How smart managers make dumb decisions and why shareholders encourage them

Why do they do it?

A limited capacity to see the whole picture -- known as "bounded rationality" -- combined with a faulty ethical compass are two big reasons, shows a new study from the University of Toronto's Rotman School of Management. The study also finds that shareholders are just as guilty of the same weaknesses and that insider trading is linked to earnings manipulation.

"For a long time we've asked ourselves, 'How come smart, rational people carry out short-term schemes that in the long-term undoubtedly are going to sink them?'" says author Ramy Elitzur, who holds the Edward J. Kernaghan Professorship in Financial Analysis and is an associate professor of accounting.

"The answer is -- we're not rational. We're rational only in a limited sense."

The study bases its findings on a model of the manager-owner relationship over time. The model is also noteworthy for combining principles of game theory -- used to predict strategic behaviour -- with the idea of bounded rationality -- that our decisions are always made within the limits of available time, information, and the human capacity to analyze it.

"It tells us, for example, that if we would like to have managers who engage less in earnings manipulation and in insider trading, we should look for managers who are more ethical and suffer less from bounded rationality," says Prof. Elitzur.

That's not a trivial finding, he says, because the model also shows that choosing less ethical managers may be in the best interests of current shareholders, but not future ones. Unless current shareholders also suffer a penalty for such a choice, they will encourage unethical and damaging behaviour. Some provisions in the U.S. Senate's Financial Regulation Overhaul bill from 2010 help to guard against these tendencies, the study says.

The case of Enron is well-known. The scandal at Satyam Computer Services was dubbed "India's Enron," and broke in 2009. Prior to his resignation, Satyam's chairman Ramalinga Raju admitted to years of systematic inflation of earnings and assets, beginning with small manipulations of account statements that eventually got out of control.

Prof. Elitzur says that it took a decade to develop his model and get it published partly because of initial resistance to his findings.

"Many accountants believed that markets are efficient and as such, a lot of the issues of earnings management would be corrected by the markets," he says. "But this belief has changed over time, and we understand better now that earnings manipulation occurs and does indeed affect markets."