“Corporate apologies: Beware the pitfalls of saying sorry”

Interesting, but somewhat cynically sounding piece on “Corporate apologies: Beware the pitfalls of saying sorry” by Jeffrey Pfeffer. (Source:  http://jeffreypfeffer.com/2015/10/corporate-apologies-beware-the-pitfalls-of-saying-sorry/). I see 2 hypotheses, which would be interesting to test empirically: a) Power -> +probability of apologizing OR b) Apologizing –> -power. Or is there a feedback loop: power (t1) –> apologizing (t2) –> power (t3)? But we should also ask: In whose interest would this kind of research be done? (I don’t have a simple answer.)