Entrepreneurship sounds a lot like managing research. From Marc Andreessen’s blog:
First, it’s not clear to me that a company manager typically really knows whether or not her stock is undervalued — or accurately valued, or overvalued. Academic researchers and reporters tend to assume that a company manager must have inside information that gives a better indication of accurate company valuation than the information already known to the public professional investor. I think it is equally likely that the company manager would simply have more detailed information about operations — information that might not be all that useful in determining valuation of a large, complex public company — and less information about other companies and other outside factors — because, after all, the manager is busy running her own company; she doesn’t have that much time to think about outside factors that might be really important.
Add this to the level of positive bias you would typically expect a company manager to have about her own company, and I think it’s pretty clear why it’s not atypical for company managers to buy back their own stock at inappropriate times in the real world: they don’t really know whether their stock is undervalued or not.
Second, consider that company managers are often strong-willed, high-ego individuals who hate to lose and hate to be told that they are wrong — and hate to see their stock price fall, and hate for investors to pressure them to fix problems in the business.