Explaining the premiums paid for large acquisitions: evidence of CEO hubris

Administrative Science Quarterly,  March, 1997  by Mathew L.A. Hayward,  Donald C. Hambrick

• E-mail
• Print
• Link

In model 3, we included the hubris factor, and this was positively associated with premiums. Overall, therefore, Roll's hubris hypothesis received considerable support. Three different sources of hubris -- recent organizational success, media praise for the CEO, and the CEO's self-importance -- as well as the hubris factor were significantly and positively associated with the size of acquisition premiums.

Model 3 also included the three variables measuring lack of board vigilance, none of which had main effects on premiums. Hypotheses 4, 5, and 6 predicted interaction effects, however, based on the argument that the manifestation of CEO hubris in large acquisition premiums would be accentuated when board vigilance was weak. To test these hypotheses, we estimated three regressions, alternately adding each (hubris factor x lack of vigilance) interaction term to model 3. The results for these three interactions are presented in Table 3.

Table 3 Acquisition Premiums: Interaction Terms between Hubris and Board Vigilance(*)

                               Interaction   Standard
Interaction term               coefficient     error     [R.sup.2]
Hubris factor x combined
CEO/chair                       .364(***)     .157       .367(***)
Hubris factor x % of
inside directors                .620(***)     .242       .374(***)
Hubris factor x smallness of
outside director holdings       .020          .029       .330(***)

(*) p [is less than] .10;

(**) p [is less than] .05;

(***) p [is less than] .01.

(*) The results are for the addition of respective interaction terms to model 3; reported are coefficients, standard errors, and R squares.

Two results for the interactions were significant: those involving combined board chair and CEO positions and the percentage of inside directors. When there was a combination of hubris and the CEO was also board chair, or when there was hubris and a large percentage of inside directors, the premiums were particularly large. The other interaction term, smallness of outside directors' holdings, was not significant. These results support our expectations about board vigilance.

Effects on Shareholder Returns

The immediate returns and the one-year returns following the acquisitions were -4 percent and -11 percent, respectively (see Table 1). Both means were significantly less than zero (p [is less than] .05). Thus, in line with considerable prior research, these acquisitions in general reduced shareholder value (Agrawal, Jaffe, and Mandelkar, 1992). The correlation between acquisition premiums and immediate and one-year returns were -.09 (n.s.) and -.26 (p [is less than] .01), respectively. This provides some preliminary evidence that larger premiums harm the acquiror's shareholder wealth.

Table 4 shows the results of the regressions for post-acquisition performance. Models 4 and 6 include all control variables. Of those, only the financial synergy variable was associated with subsequent shareholder performance and only for one-year returns (p [is less than] .01). None of the board vigilance variables had main or interaction effects on shareholder returns, nor did they alter the effects of other variables. Therefore they are not included in the performance models presented.