Predicting Corporate Fraud

Accounting Professor
Robert Davidson describes research he's conducted that may offer a way of predicting which corporations might commit fraud.
  • Transcript


Robert Davidson: How might you be able to differentiate between the executives who ultimately decided to commit fraud and those who didn't? You look at all these incentives and over a 20 year period there's thousands of executives who have strong incentives from their career perspective, from the value of their stock options to inflate the numbers, and only a tiny fraction, you know, less than one percent, that we know of, ever wind up doing this. So maybe there's something unique that we can find about these people.

We started brainstorming and thinking about just anything about someone's lifestyle, beliefs, personal behaviors that might be relevant in this area. And we had a list of all sorts of things like if they had extramarital affairs or we thought about trophy wives or just anything. And some of the things were more just sort of joking around. But then we had a list of things that we thought would make good sense. One of them was if an executive had previously broken the law. Another one would be the relative materialism or frugality of an executive; how they spend their money. And it turns out that we're able to get pretty good data on these two things for a large number of senior executives. So instead of just having an idea, it's actually something that we can test. So in this paper we found that executives who had previously broken the law were maybe two and a half to three times more likely to commit accounting fraud in the analysis we did.

What was perhaps a little more surprising was that if we looked at executives whose only legal violation was a minor speeding ticket or some other traffic violation, we still find significant results. Now, they're definitely weaker but they're still meaningful and significant in the tests that we run. We go one step further. When the SCC investigates a firm they usually name the people that the evidence suggests specifically perpetrated the fraud. And when just looking at who the SCC actually singled out, we found those with these prior legal violations were six to seven times more likely to be the one the SCC indicates had actually committed the act.

The second characteristic we looked at, depending on what area of say psychology or sociology you're looking at, you can think about it as either frugality or materialism, they're roughly two sides to the same coin. I mean if you live a really frugal lifestyle, you're not really materialistic, and if you're really materialistic you're not frugal.

So we were able to get pretty good data on cars, boats and real estate that an executive owns. Ideally we'd have paintings or huge diamonds or something of that sort but you really can't accurately collect data like that for a large group of people. So we came up with just a binary measure whether we treated an executive as frugal or materialistic or unfrugal depending on the value of any vehicles they owned, the length of any boats they may have owned and then sort of an excess value of their real estate. We realized there's a cost to living. People need to pay to live. And depending on where an executive works that cost can vary dramatically. So we wound up basically subtracting the average cost of living for wherever they happen to be and we said if an executive's home was more than double what the average is, we'd consider that relatively unfrugal.

We'd found some interesting research that suggested frugal CEOs placed more emphasis on controls and on monitoring. And we thought strong monitoring and good controls probably reduces the likelihood that fraud takes place. So we didn't find a really strong theory to suggest that these materialistic CEOs would commit fraud themselves, but we thought it was reasonable that if they weren't placing emphasis on controls that somebody in the firm might do it. And that's really what we found. We didn't find that these materialistic CEOs were accused specifically by the SCC of committing fraud, but we found fraud was much more likely to happen at their firms.

And then just as a follow up we thought well, if this really is related to corporate governance broadly or firm controls and monitoring, then these CEOs should also be more likely to have to restate financial statements just because of an error. So we collected data on restatements that were not intentional and basically performed the same analysis and we found they were more likely to have errors take place at their firms. So we think this has to do with a culture that they set in the firm and the relative lack of emphasis on monitoring and controls.

Directed/Produced by Jonathan Fowler, Elizabeth Rodd, and Dillon Fitton