Networks, Information Transfer, and Status Conferral: The Role of Social Capital in Income Stratification among Lawyers
2009; Taylor & Francis; Volume: 50; Issue: 1 Linguagem: Inglês
10.1111/j.1533-8525.2008.01133.x
ISSN1533-8525
Autores Tópico(s)Sports, Gender, and Society
ResumoAbstractThe focus of this article is to examine how and to what extent social networks serve to alleviate information problems on both demand and supply sides of the market, and how this mechanism contributes to income inequality among legal service providers. The empirical context of this study is a random sample of urban lawyers in Chicago. The findings indicate that a lawyer benefits not only from having access to social capital that provides timely and novel information related to solving work-related tasks, but also from the endorsement by high-status network partners. In addition, empirical analyses reveal that the returns on ties to high-status others, net of control, and other social capital variables vary according to the level of client uncertainty in the market. ACKNOWLEDGMENTSI gratefully acknowledge Edward Laumann, Mathew Bothner, Damon Phillips, Ross Stolzenberg, and Toby Stuart at the University of Chicago for their helpful feedback on earlier drafts of this paper. I also thank John P. Heinz and Robert Nelson at Northwestern University for their critical input. This research was supported by the American Bar Foundation.Notes1 See, for example, Stuart and Ding's (2003) Stuart, Toby E. and Waverly Ding. 2003. "Why Do Scientists Become Entrepreneurs?" American Journal of Sociology 112:97–144.[Crossref], [Web of Science ®] , [Google Scholar] study on ties to high-status actors and academic entrepreneurship.2 This differentiates the network conception of status from the asocial economic conception of reputation, as reflected in, for instance, Williamson's (1991) Williamson, Oliver E. 1991. "Comparative Economic Organization: The Analysis of Discrete Structural Alternatives." Administrative Science Quarterly 36:269–96.[Crossref], [Web of Science ®] , [Google Scholar] writing. For Podolny and other network scholars, "an actor's status derives not from past demonstrations of quality; it derives from the status of the actor's exchange partners" (Podolny 1993 Podolny, Joel M. 1993. "A Status-Based Model of Market Competition." American Journal of Sociology 98:829–72.[Crossref], [Web of Science ®] , [Google Scholar]:460).3 This is because "status lowers the transaction costs associated with the exchange between buyer and seller. Implicit and explicit promises of a higher status producer regarding product quality are more likely to be accepted; therefore, the higher status producer need not devote as much time or expense to convincing the buyer or relevant third parties of the validity of its claims" (Podolny 1993 Podolny, Joel M. 1993. "A Status-Based Model of Market Competition." American Journal of Sociology 98:829–72.[Crossref], [Web of Science ®] , [Google Scholar]:838). Also related to status, according to him, are the advantages of lower advertising and financial costs.4 In that piece, the authors explicitly conceive of ties to other lawyers as a source of social capital. That is, their focus is on the familiar story (à la Burt, Granovetter, and Lin) about the information and influence benefits embedded in social networks. Insofar as they talk about "status," it is in the specific context of the linkage between the prominent network alters and the quality of information and the extent of positive influence that can be gained for the focal actor. In Podolny's language, this is exactly about the network-as-pipes argument dealing with egocentric uncertainty. The additional argument and the supporting evidence presented in this article also highlight something that was not specifically addressed by Sandefur et al., namely, how networks reduce altercentric uncertainty and how this in turn affects lawyers' income levels.5 In their words, "New organizations are vulnerable because their participants are strangers. Efficient organizations require trust among members; and trust takes time to build. New organizations are also vulnerable because they have to create organizational roles and routines. Inventing and refining roles and routines take time and effort precisely when organizational resources are stretched to the limit" (p. 245).6 Barron, West, and Hannan (1994) Barron, David N., Elizabeth West, and Michael T. Hannan. 1994. "A Time to Grow and a Time to Die: Growth and Mortality of Credit Unions in New York, 1914–1990." American Journal of Sociology 100:381–421.[Crossref], [Web of Science ®] , [Google Scholar], however, show that the effect of aging on organizational survival disappears "once relevant conditions, such as size, have been considered" (p. 414).7 One of the anonymous reviewers suggested that I use overall legal experience, not organizational tenure, in order to keep solo practitioners in the regression models. But the correlation between age and experience is high in the data set. And since age is used as one of the two proxies for client uncertainty, I opted for the organizational tenure variable instead. There is also a more theoretical reason for excluding the solo category in estimating additional models. This is to demonstrate that returns on ties to notables matter even for firm lawyers who deal with relatively sophisticated corporate clientele, often consisting of in-house counsel members who have a great deal of knowledge about and experience in the legal services industry.8 Freeman's (1977) Freeman, Linton C. 1977. "A Set of Measures of Centrality Based on Betweenness." Sociometry 40:35–41.[Crossref] , [Google Scholar] betweeness centrality measure, as elaborated by Wasserman and Faust (1994) Wasserman, Stanley and Katherine Faust. 1994. Social Network Analysis. Cambridge, England: Cambridge University Press.[Crossref] , [Google Scholar], is defined as: for i distinct from j and k, where gjk(ni) is the number of geodesics linking the two actors (j and k) that contain actor i. Since this index's values depend on g, it is standardized. This measure thus varies from a minimum of zero, attained when ni falls on no geodesics, and a maximum of one when the ith actor falls on all geodesics, that is, (g − 1)(g − 2)/2, and is defined as: C′B(ni) = CB(ni)/[(g − 1)(g − 2)/2]. The Bonacich (1987) Bonacich, Phillip. 1987. "Power and Centrality: A Family of Measures." American Journal of Sociology 92:1170–82.[Crossref], [Web of Science ®] , [Google Scholar] power measure is formally defined as: , where α is a scaling factor, β is a weighting factor, and R is a relational matrix. Both centrality measures are calculated using the network software package UCINET (version 5).9 The Network Module in the GSS (1985) also contains data on the survey participants' network relations. The average number of people with whom they discuss "important matters" for the American population reported by Marsden (1987) Marsden, Peter. 1987. "Core Discussion Networks of Americans." American Sociological Review 52:122–31.[Crossref], [Web of Science ®] , [Google Scholar] is 3.01, which is comparable with our network size of 3. Although this number was set by the questionnaire design in our analysis, it conforms to a national average concerning people's "core discussion networks." In their study of social ties and voluntary organizational membership, McPherson, Popielarz, and Drobnic (1992) McPherson, J. Miller, Pamela A. Popielarz, and Sonja Drobnic. 1992. "Social Networks and Organizational Dynamics." American Sociological Review 57:153–70.[Crossref], [Web of Science ®] , [Google Scholar] report a smaller mean of 2.42 alters per respondent, which they see as consistent with the relatively smaller size of the city where their survey was conducted.10 According to Burt, this measure is expressed as: , q ≠ i, j, where pqj is the proportional strength of q's relationship with j, as pij is the proportional strength defined above of i's relation with j. This measure varies from a minimum of pij squared (j disconnected from all other contacts), up to a maximum of one (if j is your only contact). The sum of the previously mentioned equation across contacts j measures the aggregate constraint on your entrepreneurial opportunities within the network.11 There may be an important data-related reason for the statistical insignificance, as one of the anonymous reviewers pointed out. The respondents were asked to name only three colleagues with whom they discuss law-related matters on a regular basis. Given the nature of the question and the relatively small number of network alters named, it is likely that these colleagues constitute "strong" ties, meaning the members of the discussion network most likely know each other quite well. If this is true, then the information given by the respondents may be inadequate to fully capture the theoretical argument behind Burt's notion of structural constraint, which has to do with gap between nonredundant contacts.12 The empirical support, however, is conditional. When the solo practitioners are dropped from the regression analysis (Table 4), these coefficients do not reach the conventional level of significance. This suggests that lawyers who have a kin-related network structure are more likely to be solos. They are also less likely to have elite law school graduates as their network colleagues. Both observations make sense, since there is a great deal of homophily that drives the interaction among lawyers, which is largely based on their specific practice settings. Because solos occupy the lower stratum of the professional hierarchy, their choice in network partner may be relatively limited, relegating them, on average, to interact with those to whom they are tied by kinship. Elite partners are also difficult to come by for solo practitioners, since lawyers with relatively high status and human capital seek other similarly elite ("homophilous") colleagues.13 I am grateful to one of the reviewers of an earlier version of this manuscript for bringing this issue to my attention.14 As one of the reviewers aptly pointed out, given the cross-sectional nature of the data, this study cannot fully deal with the standard endogeneity and unobserved heterogeneity critiques of most sociological network analyses. The cross-sectional data used in this study, with no sources of exogenous variation, leave open the possibility that higher-quality attorneys in fact have high-status affiliates. Despite data limitations, efforts were made to specify the network causal effects by including a host of relevant controls.15 I also tested for the interaction effect expressed as a product of the mean-deviated values. These additional models were estimated because of the multicollinearity issue because of high correlations between the status variables and the interaction terms. In their study of the network status effect on firm growth in the semiconductor industry, Podolny, Stuart, and Hannan (1997) Podolny, Joel M., Toby E. Stuart, and Michael T. Hannan. 1997. "Networks, Knowledge, and Niches: Competition in the Worldwide Semiconductor Industry, 1984–1991." American Journal of Sociology 102:659–89.[Crossref], [Web of Science ®] , [Google Scholar] face a similar methodological problem and undertake the conventional practice of mean-deviating the variables of interest. I pursued the same strategy and found that interaction terms remain unchanged and the statistical result consistently supports the hypothesis concerning the network-status contingency argument.
Referência(s)