Solving the Legal Profession’s Diversity Problem

Here is a familiar fact pattern in large U.S. law firms.

  • Time 1. Partners come together and agree that diversity is part of their firm’s core values; they review the firm’s bleak statistics, particularly at the partnership level, and agree they can and will do better.
  • Time 2. Through significant time and expense, they successfully recruit a diverse class of incoming associates.
  • Time 3. A disproportionately large number of female and diverse associates leave the firm.
  • Time 4. The remaining associates eligible for partner are primarily white men.
  • Time 5. Partners come together and agree that diversity is part of their firm’s core values; they review the firm’s bleak statistics, particularly at the partnership level, and agree they can and will do better.

Why does this cycle repeat itself? As a long-time law firm researcher who has seen this cycle play out over several iterations, I can tell you that it is easy for a group of lawyers, especially those new to leadership, to convince themselves that they can solve the profession’s diversity problem through greater moral resolve. Yet, if the root causes are not moral in nature, we won’t make much progress.

In this article, I ask readers to consider the possibility that the profession’s lack of progress on diversity is a systems problem rather than a failure of moral resolve.

What does it mean to have a systems problem? Every firm has a system of recruitment, selection, development, feedback, evaluation, and promotion that enables law graduates to enter as legal novices and, through years of effort, acquire the skills, knowledge, and experience necessary to become partners. At most law firms, however, this system is driven more by tradition and past practice than science. Further, the system seldom places explicit or rigid demands on partner-owners because partner-owners prize their autonomy and are given the greatest rewards for bringing in business. To the extent the system relies on measurement, the quality of the data is uneven and under- analyzed. Stated another way, the “system” for creating successful lawyers and partners is not much of a system at all. And in this ignorance lies the cause of our diversity problem.

For the last several years, I have shifted my focus from academic to applied research. Although academic ideas can be elegant, compelling, and important, their major limitation is that we don’t really know if they will work in actual practice. Applied research attempts to sort this out, usually through social scientists hired by organizations that are hungry for a competitive advantage. The goal of applied research is to find solutions to important problems and then make them cheap and simple to implement. Law has a shortage of applied researchers, partially because the profession has been so prosperous for so long (what’s there to fix?) and partially because lawyers tend to be uncomfortable with data and statistics. Yet, these background factors are starting to change.

In this article, I am going to share what I have learned through my applied research as it bears on the problem of law firm diversity. The bottom line is that the problem is fixable. If we design and implement a better system, out the other side will flow successful diverse attorneys in roughly the same proportion as the number we managed to hire several years earlier.

Further, the stakes are hardly academic. Organizations with a reliable system for creating diverse lawyers will have a competitive advantage for attracting clients and the best entry-level talent. Likewise, esteem and accolades await the leaders who finally make a breakthrough on law firm diversity.

You Have to Start with a Theory

An intelligent system is invariably built upon a theory drawn from multiple sources. One high quality source is published empirical research. A second is one’s own professional work experience: “When I have tried X, Y usually happens” — so we rely on X. Finally, a subset of our theories will be based on pure reason: “Based on our collective knowledge and experience, this is the best approach for this problem.” Figure 1 is a summary of my own theory for creating high-performing partners.

In narrative form, I am saying that the creation of high-perform- ing partners is influenced by five factors: (1) aptitude, also known as cognitive ability; (2) motivation, which is primarily a function of values alignment between the lawyer and the substance of his or her work; (3) the type and quality of work experience that a lawyer receives during his or her early career; (4) the quality, quantity, and timeliness of training and feedback; and (5) the presence and quality of a mentoring or coaching relationship.

The model can also be broken down into selection and development components. A law firm optimizes elements (1) and (2) through a process of accurate selection at the point of hiring.

The less accurate the selection, the higher the lawyer attrition due to poor fit for aptitude and motivation. A firm can optimize (3), (4), and (5) by designing and implementing systems for professional development. The better the design and execution of the interconnected systems, the faster and higher the lawyer’s growth trajectory.

What is the relative importance of these factors? This is a good question that no one can answer with any degree of precision, primarily because we are in the early days of applied research within the legal profession and the required data have not yet been collected and analyzed. The best we can do is to start with a theory that is consistent with the data we do have and continuously improve our knowledge through measurement.

It has been my experience, however, that lawyers often have strong opinions on what does and doesn’t matter. These views on lawyer selection and development essentially create a series of default settings based on conventional wisdom and past practice. I have enough knowledge of the social science literature and enough experience doing sophisticated applied research in law firms to conclude that many of these default settings are wrong.

Below is a summary of what I know about each of the five components in my five-factor model. One by one, and cumulatively, these model components provide me with optimism that law firm diversity can be dramatically improved, particularly at the partnership level.

(1) Aptitude

Intelligence is an obvious baseline requirement for successful lawyers, including high-performing partners. Yet, what quantum of intelligence, or IQ, is needed to perform at a high level? There is a natural presumption that more intelligence is better. Thus, many legal employers favor highly selective law schools on the theory that these schools only admit students with high LSAT scores and strong undergraduate GPAs. If a candidate graduated from a non-elite law school, law school grades tend to become much more important in hiring decisions.

These academic filters tend to have a negative impact on the ability of law firms to recruit more diverse entry-level candidates. Yet, a more fundamental question is whether heavy reliance on academic proxies truly produces a better candidate pool. If lawyers and law firms operate on the presumption that the answer to this question is yes when the actual empirical answer is no, all sorts of negative consequences follow:

  • Partners and professional staff waste time and resources on selection criteria that don’t matter;
  • Higher quality candidates who excel on dimensions such as motivation and values alignments are never interviewed at all;
  • Developmentally rich work assignments are disproportionately allocated toward majority associates because, as a group, they tend to perform better on academic proxies.

The heavy emphasis on academic markers may be misplaced because law school graduates are already part of a heavily filtered population. To become a licensed lawyer, the typical path is to complete a four-year college degree, obtain an LSAT score high enough for admission into an ABA-accredited law school, complete three years of law school, and pass a state bar examination. This is what psychologists refer to as a “range restricted” population — compared to the general population, this group has very high cognitive ability. This fact raises a simple, testable empirical question: Do marginal gradations on a handful of academic measures reliably signal greater potential to become a high-performing lawyer?

Several years ago, this question was taken up by Professors Marjorie Shultz and Sheldon Zedeck of the University of California at Berkeley, and, remarkably, the answer was no. Drawing upon the methodology of industrial and organization- al (IO) psychology (Zedeck’s area of expertise), the researchers identified a set of 26 lawyer effectiveness factors (see Figure 2 next page). Behaviorally anchored rating scales (BARS) were then created to measure lawyer effectiveness on a 1 to 5 scale, with increments defined by specific, concrete examples of lawyer behaviors. The researchers then used the BARS to obtain peer and supervisor evaluations for over 1,100 law alumni of UC Berkeley and UC Hastings and approximately 200 UC Berkeley law students. In turn, these measurements were correlated with participants’ undergraduate GPAs, LSAT scores, and law school grades.

Figure 2. Shultz and Zedeck - 26 Lawyers Effectiveness Factors.

The results of the Shultz-Zedeck study suggest that academic factors are not very reliable proxies for future lawyering potential. Among the law school graduates in the sample, factors such as Analysis and Reasoning, Researching the Law, Writing, and Problem Solving showed modest, positive correlations with grades and LSAT scores (between 0.10 and 0.15, p > .05). Yet, LSAT scores and first-year grades were also negatively correlated at statistically significant levels with Networking (-.122) and Community Service (-.096). In the student sample, undergraduate GPA was positively correlated with no effectiveness factors but negatively associated with Practical Judgment (-.169), Seeing the World through the Eyes of Others (-.170), Developing Relationships (-.195), Integrity (-.189), and Community Service (-.152).

The Shultz-Zedeck study also documented numerous job- relevant markers of future success as a lawyer that are likely crowded out by over-reliance on academic proxies. For example, Shultz and Zedeck correlated lawyers’ BARS scores with performance on the Hogan Personality Inventory (HPI), an established, off-the-shelf personality assessment that has been validated on a large sample of knowledge workers. Overall, the HPI provides much stronger signals for lawyer effectiveness:

  • On the HPI Adjustment construct, which measures emotional stability and steadiness under pressure, alumni lawyer scores were positively correlated at statistically significant levels with 22 of the 26 effectiveness factors (ranging from .072 to .220) and negatively correlated with none.
  • On the HPI Prudence scale, which measures self-control and conscientiousness, alumni lawyer scores were correlated with 18 effectiveness factors (ranging from .071 to .189) and negatively correlated with none.
  • On the HPI Ambition scale, which measures achievement and leadership orientation, alumni lawyer scores were positively correlated with 14 effectiveness factors (ranging from .076 to .239) and negatively correlated with none.

These correlation patterns strongly suggest ample opportunities for law firms to engage in better selection by relaxing their emphasis on academic proxies and improving their focus on job-relevant behaviors. Moreover, on these broader measures of lawyer effectiveness, Shultz and Zedeck documented no performance gap based on race and gender.

These findings may surprise some law firm partners who are wedded to the idea that pedigree — perhaps the pedigree that they possess — is a strong proxy for lawyer potential. Yet, the Shultz-Zedeck findings are broadly consistent with what we have found on numerous client projects.

For example, I have been a part of numerous internal law firm studies designed to build success profiles based on résumés and transcripts. The purpose of these success profiles is to identify future high performers. One of the most persistent results across virtually all of these studies is that attendance at an elite law school is seldom a marker of future success and often a slight negative predictor.

Another persistent finding is that law school grades tend to predict future performance within a law firm. Yet, grades are a function of both aptitude and effort. Since law graduates at more elite law schools tend to have higher LSAT scores, and attending an elite law school does not predict future high performance, the predictive power of law school grades within law firms is probably attributable to higher levels of motivation.

In summary, the proportion of law school graduates with the requisite aptitude to become high-performing partners is probably larger than most law firm partners would tend to believe. Reducing a firm’s reliance on academic proxies will increase the number of candidates who could be considered for hiring, which in turn expands the number of diverse applicants who might be eligible for an interview.

(2)  Motivation

In my five-factor model for creating high-performing partners, my only other selection criterion is motivation. As noted above, I view motivation as primarily a values alignment between the lawyer and the substance of his or her work.

My theory runs as follows: High performance inside a law firm requires relentless focus on other people’s problems, typically legal problems mixed together with business, professional, and personal dimensions. Is a candidate motivated to solve these types of problems for 50+ hours per week? If the answer is yes, he or she will be sufficiently self-directed to take advantage of the developmental opportunities that the firm will provide. In- deed, it is through these opportunities that the lawyer becomes intelligent and capable. The reason is deliberative practice rather than native intelligence. If the answer is no, there is a values misalignment and the candidate is better off in a work environment where the quality and quantity of problems are a better fit.

The importance of motivation as a selection criterion flows from my applied work with law firms. In these projects, we frequently use an assessment tool called the Achievement Motivation Inventory (AMI). The AMI measures 17 dimensions of achievement motivation, which are grouped into three broad categories: (1) Ambition, (2) Self-Assurance, and (3) Self-Control. We also use another assessment called the Management Development Questionnaire (MDQ), which measures various work behaviors that map onto a competency model similar to the Shultz-Zedeck 26 effectiveness factors.

One of the most persistent findings we observe is that high- performing partners tend to score significantly higher on the AMI, particularly on dimensions related to self-assurance and self-control. The advantage of these higher levels of motivation is that the lawyer has the emotional and mental staying power to acquire, over a long period of time, job-relevant skills and knowledge. The advantage of this greater staying power can be observed on the scores of the MDQ. On a 200-point scale, the typical law student scores a 101 compared to 107 for an associate, 115 for a client service partner, and 123 for a high-perform- ing partner, mirroring the progression within a law firm.

Assessment tools like the AMI and MDQ are useful tools for professional and organizational development, but they should not be relied upon as a primary selection tool for entry-level employment. This is because high-aptitude candidates will try to figure out what the employers want to see and hear. A better alternative is a well-structured behavioral interview that asks questions that map onto a set of predefined skills and competencies necessary to be successful at the firm.

The core principle underlying a behavioral interview is that past behavior is a relatively good predictor of future behavior. If you are looking for someone with initiative, or practical problem solving skills, or team work, ask questions that elicit specific, concrete examples of these behaviors and score the candidates based on the quality and quantity of the examples they provide. The scores on these various job dimensions tend to cluster together, which I interpret as motivation to learn, improve, and perform at an overall high level.

I have helped construct these types of interview systems for several law firms. These projects typically include statistical analysis to monitor the impact on diverse candidates. Figure 3 is a chart that compares interview scores of entry-level candidates at one Am Law 200 firm over three years of recruiting (n = 350). Scoring is on a nine-point scale where scores of 7.0 or higher were usually necessary to receive an offer. Each interview was identical in format, structure, and content. The interviews were scored in a panel format by law firm partners. The typical number of interviewers was four.

Figure 3. Scores on Structured Panel Interviews at Am Law 200 Firm, by Gender and Minority Status.

The most heartening aspect of Figure 3 is that the familiar performance gap for female and minority candidates is not present. In fact, in this relatively large sample, female and minority candidates tend to do slightly better than their white, male counterpoints. These results are also consistent with the findings of the Shultz-Zedeck study. Simply put, if a law firm applies job-relevant criteria in a uniform, structured way to a diverse array of entry-level candidates, they should expect a roughly equal proportions of diverse candidates to receive scores in the “highly qualified” range.

Not surprisingly, candidates who have been put through this structured panel interview (SPI) process are much more likely to accept an offer of employment. As a result of the SPI process, the firm in Figure 3 increased its yield from 33% to 48% over a three-year period despite becoming more selective in making offers. Minority yield rates were significantly higher. Through a subsequent debriefing process, minority candidates told our researchers that they perceived the process as being more thorough and fair, thus making them more confident that the firm had an overall plan for their professional development.

(3)  Experience

Parts (1) and (2) of my model suggest that proper selection criteria and valid and reliable selection methods will yield a group of high-potential entry-level female and diverse lawyers in roughly the same proportion as the overall law school population. The next step is bringing that high potential to fruition.

There is strong empirical evidence documenting that the quantity and quality of work assignments are a major driver of professional development. Those who receive these opportunities early in their careers tend to thrive, as they can immediately leverage their enhanced skill set to obtain even better assignments in the future. Conversely, those who are assigned more repetitive tasks – often to serve the immediate needs of the firm, a partner, or a specific client – are often put at a permanent disadvantage, as they lack the skills, experience, and accomplishments to compete with lawyers of the same age and billing rate who received better early assignments.

The exact pattern described above was documented in a large 1,000-lawyer U.S. law firm. The research was done by Professors Forrest Briscoe of Penn State University and Katherine Kellogg of MIT and published in the American Sociological Review, the leading peer-reviewed journal in the field of sociology. Their core research question was focused on the circumstances under which a knowledge worker could use a work-family program that offered a reduced or flex-time policy (often to accommodate child care responsibilities) without it have a damaging impact on the worker’s long-term career progression and pro- motion prospects. The work context happened to be a law firm.

Through a carefully designed study covering 958 associates who entered the firm between 1997 and 2005, including 71 who participated in the work-family program, the researchers learned that program participants were more likely to fall behind in pro rata performance pay and more likely to exit the firm. This finding is after controlling for law school rank, race, gender, undergraduate grades, office location, and practice department.

Yet, the researchers also found that the negative outcomes associated with the work-family program were significantly reduced when the associates were exposed to high-quality work assignments during their first few months at the firm. For all associates, these types of early career assignments were associated at statistically significant levels with higher performance pay and longer tenure. Yet, for program participants, it had a significantly larger impact, essentially inoculating them from the negative consequence experienced by their peers. Observed Briscoe and Kellogg, “Exposure to powerful initial supervisors helps employees gain access to reputational-building project opportunities, which in turn allows them to build a significant track record with a wide range of supervisors and clients by the time they use the reduced-hours program.”

The implication of the Briscoe-Kellogg study is that high-quality work assignments are an essential ingredient to a lawyer’s career progression. An associate who has successfully climbed the first one or two rungs of this career ladder is both more visible and attractive to other partners. Under the free market assignment systems that so many law firms embrace as part of their culture, associates who get early breaks are also set up for future career-enriching opportunities, largely because more experienced junior associates help partners build their practices.

Now, what is the likelihood that these networks of developmentally rich early career assignments are blind to both race and gender? Our work with law firms has given us a window to explore this question. For example, we have built several data-driven success profiles of incoming associates. The concept is simple. Identify high-performing senior and mid-level associates and, using multivariate statistical analysis, work backward to identify entry-level attributes on résumés and transcripts that can be used to predict future success.

When possible, we like to include demographic information as a control variable to ensure that a negative race or gender impact is not baked into a firm’s success profile. Because the legal profession has such a bleak record on diversity, it should come as no surprise that a statistical analysis of associate records is going to document some unpleasant facts, such as lower evaluations and higher rates of attrition for specific subgroups. But these bad outcomes are not the same thing as discrimination, as they could result from poorly constructed systems.

When the data are available, we dig deeper and try to under- stand why females and non-white men are getting lower performance scores. To date, we have never encountered systemic racial or gender bias: partners, regardless of their race or gender or the race or gender of the associate being evaluated, tend to agree — “This associate is excellent,” “This associate is below average,” etc. Yet, in large datasets involving over 250,000 associate hours, we have observed very large gender- and race-based patterns in which junior associates gravitate toward supervisors who match their own race and gender. Although this dynamic may be comfortable for associates, their female or minority supervisors tend to control fewer important client relationships and thus have fewer developmentally rich assignments to allocate.

In contrast, white males who attend elite law schools are ideally situated to receive these types of opportunities because their com- fort zones overlap with the power center of most large law firms.

The longer law firms ignore the profound effect played by work assignments, the longer the profession will be plagued with a diversity problem. Again, this is not a problem of a moral deficit; it’s a systems problem. Yet, I am among the group of lawyers who believe that it is morally wrong not to fix this system. Moreover, it is just bad business.

(4) Training and Feedback

To reiterate my basic claims thus far, the formula for a high- performing partner is (1) a lawyer with a high cognitive aptitude who (2) possesses a high motivation to learn and be successful in a legal service organization and (3) is given the opportunity to do progressively more challenging work so as to signal his or her capabilities to colleagues and clients. Components (1) and (2) raise issues of selection. Component (3) is entirely developmental. Through work opportunities, law students and junior lawyers become someone who appears to others as smart or naturally gifted. Yet, few law firm partners understand that a system is at work. Hence, the system is poorly tooled and tends to produce results that are not in alignment with our professional values and business goals.

The fourth component of my model is training and feedback. Experience alone is unlikely to round out a lawyer’s professional development. For issues of quality, cost, or both, there will be times when skills and knowledge are best taught through formal training. Feedback is a communication loop that aids a lawyer in connecting together tacit information embedded in both lawyer training and experience, thus enhancing judgment and overall lawyer effective- ness. Feedback accelerates professional development. Feedback is also very expensive because it requires a practice master to closely observe performance and communicate subtle points in a manner that the less experienced lawyer is able to hear and absorb.

Because legal service organizations lack a systems perspective, they routinely confuse the cost of feedback, which tends to be short- term and personal, with the value of feedback, which is short-, medium-, and long-term and affects the competitiveness of the entire enterprise. Specifically, an organization composed of lawyers who are too busy to provide high-quality feedback to high-potential people is an organization with fewer high-performing lawyers. Further, the problem only compounds and worsens over time.

The best way to illustrate the importance of training and feed- back to professional development, including the large implications for female and diverse lawyers, it to relate a clear success story. In my research, the most compelling knowledge worker example I have come across is the Bell Labs study, which was written up nearly 20 years ago in the Harvard Business Review and expanded upon in a best-selling business book.

The basis for the study was an antitrust consent decree between the Department of Justice and AT&T, which forced divestiture of AT&T operating units and created the seven “Baby Bells.” One of the implications of the settlement was that AT&T would no longer have the monopoly profits to subsidize the famed

Bell Laboratories, which had funded the research of several Nobel Prize winners. Despite decades of success, there was no guarantee that Bell Labs could survive in the private sector as a pure science think tank.

To increase their odds of success, Bell Lab executives commit- ted the organization to an internal study that would enable them to identify top-performing engineers. The reasoning was simple: If the organization could hire and develop more “ten- or twenty-for-oners,” productivity would skyrocket and the organization would become a magnet for paid client work. To run this project, Bell Labs engaged IO psychologist Robert Kelley, whose prior work specialized in productivity assessments in the emerging “gold collar” sector.

In Kelley’s study, an engineer was designated a top performer if he or she was identified as such by peers, managers, and (eventually) the organization’s clients. Kelley polled managers and workers to generate theories of success based on various cognitive, psychological, and social factors. In turn, these theories were tested using a large sample of engineers and a two-day battery of tests designed to measure 45 alleged at- tributes of success. Yet, despite the tremendous expenditure of time and resources, Kelley and his colleagues came up empty: There was no appreciable relationship between status as a star performer and any of the cognitive, psychological, social, or environmental factors. Attempts to reanalyze the data were equally fruitless.

Puzzled by these findings, Bell Lab executives extended the study so that Kelley’s research team could generate new theo- ries of star productivity based on observation rather than self- reporting. For the next two years, Kelley and his researchers examined the work habits and strategies of Bell Lab engineers. At the end of this process, they identified nine work strategies that distinguished star performers from the middle-of-the-road engineers. In relative order of importance, they included:

  1. Taking Initiative. Top performers took responsibility above and beyond their stated jobs, volunteering for new activities and promoting new ideas;
  2. Networking. Top performers were deft at tapping into coworkers’ expertise and shared their own knowledge with those who needed it;
  3. Self-Management. Top performers were very good at regulating their own work commitments, time, performance level, and career growth;
  4. Perspective. Top performers understood their jobs within the larger context of the organization and could analyze problems from the viewpoint of customers, managers, and team members;
  5. Followership. Although perceived by others as leaders, top performers excelled at setting aside their own agendas and using their talents to help other leaders accomplish the organization’s goals;
  6. Teamwork. Top performers were more willing to assume joint “ownership” of goal setting, group commitments, work activities, schedules, and defusing conflict among group members;
  7. Leadership. Top performers had the ability to formulate, state, and build consensus on common goals and then work to accomplish them;
  8. Organizational Savvy. Top performers recognized and thus could navigate competing interests within the organization;
  9. Show-and-Tell. Top performers typically had the ability to present their ideas persuasively in written or oral form.

One of the most striking features of Kelley’s research was the propensity of average workers to draw the wrong lessons from the success of top performers. Average performers tended to invert the order of priority and thus focus on organizational savvy and show-and-tell, which they surmised was the key — based on the success of the stars — to impressing management.

Similarly, middle performers viewed networking as staying “in the loop” on office gossip and getting to know people who could help their careers. Top performers, in contrast, viewed networking as a bartering system in which the cost of admission was technical expertise and staying in the loop required a sincere commitment to reciprocity. Kelley reported that star performers got their phone calls returned faster than their middle-performing peers, who were typically receiving bad answers slowly — hardly a recipe for career success.

Yet, the most remarkable finding of the Bell Labs study was that the star performer work strategies were found to be teachable — an outcome verified using the controlled experiment methodology, which is the gold standard for empirical research. The study documented that engineers who received star performer training (one day per week for several weeks) tended to post statistically significant gains in productivity over the next year as compared to the control group that did not get the train- ing. Yet, the gains were the most dramatic among female and minority engineers — four times larger than for white males. In contrast, within the untreated (or control) group, female and minority engineers’ performance tended to deteriorate on several dimensions over the next several months.

According to Kelley, there are two main reasons why the produc- tivity levels of female and minority engineers disproportionately soared after receiving the star performer training. First, these engineers undertook proactive measures to break into knowledge networks that were based on expertise rather than gender or race (success factor #2). Second, with the benefits of the training, these engineers engaged in better self-management to deal with incoming requests from coworkers (success factor #3). In many cases, the purpose of the requests was to showcase the company’s diversity rather than tapping into the engineer’s developed skill set. When played out over several iterations, the disparities between white males and female and diverse knowledge workers gets larger, not because of a gap in innate ability, but because of a systems failure in training and feedback.

(5) Coaching and Mentoring

The fifth and final component in my model is coaching and mentoring. A strong coach and mentor is often the vehicle through which a young lawyer receives developmentally rich work experience (3) and high-quality training and feedback (4). Yet coaching and mentoring is its own freestanding component because when it is done well it becomes an intense personal connection where talented professionals choose to allocate their valuable time and resources toward the success of others. Conversely, understand- ing the nature of the investment being made, the person being mentored experiences a mixture of heightened motivation and gratitude that enables him or her to persevere through virtually any professional hardship in order to reach long-term goals.

One of the best examples of the power of mentorship is New York City business lawyer Walter Carter, who served as a mentor to many of the leading corporate lawyers of the early 20th century. Carter’s accomplishments on this front were chronicled in a 1954 book entitled Walter S. Carter: Collector of Young Masters. According to the book’s author, Otto Koegel, Carter’s gift was spotting promising young talent and bringing them along as corporate lawyers who were capable of counsel- ing executives of large financial and industry enterprises.

An appendix at the back of Koegel’s book is a folded poster with a family tree of Carter’s lawyer progeny. One of the first nodes on the family tree is Paul Cravath, who worked for Carter as a junior lawyer. The subsequent branches document Cravath’s departure and movement to a firm that would later become Cravath, Swaine & Moore, where Cravath designed and implemented the “Cravath system.” According to the firm’s history, the Cravath system is largely credited with the firm’s eventual leadership position among Wall Street firms. The firm history also cites Walter Carter’s training principles as the basis for the system. Other branches on the Carter family tree connect founders or leaders at many familiar powerhouse firms of the 21st century, including Milbank Tweed, Willkie Farr, Cadwalader, Shearman & Sterling, and Hughes Hubbard.

I have also observed something similar to Carter’s impact on future leading lawyers, albeit within the context of a government agency. Colleagues in the securities bar have observed the phenomenon of “Sporkin’s kids,” referring to the many influential lawyers who worked under Stanley Sporkin during his long and distinguished tenure at the Securities and Exchange Commission (SEC). Many of Sporkin’s SEC protégés lacked the pedigree of an elite law school, yet they went on to become some of the most sought after and influential securities litigation lawyers of their generation. They include Edward

Herlihy of Wachtell Lipton (George Washington Law), William McLucas of WilmerHale (Temple Law), and Ralph Ferrera of Proskauer (Cincinnati Law).

After two decades at the SEC, Sporkin became general counsel of the CIA and then a prominent federal judge. In preparation for writing this article, I contacted Judge Sporkin to ask him about this track record of mentorship. He commented that his philosophy was to look for intelligent young lawyers who would approach their jobs “with enthusiasm.” In Sporkin’s view, the law school attended was a poor proxy for these intangibles (Sporkin himself attended Yale). Further, according to Sporkin, it was critical that there be values alignment between the young lawyer and the mission of the agency. Otherwise, the lawyer could not keep up with the demands of working in his office. (Compare Sporkin’s observations to the Motivation factor outlined in this article’s five-factor model.) Judge Sporkin expressed gratitude for the lack of bureaucracy in the 1960s, 1970s, and 1980s, which enabled him to hire so much raw talent according to his own criteria. He related the story of meeting a young Ralph Ferrera, who pleaded with Sporkin for an opportunity to work at the agency. Sporkin lacked the budget to hire him, so Ferrera worked for free until a formal staff position became open. The rest, as they say, is history.

In my experience, law firms undervalue the importance of coaching and mentorship. Carter and Sporkin had the power to make these investments on their own. Yet, today’s modern law firm emphasizes the production of revenues. The cost of nonbillable time can be readily calculated; the same cannot be said, how- ever, about the value of nonbillable time. Partners who have given little thought to the power of professional development are most likely to resist large investments. They lack the systems perspective of Paul Cravath. I have studied lawyer development for over a decade. I think these partners are trading dollars for pennies.


The purpose of this article is to create a roadmap for solving the legal profession’s longstanding diversity problem. The solution is to end the moral handwringing and to create a system for se- lecting and developing lawyers. Yes, it will be expensive in time, money, and political capital, but not nearly as costly as wasting raw human potential. Glory, and possibly organizational riches, will accrue to the law firm leaders and general counsel who are brave enough and wise enough to demand that we go down this road. The time has come to fix this problem once and for all.

DiversityBill Henderson