form. The story in its strongest form, then, implies a circularity that is inconsistent with the
empirical evidence presented in this paper. Put simply, transactional form must be doing at least
some of the work in causing the differences in outcomes reported here.
Causation concerns with respect to choice of transactional form are less easily dismissed. It
is possible that a controller is more likely to hire experienced outside counsel when it knows that
it will execute its freeze-out via tender offer – that is, a client-driven effect rather than a lawyer-
driven effect. In order to attempt to isolate the causal chain running from choice of outside
counsel to transactional form chosen, I use a two-stage instrumental variable (IV) approach.
Following the methodology introduced by Maddala (1983:246) and illustrated by Comment &
Schwert (1995), I first run a probit regression to predict whether the controller will choose
outside counsel that has substantial M&A experience. In the second-stage regression I replace
the dummy variable EXPERIENCED with predicted values from the first-stage model. In these
regressions (unreported), I find that the coefficients for EXPERIENCED and
EXPERIENCED*CONTROL are consistently positive, as in Table 5, but they are not
statistically significant at any conventional level. I therefore cannot rule out the possibility that
the apparent connection between experienced outside counsel and transactional form is in fact
driven by client selection of outside counsel.
5. Case studies
A comparison of two recent and high-profile freeze-outs provides some texture to the
empirical findings presented in Part 4. In August 2004, Cox Enterprises (“CEI”) announced a
$8.3 billion merger freeze-out of the Cox Communications (“CCI”) minority shareholders. In
January 2005, News Corp. (controlled by Rupert Murdoch) announced a $6.0 billion tender offer