Lawyers, particularly contingent fee lawyers, are gatekeepers. They control the flow of civil cases into the courts. While lawyers can exercise this gatekeeping role in ways that either encourage or discourage potential litigants, the currently popular image of lawyers is that they stir up trouble. In A Nation Under Lawyers, Mary Ann Glendon argues that this need not be, nor has always been, the case; she quotes, at least twice, an observation attributed to Elihu Root that "About half of the practice of the decent lawyer consists in telling would-be clients that they are damned fools and should stop".(2) Glendon implies that today lawyers are more interested in encouraging clients to sue than in being the "decent lawyers" of yesteryear.

The image that lawyers stir up cases is particularly associated with contingency fee lawyers. This is not surprising because of the apparent logic of the contingency fee: the lawyers get a cut of whatever they recover, and without cases there is no cut to get. Various interests--physicians, accountants, auto executives, chemical companies--blame contingency fee lawyers for much of what those interests view as crises arising from the liability system. Undoubtedly, there are lawyers who push the edge of the liability frontier, or who engage in practices pejoratively referred to as ambulance chasing. However, the day-to-day reality of most contingency fee legal practices is very different from this image. While virtually every contingency fee practitioner wants to find highly lucrative cases, such cases are relatively rare. Many cases presented to lawyers are not winnable, or do not offer a prospect of even a moderately acceptable fee. The contingency fee practitioner seeks to choose cases that offer a high probability of providing at least an acceptable return, hoping to find some fraction of cases that present the opportunity to generate a significant fee.

Thus, while lawyers may encourage or seek out cases, the contingency logic suggests that contingency fee lawyers should reject a large number of cases that potential clients bring to them. Lawyers evaluate potential cases in terms of the risks involved and the potential returns associated with those risks. An attorney will reject cases which do not satisfy the attorney's risk/return criteria. Thus, contingency fee lawyers resemble portfolio managers, choosing to "invest" (their time) in risky cases hoping to obtain adequate or better returns. (3)

Most of the above is speculative. Is there any published information that provides systematic insights into case screening practices of lawyers? The answer, surprisingly, is almost none! There have been occasional articles in the legal press that describe case screening practices of top end law firms.(4) There are prescriptive articles in periodicals targeted to the legal profession that discuss what should be considered in screening cases.(5) However, I have been able to identify only one, 25 year old, published study, and it focuses solely on medical malpractice.(6)

A new survey of contingency fee practitioners in Wisconsin provides the first systematic data on case screening patterns and practices. A total of 511 Wisconsin attorneys whose practices involve contingency fee work responded to this survey.(7) The survey covered a variety of aspects of contingency fee work, but the focus here is on screening of potential cases. This survey is part of a larger project on contingency fee legal practice in Wisconsin (see sidebar).

Contact Volume

Among the 455 survey respondents who provided usable data, the number of potential clients contacting the respondents or their firms ranged from 1 to 5,000.(8) (The number of clients accepted ranged from 0 to 600.) Table 1 summarizes the volume of contacts. Overall, these lawyers contacts from 53,584 potential clients. Seven respondents reported more than 1,000 contacts; this represents an average of about least 20 contacts (or more) a week, or four per day; at the other extreme, almost half (47%) report about two or fewer contacts per month (25 or less over the year).

One problem with the figures above is that some lawyers (about 10% of the 455 respondents) work in firms where case screening is handled on a firm-wide basis while other lawyers handle screening on an individual lawyer basis either in a firm or as solo practitioners. Preliminary analyses indicated that the general pattern in the results is the same regardless of whether the "firm-level" screeners are included or excluded. Consequently, because I will be considering individual characteristics as control variables in the analysis that follows, much of the following discussion focuses only on those lawyers in firms where individual lawyers are responsible for screening cases (footnotes will indicate where this is not the case). Looking only at those respondents who were individually responsible for case screening, the range of contacts is from 1 to 2,500; aggregating across all of these respondents, there were 40,518 contacts. The right two columns of Table 1 break these figures down by contact volume.

Selectivity in Taking Cases

How many the potential cases do lawyers accept? The number of clients accepted ranged from 0 to 600. There are at least two ways to convert the number accepted into acceptance rates. First, we can look at it from the perspective of the lawyer by asking what is the typical proportion of potential cases lawyers accept? This involves looking at mean or median acceptance rates across the sample of lawyers. Alternatively, from the viewpoint of the potential client, one can ask what is the likelihood that a randomly selected client calling a randomly selected lawyer will have his or her case accepted by that lawyer? To look at this, the best estimate involves aggregating across lawyer: adding up the number of cases accepted across all of the lawyers and the number of contacts received across all of the lawyers and dividing the two figures. In the discussion that follows, I present both types of estimates.

Overall, lawyers reported accepting cases from a mean of 46% (median 45%) of the potential clients who contacted them; the "midspread"(9) is 24% to 67%. Aggregating across the 455 lawyers, the lawyers accepted 16,519 (of 53,584) cases for an acceptance rate of 31%. Eliminating the seven respondents reporting 1,000 or more contacts, the mean acceptance rate is 47% (median 48%). These firms had contacts from 36,884 potential clients; they accepted 15,224 (41). As shown in Table 2a, there appears to be a fairly clear linkage between volume and selectivity. For those lawyers or firms receiving about one and a half or fewer contacts per week, the acceptance rate tends to be on the order of 50%; for those with more than one and a half and up to about 20 contacts per week (1,000 cases per year), the acceptance rate is a little under 40%; for the very high volume practices with more than 20 contacts per week, the acceptance rate drops off sharply to under 10%. Table 2b eliminates those respondents who work in firms which screen at the firm level. Other than reducing the number of higher volume respondents, the general pattern is essentially the same.

Correlates of Selectivity

There are a variety of variables which might influence acceptance rates: gender, experience, specialized nature of practice or firm, size of community. Because these might be correlated with contact volume, which the analysis above shows is clearly related to acceptance rate, I controlled for volume using two categories: low (75 or fewer contacts over the year) and medium (76 to 1,000 contacts). Eliminating the respondents in firms where screening is a firm-level function (which is necessary to look at any types of attorney characteristics), there are only four high volume (over 1,000 cases per year) respondents; I report the characteristics of those four respondents in the right-most column of Table 3.

The table shows acceptance rates controlling for ten different variables which one might expect to be related to how selective lawyers are in their acceptance of cases. I show statistically significant (at the .05 level of better) differences in bold italics.(10) Of the 20 comparisons shown in the table (10 comparisons done separately for low and medium volume practices), only six achieve statistical significance. Five of these are for the low volume lawyers:

• Women are more selective than men.
• Lawyers with more than 20 years experience are less selective than those with 20 or fewer years of experience.
• Lawyers describing their practice as primarily personal injury plaintiffs or general practice are less selective than are those who are primarily personal injury defense or other types of litigation.
• Lawyers whose contingency fee caseload is predominantly (50% or more) personal injury are less selective than those whose contingency fee caseloads are not dominated by personal injury cases.
• Selectivity decreases as the lawyer's dependence (in terms of income) on contingency fee work goes up.
• The only statistically significantly difference among the medium volume lawyers is that those whose caseloads involve 10% or more medical malpractice work are more selective than are those with little or no medical malpractice work. The absence of statistically significant patterns for medium volume respondents probably reflects the smaller number of such respondents compared to the low volume respondents; the directional patterns for some of the variables mentioned above were similar for medium volume lawyers even though the differences did not achieve statistical significance (e.g., the percent of cases accepted by men was higher than that by women, selectivity decreased as dependence on contingency work increased, and selectivity increased as non-PI contingency fee work increased).

The only discernible pattern in Table 3 is what appears to be an inverse relationship between selectivity and dependence on contingency fee (particularly personal injury) work: selectivity decreases as dependence on contingency fee cases increases. That is, lawyers who have substantial work that is not contingency fee based are able to be more selective in the cases that they accept. An established lawyer whose practice is largely noncontingency fee work can afford to be more selective. For these lawyers, the question is why take contingency fee cases that will be less lucrative than hourly fee work? One lawyer I talked to put it bluntly when he said, "I'm in contingency fee cases to beat my hourly rate." What the table does not show is that at least some of those who are most dependent upon such work are among the most selective in which cases they take. At the extreme is the lawyer who reported that his/her firm as had 5,000 contacts per year but accepted only 200 cases (an acceptance rate of 4%). However, most lawyers with substantial dependence upon contingency fee cases are not able to be as selective as either those lawyers or firms who "cherry pick" among potential cases or those who combine contingency fee work with hourly fee work.(11)

The most striking aspects of Table 3 are the relative lack of variation (taking into account the volume variable) and the absence of any identifiable group that accepts "most" of the cases that are screened. In fact, almost no categories accept substantially more than 50% of the cases screened, and the highest acceptance rate is only 60%. Simply stated, contingency fee lawyers generally turn down at least as many cases as they accept, and often turn down considerably more than they accept.

When Are Cases Declined?

Typically, the first contact between lawyer and client comes in a phone call initiated by the potential client.(12) This first call is extremely important because, on average, lawyers declined about half (51%) of all eventually declined cases based on this initial telephone contact (see Table 4a). Looking at aggregate numbers of cases declined (i.e., adding up all of the cases declined across the respondents), the rate of decline based on initial telephone contacts increases as call volume goes up (see Table 4b), with the low volume lawyers declining 53% of the total declined after the first phone conversation, medium volume 60%, and high volume 83%. Lawyers drop about 5% of the declined cases after a potential client fails to keep a first appointment. Most of the rest of the cases declined were declined after the first in person meeting. Thus, lawyers make relatively quick decisions on most potential contingency fee cases.

Screening Criteria

To what degree do lawyers have explicit criteria, such as amount of damages, for accepting cases? I asked the respondents whether there was a minimum damage figure for each of three types of cases: auto accidents, medical malpractice, and products liability. Most lawyers (94%) provided a response for auto accidents, but only 43% stated that there was a minimum for such cases (the median minimum $5,000). Substantially fewer responded regarding medical malpractice (61%) or products liability (69%), reflecting that many lawyers did not handle these kinds of cases. Of those who did respond, a higher percentage reported a minimum damage figure: 61% for medical malpractice (median $100,000) and 60% for products liability (median $75,000).

Finally, why do lawyers decline cases? I asked lawyers what percentage of cases declined were due to lack of liability, due to low damages or inadequate fee potential, due to both lack of liability and low damages, due to falling outside the area of the lawyer's practice, and due to other reasons. Not surprisingly, lack of liability and inadequate damages (singly or together) are the dominant reasons for declining cases, accounting for about 80% of the declined cases. Table 5a shows the aggregated figures for all lawyers, for all lawyers omitting the four high case volume lawyers, low volume lawyers and medium volume lawyers. Lack of liability alone accounts for the largest proportion of cases declined, particularly for those lawyers with a higher volume of contacts from potential clients. Excluding the high volume lawyers (who are not shown as a separate column in Table 5 because there are only four such respondents(13)), only about a quarter of the declined cases were solely due to low damages.

One question that this analysis cannot answer is whether lawyers turn away too many or too few cases, or whether they turn away the right cases. Many people have lamented the supposed growth in "litigiousness" among the American population, but it is not necessarily self-evident that Americans are too eager to seek compensation when under our law compensation is due.(14) We might use Elihu Root's injunction, "about half of the practice of the decent lawyer consists in telling would-be clients that they are damned fools and should stop," as one possible measure. If we take "half of the practice," to refer to the proportion of potential cases accepted, then most contingent fee lawyers achieve this measure of decency.

While my survey data do show variations among lawyers in selectiveness, they do not provide direct information on variations among types of cases. My observations, however, make it clear that lawyers are more selective in some types of cases than in others. To some degree this reflects the simple fact that injury victims are more knowledgeable about whether they are entitled to compensation for some types of injuries (auto accidents) than for others (slip and falls, medical malpractice, etc.). I spent one month observing in a law office during the middle of a Wisconsin winter; during that time, there were several significant snow falls, and I listened in on a number of calls from persons who had slipped on snow or ice. Most of these cases were turned away simply because Wisconsin law allows property owners a period of time to remove snow or ice from sidewalks before they become liable for injuries arising from falls on their property.

Medical malpractice is one of the areas most talked about as needing reform to provide relief to medical providers from increasing law suits. During three months of observation in three different law practices, I saw lawyers dealing with contacts from 14 potential medical malpractice clients; only one of these contacts resulted in a retainer being signed, and it did not involve the potential malpractice aspects of the case. Lawyers are extremely cautious in accepting medical malpractice cases, and the lawyers I observed spent a lot of time explaining to these potential clients why their negative medical outcome did not constitute malpractice, or the difficulty in establishing that it did arise from malpractice, or that even if it was malpractice, the ultimate medical outcome was probably not affected by the error (and the interim consequences did not give rise to damages that made pursuing the matter financial attractive). For example, one potential client had gone in for a surgical procedure to correct a swallowing problem that involved insertion of an instrument down his esophagus. During the procedure the esophagus was injured necessitating surgery through the chest. The potential client was upset because (1) he had been in substantial pain in the recovery room and there was a delay in realizing the problem with the esophagus, and (2) the recovery from the more major surgery was several months longer than it would have been from the simpler procedure. As it turned out, a torn esophagus was a known risk of the simpler procedure (and the client has been warned of that risk as indicated by a signed informed consent). Furthermore, there was a significant chance that a more invasive procedure (through the chest) would have been needed even without the injury to the esophagus because the simpler procedure was substantially less than 100% effective. In this case, the lawyer explained to the potential client that the physician who conducted the original procedure probably had not committed medical malpractice (even if the recovery room staff had been slow to realize that there was a problem), and that there would be questions about damages because the final result was very good (i.e., the original problem had been corrected) and there were no residual problems from the chest surgery.

One of the standard laments of proponents of change is that lawyers pursue too many cases that prove to be unfounded, and this is particularly the case for medical malpractice cases. Is this the result of poor screening decisions, or is there another explanation? Michael Saks has nicely laid out the dilemma that arises in assessing case selection in medical malpractice litigation.(15) Injuries during medical treatment are common, but only a very small fraction of those injuries is attributable to negligence. The injured persons are not very good at determining whether their injuries are due to negligence, and so large numbers of persons whose injuries are not due to negligence are likely to seek legal counsel. Let us assume the following:

• 100,000 persons experience injuries during medical treatment;
• 1,000 (1%) of these were injured due to negligence and the injuries of the other 99,000 were not due to negligence;
• 500 (half) of those injured due to negligence seek legal counsel as do 10% (9,900) of whose injuries were not a result of negligence;
• the lawyers make the correct decision in 90% of the cases;
• this means that 450 (90% of 500) of the cases accepted did involve negligence and 990 (10% of 990) did not.

The bottom line is that even though those who actually experienced medical negligence are much more likely to seek legal counsel than those who were not so injured (50% versus 10%), and lawyers make the correct decision 90% of the time, more than two-thirds of the cases pursued did not involve negligence. By these figures, right around 1% of injuries not arising from medical negligence leads to legal claims, a percentage which is about the same as the rate of negligence among medical injuries.(16)

This research makes it clear that contingency fee lawyers do operate as gatekeepers: they turn away substantial numbers of potential clients, most often because those potential clients simply do not have a basis for pursuing the case they bring to the lawyer. The contingency fee structure means that lawyers carry out this function in large part as an exercise in economic self interest. That is, lawyers try to choose cases that they believe will yield fees at least equal to what they could earn from either nonhourly fee cases or from other contingency fee cases. In a sense, the lawyer is making decisions about which cases to include in his or her portfolio of risks. The lawyer knows that some cases will fail to yield a fee sufficient to compensate for the expenses of those cases while other cases will yield a profit which will at least offset the "unsuccessful" cases,(17) and hopefully go beyond that to yield a profit across the entire portfolio.(18) Some lawyers explicitly include in their portfolios a mix of hourly and contingency fee cases on the expectation that the former will cover the overhead and the latter will produce the profits; others have a mix of cases simply because that is the nature of the clientele they are able to attract.

There are other types of potential gatekeeping mechanisms. One would be to ban the contingency fee and require that potential litigants be prepared the bear the cost of seeking recompense for injury or other losses. This shifts the risk burden from the lawyer to the potential litigant. In some countries where contingency fees are prohibited, there have grown up systems of legal insurance to shift the risk.(19) Such a system would disadvantage those at the bottom of the economic ladder because "legal expense insurance" would probably be one of those "extras" which would be the first item to be cut from a tight budget. Another way of gatekeeping, which also focuses on the incentive structure/risk preferences of the potential litigant is fee shifting, where the loser of a lawsuit is required to pay the winners costs. This system serves to disadvantage the middle class which has to choose between foregoing compensation or putting at risk other resources (e.g., savings); those at the bottom of the economic ladder are essentially "judgment proof" and have nothing to lose through the fee shifting mechanism.(20)

Interestingly, in 1995 England, which had both a ban on contingency fees and a fee shifting regime, relaxed the former by permitting a variant which the government preferred to call "conditional fees."(21) At that same time, the Law Society (the professional organization of solicitors) came forth with an insurance scheme to ameliorate the problems created fee shifting (i.e., the client can be insured against the risks of the fee shifting rule).(22) The result is gatekeeping may be shifting from incentives focused on the potential litigant to incentives focused on the solicitor. It is too early to know what impact this has had, or will have, on the English court system.

To remedy the lack of good contemporary data on screening of potential clients, and the more general dearth of systematic information on contingency fee legal practice, the Professor Kritzer undertook a multi-faceted study of contingency fee practice focusing on lawyers in Wisconsin. This study, which is funded by the National Science Foundation, involved three separate data collection components.

The first was a mail survey of Wisconsin contingent practitioners; the sample for the survey was based on the State Bar of Wisconsin's Litigation Section mailing list. The survey was executed for Professor Kritzer by the University of Wisconsin Survey Center in a manner designed to preserve the confidentiality of the respondents. A total of 511 useable questionnaires were returned; this constituted a 48% response rate. Most of the information in this article draws upon the survey data.

The second aspect of the data collection involved full time observation of lawyers at work by Professor Kritzer. He spent approximately one month in each of three different contingent fee practices. While present in each practice, his formal status was that of a paralegal, and in each practice he did provide some assistance with research and other activities. All of the practices were within a one hour driving distance of Madison. The practices were chosen to reflect different types of settings: Two were practices specializing in contingent fee cases, one relatively high volume and one low volume; the third was a mixed trial or court practice in which contingent fee work constituted about 20% of the lawyer's work. The specific practices were chosen based on a combination of personal contacts and suggestions by persons knowledgeable about local practitioners. Only one practice that was originally contacted declined to participate.

The third part of the data collection consisted of semi-structured interviews with approximately 50 contingent fee practitioners, insurance defense counsel, and current or retired insurance adjusters. The plaintiffs' lawyers interviewed were chosen by selecting from directories and yellow page listings covering an area within about 100 miles of Madison. Defense counsel or firms were identified by the plaintiffs' lawyers, and, insurance adjusters were identified by defense counsel.

1. The original version of this paper was prepared for presentation at 1996 Annual Meeting of the American Political Science Association, San Francisco, California, August 28-September 1. A short version of the analysis presented here previously appeared as "Holding Back the Floodtide: The Role of Contingent Fee Lawyers," 70 (3) Wisconsin Lawyer 10. The research reported in this paper is supported by National Science Foundation Grant No. SBR-9510976. Research assistance was provided by J. Mitchell Pickerill.

2. Mary Ann Glendon, A Nation Under Lawyers: How the Crisis in the Legal Profession Is Transforming American Society (Cambridge: Harvard University Press, 1994), quoted at pp. 37 and 75.

3. Defenders of the legal profession would also cite professional norms, or court rules (such as federal Rule 11 which provides for potential sanctions against lawyers for bringing meritless cases), as providing further motivations for rejecting cases.

4. Mark Crane, "Lawyers Don't Take Every Case," National Law Journal (January 25, 1988), pp. 1, 34.

5. See, for example, William Trine and Paul Luvera, "Pros and Cons of Accepting a Case," 16 Trial 16 (May, 1994).

6. Stephen C. Dietz, C. Bruce Baird, and Lawrence Berul, "The Medical Malpractice Legal System," in Appendix: Report of the Secretary's Commission on Medical Malpractice (Washington: Department of Health, Education and Welfare, 1973 [DHEW Publication No. (OS) 73-89]), pp. 95-101. This study found that, overall, attorneys accepted only about one in eight potential medical malpractice clients who contacted them with possible claims

7. The response rate was approximately 48%.

8. Twelve respondents who provided information on screening are omitted from this total because their information was for firm-level patterns and I had multiple respondents from their firm. Including them would have meant double counting. My previous reports of analyses of these data did not exclude these respondents (which is why some specific figures here differ from previously reported figures).

9. The midspread is the distance from the first quartile to the third quartile: 25% of the lawyers reported accepting 23% or fewer of the potential clients and 25% reported accepting 67% or more.

10. As an initial comparison, I did a t-test between the acceptance rates for low and medium volume lawyers; the test was significant (t=3.72, p< .0001).

11. One minor pattern is the difference between male and female attorneys, with females accepting a lower percentage of cases. Additional analyses indicated that this was probably not simply a result of women working in practice settings where the percent of cases accepted was lower. Why women tend to accept a lower percentage of cases than do men is not clear.

12. A very small number of lawyers or firms initiate the contact by a post-incident mailing; only eight of the respondents indicated that any of their clients were obtained as a result of direct mail contacts.

13. These four lawyers report declining a total of 7,145 cases, 68% solely on liability grounds, 7% due to low damages, 5% due to a combination of low damages and lack of liability, 12% because they were outside the lawyers' areas of practice, and 8% for other reasons).

14. See Richard L. Abel, "The Real Tort Crisis -- Too Few Claims," 48 Ohio State Law Journal 443 (1987).

15. Michael J. Saks, "Do We Really Know Anything About the Behavior of the tort Litigation System--and Why Not?" 140 University of Pennsylvania Law Review 1147 at 1193-1196 1992); the numbers I have used differ somewhat from those used by Saks, but the pattern is the same (although the numbers Saks uses produce an even more extreme result).

16. See ibid., p. 1196n150.

17. I put "successful" in quotations to indicate that I mean something more than simply winning or losing. A case may be unsuccessful if no recovery is obtained, if an unexpectedly small recovery is obtained, or if the costs of obtaining the recovery (i.e., the amount of lawyer time) are much higher than expected. For more on these various risks see Herbert M. Kritzer, "Rhetoric and Reality . . . Uses and Abuses . . . Contingencies and Certainties: The American Contingent Fee in Operation," DPRP Working Paper, Institute for Legal Studies, University of Wisconsin, 1996.

18. For an analysis of the returns of contingent fee practice based on the Wisconsin survey, see Herbert M. Kritzer, "The Wages of Risk: The Returns of Contingency Fee Legal Practice," DPRP Working Paper, Institute for Legal Studies, University of Wisconsin, 1996.

19. This is perhaps most developed in Germany; see Erhard Blankenburg, "Changes in Political Regimes and Continuity of the Rule of Law in Germany," in Herbert Jacob, Erhard Blankenburg, Herbert M. Kritzer, D. Marie Provine, and Joseph Sanders, Courts, Law & Politics in Comparative Perspective (New Haven: Yale University Press, 1996), p. 298.

20. See the recent study of fee shifting in Alaska (the one state which has a blanket fee-shifting rule, although limited is degree), Susanne Di Pietro, "The English Rule at Work in Alaska," 80 Judicature 88 (1996). A similar pattern was reported when Florida briefly imposed a fee shifting rule in medical malpractice cases; see Edward A. Snyder and James W. Hughes, "The English Rule for Allocating Legal Costs: Evidence Confronts Theory," 6 Journal of Law, Economics, and Organization 345 at 356 (1990).

21. The American contingency fee is a commission system which takes on the contingency element through the fact that 33% of nothing is nothing. The English "conditional fee" system allows the solicitor to add a bonus to his or her usual fee (of up to 100%), but this fee is not a direct function of the amount recovered; see Herbert M. Kritzer, "Courts, Justice, and Politics in England," in Herbert Jacob, Erhard Blankenburg, Herbert M. Kritzer, D. Marie Provine, and Joseph Sanders, Courts, Law & Politics in Comparative Perspective (New Haven: Yale University Press, 1996), pp. 136.

22. Ibid., p. 147.