against the controller because of their greater sophistication, on average, relative to other
coordinate more effectively among themselves, within certain legal constraints, which may give
them greater bargaining power against the controller.
Finally, I control for two target and deal characteristics: whether the target is incorporated in
Delaware, and whether the controller is a founder or family group. On the latter, FAMILY is set
to 1 for targets in which the controller is a founder or family group. The intuition here is that
families may be less likely to “turn the screws” on the SC than a corporate controller, regardless
of the transactional form that is used.
All models are run as ordinary least squares (OLS) regressions. Because Cook-Weisberg
reported in Tables 3A & 3B.
[insert Tables 3A & 3B here]
Examining first offers, the TENDER coefficient is negative in Panel A, consistent with
Hypothesis H1, but not statistically significant. However, when final offers are examined, the
TENDER coefficient becomes negative and significant at 90% confidence using a 30-day
baseline trading price, and at 95% confidence using a 60-day baseline trading price. Taken
together these findings suggest that the most important (though not necessarily only) driver of
outcome differences between tender offer and merger freeze-outs arises from the bid increases:
specifically, controllers in merger freeze-outs make larger concessions to the SC than controllers
in tender offer freeze-outs.40 This finding generally supports Hypothesis H1.
40 In a prior version of this paper that was posted on SSRN, I directly modeled bid increases and reported results that are consistent with this conclusion.