Determinants of institutional investor activism: a test of the Ryan-Schneider model

Journal of Managerial Issues, Summer, 2009 by Michael J. Rubach, Terrence C. Sebora

Definition of Variables

For this study, shareholder activism encompasses both traditional and relationship mechanisms with boards of directors and top management teams. The dependent variable is the practice of shareholder activism. To measure the use of traditional influencing mechanisms, the executives of the institutional shareholders were asked whether during the last calendar year their organizations performed any of the following activities: voted with a proponent of a proxy contest, became a party to a shareholder derivative suit or shareholder class action, filed a shareholder proposal, prepared a list of under-performing firms in its portfolio (targeting), or performed any other activities which the executives would consider to be the practice of shareholder activism. In addition, the executives were asked whether during the last calendar year their organizations made contacts with board members and/or top management team members of any firms in their portfolios to discuss the performance, business strategies and/or corporate governance of that particular firm. Consistent with what might be expected, the data disclosed that shareholder activism is not widespread--about 10% of the respondents practiced traditional influencing mechanisms, about 10% practiced relationship investing with boards, and about 30% practiced relationship investing with top managers. Shareholder activism data were dichotomized: "1" = performed activism and "0" = did not perform activism. The key question of this research was not how many are active but rather why do firms become active at all.

The independent variables are the factors that distinguish institutional shareholders: size, investment time horizon, performance expectations, pressure sensitivity, legal restraints, and portfolio management (internal versus external). For purposes of this study, size is a measurement of the value of the assets in an institution's portfolio at market value (Romano, 1993). The natural log was used to measure portfolio size because this transformation of the data reduced the effects of the large variances in the sizes of the institutional shareholders. For purposes of investment time horizon, institutions were classified as captive or mobile based on whether their principal constituents legally have the ability to move their invested funds (Useem, 1996). The investment time horizon variable was also operationalized as a dichotomous variable: captive = 1 and mobile = 0. Brinkley et al.'s (1988) typology for institutional investors was used to determine pressure sensitivity: "1" for pressure-sensitive, "2" for pressure-resistant, and "3" for pressure-indifferent. The respondents were asked to self-type their performance expectations (see Shortell and Zajac, 1990). Performance expectation was operationalized as a dichotomous variable: "1" = concerned exclusively with financial returns and "0" = concerned with mixed financial and non-financial performance. Portfolio management was measured as a continuous variable: what percentage of the institutional shareholder's portfolio is managed internally.