,000 and $400,000) she is best off with Lawyer B, the lawyer who quotes her the lowest hourly rate. If Alice is not very confident about her prospects of winning more than $166,000 (the break-even point where Lawyer C’s percentage fee equals the hourly fee of Lawyer B), she is likely to choose Lawyer C and expect to be charged a 30% fee. If she thinks the award will be for a larger amount, she is likely to choose Lawyer B and expect to be charged $500 per hour or $50,000 for 100 hours.

At first glance, Lawyer A seems to have priced himself out of the market for Alice’s case (in all award scenarios, Lawyer A’s percentage fee is higher than that of Lawyer C, and Lawyer A’s hourly fee is higher than that of Lawyer B). On the other hand, if Alice chooses Lawyer B, she will find that she would have saved $6,666 out of a $100,000 award by choosing Lawyer A instead. Although Alice would have lost significant amounts by choosing Lawyer A over Lawyer B in other scenarios (e.g. $5,555 out of a $166,666 award and $16,666 out of a $200,000 award), Lawyer A perhaps makes up for this by saving Alice money on the low end. Choosing Lawyer C, of course, would have helped Alice the most on the low end, but Lawyer A is $20,000 cheaper than Lawyer C for a $300,000 award and $30,000 cheaper than Lawyer C for a $400,000 award. Lawyer A thus may still be competitive because, particularly if Alice has little idea how much she is likely to recover, Lawyer A offers a compromise between the high contingent fee of Lawyer B and the high hourly rate of Lawyer C.

The foregoing discussion assumes of course that Alice knows that 100 hours of attorney time will be spent on her case. More likely, she will have to make her decision knowing only of a range of attorney time that could be spent on the case (perhaps 50 to 200 hours). If only 50 hours are spent, Lawyer B looks decidedly more attractive, and is indeed Alice’s first choice in all of the scenarios (Table 2).

Table 2.

If on the other hand, 200 hours of attorney time are spent on Alice’s case, Lawyer C is cheaper for Alice in all but the $400,000 scenario (Table 3).

Table 3.

Under the New American Rule, each attorney is required to provide the client with a nonbinding estimate of the amount of hours that will be spent on the case. This estimate should help Alice decide between Attorney B and Attorney C, although Attorney B might low-ball his estimate in order to persuade Alice to choose him. For this reason, jurisdictions that adopt the New American Rule should consider requiring disclosure of an additional statistic: for the sample of cases in which the lawyer earned an hourly fee during the previous year, the median percentage difference between the number of hours that the lawyer estimated at the beginning and the number of hours the lawyer actually worked on the case. Even if most lawyers consistently underestimate, this statistic should show which lawyers are consistently underestimating by relatively large amounts (for example 30% as opposed to 10%). Thus, a particularly shrewd client might increase a lawyer’s estimate of hours by the lawyer’s median percentage underestimation for the previous year.

Will, however, the typical plaintiff in Alice’s situation go through all of these calculations? Maybe not if left to her own resources. Software, however, could easily be made available on state bar association Web pages, or even on the U.S. Chamber of Commerce Web page, that would allow a client to create similar tables using different estimates of the number of hours that will be spent on a case and of the amount of money that will be recovered. Not every client will choose the lawyer who is the cheapest in retrospect, and neither is this a goal of the New American Rule. On the other hand, glaring differences will be discernable (for example, the advantage of choosing Lawyer A only if the client is risk averse on the low end and also very uncertain about the probable award in the case; the advantage of choosing Lawyer B over Lawyer C in most cases that require little lawyer time; and the advantage of choosing Lawyer C over Lawyer B in most time intensive cases). Clients could use this information to avoid the most egregious mistakes in choosing lawyers, and to induce lawyers to replace uniform percentage fees with competitive rates tailored to the characteristics of the case involved.

Furthermore, even if only minimal information is disclosed by plaintiffs’ lawyers (perhaps far less than suggested in this report) the simple requirement of the New American Rule that clients be offered an alternative per hour limit on contingent fees will tell plaintiffs most of what they need to know to choose a lawyer intelligently. The monthly report of the number of hours worked would also let clients know how hard the lawyer is working on the case. This is information which defendants routinely get from their lawyers and to which plaintiffs should be entitled as well. The principle behind the New American Rule is also quite simple: lawyers should impose on themselves a per hour limit on their contingent fees and then be honest with their clients about how much risk they are taking and how much more than their normal hourly rates they charge to make taking that risk worthwhile.

Need for the New American Rule

No client and no rational policy maker should begrudge an attorney charging a premium based on risk. Indeed, risk premiums make the justice system accessible to those who cannot afford to pay a normal hourly rate—win or lose. In some cases, however, the client is terribly injured, having been paralyzed or otherwise permanently disabled in an accident that was clearly someone else’s fault. The lawyer’s job amounts to gathering evidence from medical records and perhaps a few witnesses and making demand on an insurance company. The defendant is often judgment proof and, if the defendant carries the minimal insurance required by law, the insurance proceeds are likely to be somewhere between $200,000 and $300,000.34  The insurance company in turn is often willing to pay the policy limit to avoid litigation and the additional liability for unreasonable refusal to settle that could result from a jury award substantially above the policy limit.35  In such cases, it may be highly predictable that, after a decent interval, the insurance adjuster and the plaintiff’s attorney will settle and the company will send a check payable to the attorney from which the attorney deducts his fee plus any expenses incurred. The attorney will then deliver the balance to a grateful client after the client signs a release forfeiting any further right to sue.

Would that client be so grateful if he knew that his lawyer had spent a total of only ten hours on the case because the lawyer knew more work was not needed, if he knew that the lawyer’s nonlegal staff had done the bulk of the work assembling medical records and accident reports, and if he knew that the lawyer’s $100,000 fee came to $10,000 per hour. Under the New American Rule, the lawyer would have limited his fee with an hourly rate, and this rate would presumably reflect the fact that this was a low risk case—perhaps it would be as high as two times the normal hourly rate charged for noncontingent fee work in the same community. If this rate were $400, instead of a normal rate of $200 an hour without the contingency, the client would be charged $4,000 for the ten hours plus expenses. One does not need an abacus to calculate the difference between a $200,000 recovery (out of a $300,000 award) and a $296,000 recovery. Yearly interest alone on the $96,000 at 6% would be $5,760 (a lifetime annuity could yield significantly more), an amount which perhaps should go to the disabled plaintiff and his family rather than to the plaintiff’s lawyer. The lawyer might claim that he is worth the $100,000 because his reputation as a jury mesmerizing litigator frightened the insurance company into paying the policy limits. This might be true. If so, the attorney could, under the New American Rule, tell the client that even for low risk cases his contingency fee is limited at $5,000 or even $10,000 an hour. The client then has the opportunity to compare these rates with those of other lawyers who charge less, and then decide for himself whether the higher fee is worth it. This type of up-front honesty about risk adjusted per hour charges not only helps the client decide what to do, but also could help restore the public’s confidence in the ethics of lawyers.


Contingent fee abuse hurts everyone. Plaintiffs suffer when their net recovery is less because lawyers take too much. Consumers pay more for merchandise when juries make up for contingent fees by increasing awards against manufacturers36  and when excessive contingent fees from quick settlements encourage lawyers to file strike suits. The legal profession suffers from disrepute that is in reality attributable to the relatively few lawyers who abuse their market power to charge far more than their services are worth,37  even though many contingent-fee lawyers do not earn abnormally high incomes.38 

The New American Rule furthers competition by providing consumers with the information they need to overcome asymmetry of information and negotiating power in the market for legal services. Self-imposed limits that lawyers choose under the Rule send a clear message about what to expect and protect clients from unfair bargains struck at a time when clients are under emotional and often physical stress. The New American Rule also gives contingent-fee lawyers who earn only modest incomes an opportunity to break into the large cases with high probability of liability. Direct comparison will be possible between the maximum prices charged by lawyers who are happy to walk away from a successful contingent-fee case with $1,000 an hour, and lawyers who must now tell their clients if, for them, $10,000 or even $100,000 an hour is not enough.

With few exceptions, politicians and judges have not stood up to lawyers who charge excessive contingent fees. Few proposals to limit contingent fees have been implemented at the state or federal level, in part because highly successful tort lawyers have political power but also because many such proposals are perceived either to be too complex or to unduly interfere with freedom of contract. The New American Rule is an appropriate first amendment to a client’s bill of rights because it gives plaintiffs, as well as lawyers who charge reasonable fees, a fair chance.


1 The author is grateful for the financial support of the Manhattan Institute for this project, and for the helpful comments of Judyth Pendell and James Wootton.

2 Ninety-five percent of personal injury cases are taken on a contingent fee. Note, Settling for Less: Applying Law and Economics to Poor People, 107 Harv. L. Rev. 442, 448 n.23 (1993) (citing James L. Kakalik & Nicholas M. Pace, Costs and Compensation Paid in Tort Litigation 37 (1986)).

3 Model Rule 1.5 lists eight factors used to determine whether a fee is reasonable (“The factors to be considered in determining the reasonableness of a fee include the following: (1) the time and labor required, the novelty and difficulty of the questions involved, and the skill requisite to perform the legal service properly; (2) the likelihood, if apparent to the client, that the acceptance of the particular employment will preclude other employment by the lawyer; (3) the fee customarily charged in the locality for similar legal services; (4) the amount involved and the results obtained; (5) the time limitations imposed by the client or by the circumstances; (6) the nature and length of the professional relationship with the client; (7) the experience, reputation, and ability of the lawyer or lawyers performing the services; and (8) whether the fee is fixed or contingent.”). See also Model Code DR 2-106(B) (listing substantially similar factors for determining reasonableness).

4 Ted Schneyer, Legal-Process Constraints on the Regulation of Lawyers’ Contingent Fee Contracts, 47 DePaul Law Review 371, 399 (1998).

5 Id. at 400. Once in a while, a court recognizes that a 33% or even a 25% contingent fee can be excessive. See In Re Swartz, 686 P.2d 1236, 1239, 1244 (Ariz. 1984) (holding that a contingent fee of one-third of a client’s recovery was excessive given the relatively high certainty of liability and the amount of anticipated damages) and Horton v. Butler, 387 So. 2d 1315, 1317 (La. Ct. App. 1980) (holding that a one-quarter contingent fee was unreasonable where the lawyer’s only action to assist a widow in collection of a life insurance claim was to inform the insurer of the correct date that her deceased husband enlisted in the Air Force). Judicial review of this sort, however, is rare. See Schneyer, supra at 400.

6 See e.g. Herbert M. Kritzer, The Wages of Risk: The Returns of Contingency Fee Legal Practice, 47 DePaul Law Review 267, 287 (1998) (1995 survey of Wisconsin contingency fee lawyers using a sampling frame defined by the Litigation Section of the State Bar of Wisconsin and resulting in 511 usable responses to questions “on up to three cases: the case closed most recently after a trial had begun, the case close most recently after filing but before the start of trial, and the case closed most recently before filing.”). Id. at 284. Of the 989 cases about which information was provided, 53% had a flat one-third rate, 3% had a flat one-quarter rate, 2% had some other flat rate, 39% had a “variable percentage” rate and 3% had a rate that mixed various elements of the hourly and contingent fee. Id. at 287-88. “The most common pattern for those cases employing a variable percentage called for a contingency fee of 25 % if the case did not involve substantial trial preparation (or, in some cases, did not get to trial), and 33% if the case got beyond this point. The contingency fee rose to 40% or more if the case resulted in an appeal.” Id. at 286. Although Kritzer’s individual interviews revealed that some lawyers were willing to “negotiate” with clients and lower their fees to get a particularly good case, id. at 288, the data on the 989 randomly chosen cases does not reveal a propensity of plaintiffs’ lawyers to price cases individually according to the size of the likely damage award, the likelihood of success, and the number of lawyer hours likely to be expended. One-third was charged in 92% of the flat rate cases, id. at 286, and was also the most common rate for cases brought to trial (but not appealed) under a variable percentage fee. Critics of contingent fees might question this data, which was voluntarily submitted by contingent-fee lawyers in a survey conducted at the University of Wisconsin. Kritzer’s data, however, honestly reveals what critics of contingent fees have been concerned about for quite some time: the tendency of contingent fees to “stick” around a level of 33% or above in a market that is supposedly competitive.

7 ABA Informal Opinion 1521 (1986) states that “when there is any doubt whether a contingent fee is consistent with the client’s best interests [and the client is able to pay a reasonable fixed fee], the lawyer must offer the client the opportunity to engage counsel on a reasonable fixed fee basis before entering into a contingent fee arrangement.” This, however, is one of the most widely ignored rules in legal ethics.

8 See George B. Shepherd, Time and Money: Discovery Leads to Hourly Billing, 1999 Illinois Law
Review 91, 108 (discussing the strong incentive that lawyers under an hourly contract have to conduct unnecessary work). Professor Lisa Lerman recently completed an extensive study of overbilling by hourly rate attorneys at some of the country’s most prestigious law firms. See Lisa G. Lerman, Blue Chip Bilking: Regulation of Billing and Expense Fraud by Lawyers, 12 Geo. Jour. Legal Ethics 205, 209 (“[R]egulation of this type of conduct is very difficult because no one except the lawyer really knows how much time was spent and how much was billed.”).

9 ”Competent representation requires the legal knowledge, skill, thoroughness and preparation reasonably necessary for the representation.”  Model Rule 1.1. See also ALI Restatement of the Law Governing Lawyers, Section 28 (1996) (Proposed Final Draft No. 1) (providing that a lawyer must “act with reasonable competence and diligence.”).

10 Model Rule 1.7(b) and comment [6] provide that a lawyer should not allow his own interests, financial or otherwise, to materially limit his representation of a client. An attorney is also required to evaluate a settlement offer on the basis of the client’s interests, not his own, and then to abide by the client’s decision whether or not to accept the offer. See Model Rule 1.2(a).

11 Lester Brickman, Contingent Fees without Contingencies: Hamlet without the Prince of Denmark?, 37 UCLA L. Rev. 29, 127 (1989).

12 See e.g. Kritzer, supra note 6; Herbert M. Kritzer, Rhetoric and Reality . . . Uses and Abuses . . . Contingencies and Certainties: The Political Economy of the American Contingent Fee (1995) (paper published by the Wisconsin Institute for Legal Studies).

13 See Kritzer, supra note 6 at 286-87 (stating that one-third was charged in 92% of the flat rate cases, and that among “variable percentage” case, the most common arrangement involved a one-third fee if the case went to trial and a fee of 40% or more if the case was appealed).

14 Prospect theory in psychology posits that decision makers are risk averse when deciding between two options that result in a gain from their present frame of reference and risk preferring when deciding between two options that result in a loss. See Amos Tversky & Daniel Kahneman, Prospect Theory: An Analysis of Decision Under Risk, 47 Econometrica 263 (1979). Professor Jeffrey Rachlinski has demonstrated with psychological studies that plaintiffs are likely to be risk averse when weighing settlement offers against taking their chances in litigation for even larger potential gains. See Jeffrey J. Rachlinski, Gains, Losses, and the Psychology of Litigation, 70 S. Cal. L. Rev. 113 (1996). Risk aversion should affect not only a plaintiff’s decision about whether to settle, but also her choice of lawyers, particularly if she worries that her case could be lost if she tries to protect her upside gains with a lawyer that charges less than a 33% contingent fee. Contingent-fee lawyers of course can fuel these fears by disparaging the few lawyers who depart from “customary” fee schedules. Lawyers furthermore have an incentive to conceal the fact that a contingent fee can increase a client’s risk at the lower range of the recovery spectrum—the risk that the lawyer will invest very little work in a case, settle cheaply and then walk away with his one-third share of the amount that the client has recovered.

15 Herbert Kritzer has few criticisms of the existing contingent fee system, but still recognizes that clients lack useful information on fees. “The basic problem for potential clients is a lack of information. Word of mouth provides some information on who is and is not a ‘good’ lawyer, but there is virtually no information on differences in fees. Advertising would provide little information on the quality of service and/or result, but it could provide some very useful information on fee options.” Kritzer, supra note 6 at 289-90.

16 Mary Ann Glendon, A Nation Under Lawyers 54 (1994).

17 Contingent fees, however, do not solve the problem of inadequate access to legal services. Contingent fees only help the client of limited means in circumstances where a contingent-fee lawyer is willing to take the case. These are usually cases in which the client is the plaintiff, is suing for substantial amounts of money and is reasonably likely to win. The laudable goal of increasing access to the justice system furthermore needs to be balanced against societal costs of increasing incentives to file meritless suits. Other approaches to the access problem, such as loosening the bar’s monopolistic restrictions on unauthorized practice of law, may be more promising, and have less harmful side effects, than unregulated contingent fees. See generally Deborah L. Rhode, Policing the Professional Monopoly: A Constitutional and Empirical Analysis of Unauthorized Practice Prohibitions, 34 Stan, L. Rev. 1 (1981) (empirical study of bar association enforcement of prohibitions on unauthorized practice); Ralph C. Cavanaugh & Deborah L. Rhode, Project, The Unauthorized Practice of Law and Pro Se Divorce: An Empirical Analysis, 86 Yale L.J. 104, 150 (1976).

18 One solution for the uninsured is insurance that a plaintiff buys after her claim arises and she has selected a lawyer. The insurance premium would be a percentage of any future award in the case, and the insurance company would in return pay the plaintiff’s lawyer an hourly or flat fee. The premium charged by the insurer (unlike the 33% generally charged by contingent-fee lawyers) would presumably vary with several factors including the strength of the case, the amount of damages expected and the track record (both for reasonable billing and for success) of the lawyer. The plaintiff could price different lawyers with different insurance companies before selecting both a lawyer and an insurer. Such a scheme would of course violate common law prohibitions on champerty—sponsoring litigation in return for a share in the proceeds. In the United States, however, champerty prohibitions are already subject to a glaring exception: contingent fee lawyers sell litigation insurance to their clients by providing services and paying expenses in return for a share of litigation proceeds, but they are allowed to do so without competition from nonlawyer insurers. Furthermore, lawyers who refuse to work by the hour also get away with questionable conduct under the antitrust laws: tying one of their products, litigation insurance, to another, legal services. See Richard W. Painter, Litigating on a Contingency: A Monopoly of Champions or a Market for Champerty?, 71 Chi. Kent L. Rev. 625 (Symposium on Fee Shifting 1995).

19 In 1990, Parliament authorized “conditional fee arrangements” in certain “specified proceedings” subject to maximum “uplifts” set by the Lord Chancellor, who then set the maximum uplift at 100%, specifically observing that “[t]he scheme will not operate in the same way as the contingency fee system of the USA.” See Courts and Legal Services Act (1990) (c.41), section 58(3) & (4); Lord Mackay, Reducing Risks for Clients—the Introduction of the Conditional Fee Scheme in England and Wales, Law Society’s Gazette, July 5, 1995, at 10. Scotland’s Law Reform Act of 1990, section 36 allows a solicitor to agree with a client that if litigation is successful the “solicitor’s fee will be increased by a percentage” that “shall not exceed limit[s]” set forth in the Act of Sederunt, S1 1992/1879 (for the sheriff court) and S1 1992/1898 (for the Court of Session) (specifying that the solicitor’s fee may not be increased by more than 100%). See Walter G. Semple, Fees in Speculative Actions, J.L. Soc’y of Scotland, Feb. 1994, at 57.

20 “Such rate ceilings are largely cosmetic, keeping the final fee at what seems a reasonable level to the outside observer, while still permitting the lawyer covertly to pick and then milk (through underwork) the lucrative cases.” Kevin M. Clermont & John D. Currivan, Improving on the Contingent Fee, 63 Cornell L. Rev. 529, 535 (1978).

21 Lester Brickman Et Al. Rethinking Contingency Fees 20-23 (1994).

22 The New American Rule is intended to protect plaintiffs who are negotiating fee agreements with lawyers without the benefit of independent legal advice. Plaintiffs who have hired independent counsel, including in-house counsel in a corporation, to advise them in retaining another lawyer on a contingent-fee basis should therefore probably be excluded from the proposal’s requirements. “Independent counsel” should not, however, include any lawyer to whom the plaintiff is referred for advice by the contingent-fee lawyer, any lawyer who normally refers business to the contingent-fee lawyer, or any lawyer who has a professional relationship with the contingent-fee lawyer, including a fee sharing agreement. The plaintiff, furthermore, should not receive assistance from the contingent-fee lawyer in paying the fee charged by the independent counsel.

23 The proposal contemplates that contingent fees will be charged only against net recovery (recovery less any expenses for which the plaintiff is liable). Although contingent fees can also be calculated on the basis of gross recovery, ethics rules generally require contingent fee agreements to state whether a percentage fee will be based on the gross or the net recovery. See Schneyer, supra note 4 at 384, citing Model Rules of Professional Conduct Rule 1.5 (1994)

24 In the rare circumstances in which a client chooses to pay the higher of the two fees, courts and bar disciplinary committees should, if warranted by the circumstances, inquire into whether the higher fee was paid as a result of any coercive conduct on the part of the lawyer. In particular, a lawyer should not be permitted to assert a
dubious claim against the client in connection with the case or a related case and then agree to relinquish that claim in return for the higher fee.

25 "[T]he rule, which is now recognized in almost every state . . . is that a client’s discharge of a lawyer ends the lawyer’s right to recover on the contract of employment.” Charles W. Wolfram, Modern Legal Ethics 546 (1986). Most jurisdictions award the attorney quantum meruit recovery for the value of her services rendered. Id. This rule, however, is arguably unfair to a client who discharges a contingent-fee attorney. See Lester Brickman, Setting the Fee When the Client Discharges a Contingent Fee Attorney, 41 Emory L.J. 367, 368 (1992) (recommending that clients who employ an attorney under a contingent-fee arrangement not be forced to pay a quantum meruit fee upon discharge of the attorney unless the suit succeeds).

26 Some plaintiffs’ lawyers will object that estimates of hours are difficult to make when it is uncertain what litigation tactics defendants will deploy. The required estimate, however, is nonbinding and at least should give the client a general idea of how much time will have to be spent on the case. Furthermore, although the nonbinding estimate would be helpful to the client, this provision could be deleted from the proposal without unduly interfering with the smooth functioning of the New American Rule.

27 Although the client will have already entered into the fee agreement with the lawyer at this point, the monthly reports give the client early warning if the lawyer runs up unnecessary hours or pads his time sheets. If the circumstances were such that the client expected to choose the hourly rate at the conclusion of the litigation, the client could respond by firing the lawyer or taking other remedial action.

28 Means are more useful than medians for measuring whether a lawyer’s contingent fees are as a whole excessive, because means measure the extent of overcharging on the high end (above the median). For similar reasons, means are a preferable measure for determining whether jury verdicts are excessive. For this reason, scholarly literature on these issues is much more persuasive if mean figures are used. Individual clients, on the other hand, are not as interested in whether a lawyer grossly overcharged a few clients, as they are in whether they are getting a good deal. How the rate quoted to them compares with the rate quoted to the lawyer’s typical (median)
client is something the client would want to know (in fact a high mean reported by a lawyer who overcharges a few of his clients might lull prospective clients into believing they are getting a good deal when in fact they are being charged more than the lawyer’s typical client).

29 In a fully competitive market, the lawyer would not ordinarily sell his time to one client for an expected return less than he could earn by selling his time to another client. However, lawyers like other service providers (airlines, educational institutions, etc.) have an incentive to engage in, and can get away with, price discrimination (charging clients according to their willingness to pay rather than according to cost) in a less than fully competitive market. See Painter, Litigating on a Contingency, supra note 18 at 664.

30 The New American Rule has an additional advantage of establishing a body of public information that over time should assist judges in evaluating the reasonableness of contingent fees charged in individual cases.

31 The New American Rule does not regulate settlement decisions or tie lawyers’ fees to settlement decisions, but does require lawyers to convey settlement offers to their clients along with a nonbinding estimate of both the cost and benefits of turning down the offer and pursuing a higher settlement offer or verdict.

32 The Brickman-Horowitz-O’Connell Proposal deals with this problem in part by capping the hourly fees that can be charged if a case is settled as a result of an early offer under the proposal.

33 See Geoffrey C. Hazard, Jr., The Future of Legal Ethics, 100 Yale L.J. 1239, 1279 (1991) (observing that courts and legislatures are becoming more involved in regulation of the legal profession).

34 In most states, drivers are required to maintain minimal levels of liability insurance. The limits of these policies per occurrence range from $100,000 to $300,000. The insurance company is usually also responsible for paying the defendant’s lawyer. Paying for defense lawyers to litigate a losing case when a jury is certain to award the “policy limits” or more makes no sense for the company. The insurance adjuster knows this and so does the plaintiff’s attorney when he takes the case and quotes his contingent fee.

35 Courts sometimes impose liability on insurers that unreasonably refuse to settle a case within the limits of a policy. See Kent D. Syverud, The Duty to Settle, 76 Va. L. Rev. 1113, 1120-26 (1990).

36 The 33% contingent fee is so common that most jurors are aware of it even though procedural rules usually prohibit either party from informing jurors about how much of a judgment will be paid to the plaintiff’s lawyers. Jurors who want a plaintiff to walk away from a case with $X usually know exactly how much to award (150% of $X).

37 The hundreds of millions, and even billions, of dollars in fees demanded by plaintiffs’ lawyers in tobacco class actions brought by state attorneys general are only the most recent examples of extreme overcharging.

38 Professor Kritzer reports that, based on a survey of cases brought by Wisconsin contingent fee lawyers “I find that the median effective hourly rate is $132.” Kritzer, supra note 6 at 292. This statistic, if representative of contingent-fee cases brought
nationwide, demonstrates that many cases do not result in fees that are excessive and that many contingent fee lawyers do not earn excessive incomes. The statistic says nothing, of course, about whether some cases above the median are charged excessive fees. Kritzer goes on to point out that the mean effective hourly rate for his sample is a considerably higher $242., id. at 293, but that eliminating the top ten percent of cases from the sample would bring the mean back down to $136. Id. The highest “outlier” case in Kritzer’s sample generated an effective hourly rate of $2,900, and ten percent of the cases had effective hourly rates of $1,000 or more. Id. at 295. Kritzer thus correctly recognizes “the role of a relatively small portion of cases as generating the ‘profits’ across a portfolio of contingency fee cases.” Id at 293. The frequency and magnitude of high-end fees furthermore may be even greater than Kritzer’s study suggests. Participation in Kritzer’s survey was voluntary, and there is little reason for contingent fee lawyers to disclose to scholars studying whether contingent fees are excessive those cases in which they earned an effective hourly rate in excess of $1,000.


Center for Legal Policy.



In this Civil Justice Report Richard Painter, Professor of Law at the University of Illinois, reviews several proposals for legal reform aimed at protecting clients from excessive contingent fees.


Earlier Proposals for Contingent-Fee Reform

The Brickman-Horowitz-O’Connell Proposal

The New American Rule

Advantages of the New American Rule

The New American Rule in Practice

Table 1

Table 2

Table 3

Need for the New American Rule




Home | About MI | Scholars | Publications | Books | Links | Contact MI
City Journal | CAU | CCI | CEPE | CLP | CMP | CRD | ECNY
Thank you for visiting us.
To receive a General Information Packet, please email
and include your name and address in your e-mail message.
Copyright © 2015 Manhattan Institute for Policy Research, Inc. All rights reserved.
52 Vanderbilt Avenue, New York, N.Y. 10017
phone (212) 599-7000 / fax (212) 599-3494