Ending the Exodus

A firm thought its problem was rude partners, but that was just a symptom. The real problem was attrition-and the solution was mentoring.

By: Peter M. Newton

"DO YOU DO ANGER MANAGEMENT workshops?" the caller asked. In this case, the anger in need of management was that of three bull elephant partners who were trampling associates and goring other partners. Members of the management committee, including my caller, did not want to confront the rogues directly because they were among the firm's best revenue producers. Besides, they had already proven themselves to be impervious to counsel from colleagues. The caller thought that an anger management workshop for all the partners might quiet the beasts without singling them out, so that they would tolerate being caged alone with a psychologist and civilized.

Since these partners had been at the firm for a long time and had long been impossible, I asked why they needed taming now. The answer was associate attrition. This older, successful midsize firm had experienced two consecutive years of 30 percent associate attrition, and new associates salaries were rising to levels it did not wish to match.

It was now clear that the problem with my caller's firm was not emotional incontinence among a few partners but associate attrition. Abuse by partners was only one possible explanation for it. We agreed that I would interview all the associates on a confidential basis to find out how they felt about working for the firm.

What I found was surprising: The single most important factor in determining associate loyalty at the firm was not compensation, length of workday, or interest level of assignments, but the existence of a good mentor relationship. Where I found an associate who evinced a strong attachment to the firm, and to the profession of law, I learned that I was talking to a person firmly ensconced in a good mentor relationship with a partner. Where I found a person whose relationship with the firm was weak or very ambivalent, and who was imagining leaving the field for another, I learned that the person had only a weak connection with a mentor or not at all.

In one remarkable case, the strength of a mentor relationship proved even more important than a pervasively abusive work environment. An associate told me of having begun in a different firm, one that had run amok. Business was bad, and the partners were attacking each other and the associates; the firm seemed to be going to hell in a handbasket. But it did have a program in which a mentor was assigned to each new associate every year, and this associate had begun to form a strong relationship with his mentor. "So you left because the place was falling apart?" I asked. "No," the associate said, "I left because they switched mentors on me at the start of the second year. If I could have kept my first mentor, I would have stayed."

The correlation between the strength of the mentoring relationship and firm loyalty was so strong that it over-whelmed gender as a factor in associate satisfaction. Most of the associates at my client's firm were women. Few of the partners were, so most of the mentoring was male-female. While women associates were concerned about the paucity of female partners, this did not prevent many of them from enjoying developmentally rich mentor relationships with men.

Similarly, the mentoring connection easily overpowered compensation. Associate pay at my client's firm was moderate, compared with other firms in the same city, but only the associates who lacked solid mentor relationships spoke wistfully about the money and excitement of working for dot-coms. Better-connected associates almost always considered themselves well compensated.

Finally, no theme emerged in which particular partners (such as the rogue elephants) were identified as destructive. What worried the management committee did not worry the associates. Partners whom some associates found hurtful were deeply trusted by others. Conversely, partners admired by most protégés were found neglectful by some. There was not a single partner whom some associate did not deem inadequate as a mentor (some simply because those partners were seen as too busy.) Only one partner had no acolytes, and he was one of the three originally identified by my caller, but even in this case, he was rarely singled out as a problem.

What the firm needed was not anger management but help in improving its mentoring program. The finding was especially dramatic because the interviews I conducted were largely unstructured. I did not begin with the expectation that mentoring would emerge as a key variable. I was quite prepared to find that it was workload, compensation, the nature of the work itself, or some other factor.

Despite the myriad differences between associates and employees generally, my findings about the preeminence of mentorial relationships in employee retention, are consistent with results concerning employee recruitment, satisfaction, and performance found in large-scale studies of businesses. Indeed, the weight of the evidence from my own and other research is so great that it leads me to the following axiom about law firms: Assuming sensible hiring decisions, an associate who enjoys a good mentor relationship almost never fails to become a competent attorney. Good mentoring diminishes attrition, strengthens loyalty, integrates the firm by uniting the generations, makes better lawyers, and creates an ongoing firm culture of generative learning. It helps the young to feel nurtured and the middle-aged fruitful.

Given the importance of the mentor relationship, it is good to be clear about it. During the last 30 years, the word "mentor" has become badly clichéd. From Coca-Cola to Wall Street to Jiffy Lube, mentoring programs abound, although their actual purposes vary significantly and occasionally involve little beyond keeping youthful fingers out of the till. The word is now commonly applied to diverse sorts of relationships that provide little of deep or enduring value. No distinctions are recognized or made between mentors and supervisors, coaches, patrons, role models, or gurus.

The original research that discovered the importance of mentors was done at Yale University during the early 1970s by my mentor, Daniel Levinson, and others, including myself. Our studies led us to distinguish the role of mentor from those other sorts of roles. A supervisory relationship, for instance, exists primarily to provide control and evaluation. A role model or a guru inspires emulation; no relationship beyond the imaginary is required. A patron is a person in a position of power who may be called upon to provide protective intervention in times of exigency, but who may have no ongoing work relationship with the beneficiary. Each may have mentoring aspects, but the essence of each is different.

Although the mentor relationship often contains aspects of these others, its primary task is the professional development of a novice by an expert in the same field. It is comprised of chemistry-mutual attraction-and shared work. Real mentoring is a two-way street and benefits both participants; its developmental value to the protégé is matched by its contribution to the older person. In the ideal, the mentor helps the protégé form a professional dream and dare to believe in it. Typically, the relationship lasts from two to seven years and ends when the protégé outgrows it.

Unlike education in the other two fields that I know well-medicine and clinical psychology-law students receive almost no applied training. Medical students are on the wards under direct, individual supervision in their second year and increasingly for the next six to eight. Graduate students in clinical psychology have clinical placements with individual supervision throughout their six predoctoral years. After graduation they must log 1,500 post-doctoral hours under individual supervision in order to take the licensing examination. This ensures little beyond a rudimentary level of competence, but it does mean that, faced with a clinical problem to solve, the young physician or psychologist has seen it or one like it before, and probably more than once.

In contrast, first-year associates are given tasks they've never done and are often placed under great pressure involving time and real-world consequences for clients, partners, and their own careers. Associates receive little guidance while they work, yet their finished products are scrutinized with excruciating assiduity and often found wanting. In these evaluations, associates experience a harsh confrontation with the firm's standards and values. They get an intimidating look at the gap-seemingly a chasm-that exists between their own current work and what is required, between their own professional selves and that of a mature lawyer.

This gap is best bridged by a mentor. Mentors criticize but also teach; demand but also empathize; exemplify excellence but also acknowledge their own failings. They evince an ongoing interest in the larger personal lives of their protégés and the difficulties of integrating law into their lives while sharing something about their own, providing a bearable preview of life after youth.

Implementation of a successful mentor program is complex but not impossible. The program should be headed by a partner who has benefited from a genuine mentor relationship-a truly generative person, not merely the last person out of the room when a volunteer was requested. The program director needs to know the gifts and frailties of the other partners; tact and judgment will matter greatly in this position. Heads of mentoring programs should be recognized and compensated for their efforts, and they need to be given the freedom to decide which partners should be offered to associates. Some very exacting but well-intentioned partners may be good for second-year associates but deadly for beginners. (One option is to put all first-year associates in the hands of more senior associates and assign partners only during the second year.)

Associates should be allowed to choose from among available partners and encouraged to consult with older associates in making their choices. Mentor assignments should be made for a minimum of two years, although associates should be give time opportunity to change mentors after six months.

Mentors should be evaluated yearly, in ways that protect the confidentiality of the associate and the dignity of the partner. Methods should be consistent with the culture of the firm. Questionnaires are not always preferable to narrative or even anecdotal reports gathered by the program director. Obviously, the level of trust that partners and associates have in the program director's integrity is a sine qua non of a successful mentor program.

Finally, successful mentors should be celebrated and compensated, poor performers privately counseled, and all participants given the benefit of training programs. This includes associates, who often need help in figuring out what is appropriate to expect from a mentor, how best to ask for it, and what to give in return. Mentors are not sent by God to love associates unconditionally, but are people with needs of their own, which must in some real measure be met-both by protégés and by the firm that connects them.

At the firm that I studied, a mentoring program was already in place, so the firm's managing partner asked me to conduct mentoring workshops for partners and associates in all the firm's offices. In those workshops, I sought to increase participants' understanding of the dynamics of mentoring and offered recommendations for both mentors and protégés. Although we know from research that this sort of intervention in an ongoing program can reduce attrition by as much as a third, it is still too early to know whether that will be the case in this firm. But already the sense of improved moral among the associates-the feeling that they are being heard and taken seriously-is palpable.

Dr. Peter Newton can be reached at (510) 521-3848 or at pmnewton@mac.com.

Reprinted with permission from the January 2001 edition of The American Lawyer. Copyright 2001 NLP IP Company.