Moving away from "Up or Out": Determinants of permanent employment in law firms

Law & Society Review, 1999 by Gorman, Elizabeth H

Independent Variables

I use measures that tap the three aspects of work complexity discussed above: the extent to which an establishment's work requires specialized knowledge, professional judgment, and skill in negotiating and advising. The measures used here are based on the answers of respondents practicing in each of 42 fields to three questions included a 1995 survey of a random sample of practicing lawyers in the Chicago area conducted by John Heinz and Edward Laumann (see Heinz et al. 1997; Heinz et al. 1998), replicating their well-known earlier study of Chicago lawyers (Heinz & Laumann 1982) (see Table 1). For each field, Heinz and Laumann calculated the percentage of respondents who answered with a "1" or a "2" on a five-point Likert-type scale. I transformed these field-level measures into establishment-level measures as follows. In the NALP Directory, each establishment lists its primary practice areas and the number of its lawyers practicing in each area. I collected data on up to 20 practice areas per establishment. For each practice area in each establishment, I identified the most similar of the 42 fields listed by Heinz and Laumann and assigned to that practice area the appropriate field-level measures. I then weighted each practice area score by the percentage of establishment lawyers practicing in that area and summed the weighted scores to obtain a weighted average measure for the establishment.

Although my data do not contain measures of the strength of establishments' actual client relationships, I obtained estimates by using field-level measures provided by the Heinz and Laumann 1995 survey. The survey asked respondents practicing in each of 42 fields to indicate the percentage of their clients with whom their relationships had lasted for 3 or more years. A field-level measure of the stability of client relationships is provided by the mean percentage reported in each field. The survey also asked respondents to indicate the total number of clients for whom they had performed services in the previous year. The median number of clients reported by practitioners in each field provides a field-level measure of number of clients per lawyer, and the reciprocal of this number (transformed into a percent age) provides a field-level measure of the median percentage of time and revenues associated with each client. Field-level measures for both client stability and client business volume were translated into estimated establishment-level measures by weight ing each practice area score by the percentage of establishment lawyers practicing in that area and summing the weighted scores. To avoid multicollinearity when the cross product of these variables is included in the model, both variables are centered around their means.13

In the NALP Directory, establishments provide a narrative statement describing their legal practices, administrative policies, and organizational values. I measured the strength of an establishment's commitment to collegiality by a dichotomous variable indicating whether or not the establishment represented itself in its narrative statement as valuing collegiality and striving to maintain collegial social arrangements. In coding the data, I attempted to capture the substance of a firm's statement, regardless of the presence or absence of the terms collegial or collegiality. Although the narrative statements typically attempt to present the organization in a positive light, they are surprisingly diverse in content and emphasis. If the narrative statements reflected primarily a public relations effort rather than the extent of an establishment's actual preference for collegiality, we might expect a coding of "1" for a high percentage of establishments. In fact, only 20% of establishments were coded "1" on this variable. Moreover, the collegiality measure is negatively associated with the presence of formal departments, the presence of a two-tier partnership, the employee-to-partner ratio, and status as a branch office, as one would expect if the measure is indeed capturing the construct of interest.