This article by Malmendier and Tate titled CEO Overconfidence and Corporate Investment provides strong support that managerial hubris (overconfidence) can explain at least some of corporate investment patterns.
They construct a measure of CEO overconfidence based on the well-known pattern that managers often excercise employee stock options early becasue their personal portfolios are over-weighted in the options. So, they identify CEOs that have options that are deep in the money and yet don't exercise them until the final twelve months before expiration.
They find that overconfident CEOs investments (acquisitions) are more sensitive to internal cash flows. In other words, if a CEO is overconfident and has more cash, he'll use it to buy something. One implication of the study is that external monitoring is most important if you have both an overconfident CEO and high free cash flow.
Nice paper, and a clever measure.