Skip to main content

One out of 10 public companies in the United States is committing securities fraud, costing investors US$830-billion each year, but most go undetected, according to a new study co-authored by a University of Toronto professor.

And he says he fears fraud may be more prevalent in Canada.

The study by Alexander Dyck, a professor of finance at the University of Toronto’s Rotman School of Management, and two co-authors, used a real-life situation as the basis for the study: The 2001 collapse of accounting firm Arthur Andersen, and the need for its clients to find a new auditor.

Andersen had roughly 20 per cent of U.S. public companies as clients before the Enron Inc. scandal, and in the wake of its association with Enron, the firm’s former customers endured heightened scrutiny by their new auditors, investors and the public.

“What we saw post-the Andersen blow-up was that [companies] that had Andersen prior were more likely to have revealed fraud,” Mr. Dyck said in an interview with The Globe and Mail.

“There were more frauds that were revealed in the former Andersen clients in that time period than people that kept with the same auditors and didn’t have to switch out of Andersen. And so our interpretation of that is that when people looked harder, they saw a lot more stuff.”

The secret to FTX’s ‘massive, years-long fraud’: A made-up crypto asset whose price was allegedly manipulated

How FTX bought its way to become the ‘most regulated’ crypto exchange

That also meant that had Andersen not failed, all those accounting issues might never have been seen. The authors argue that only about one-third of frauds are typically uncovered, leaving the remaining two-thirds undetected. “Fraud is indeed like an iceberg with significant undetected fraud beneath the surface,” the authors write.

To measure fraud post-Andersen, the academics looked at financial misrepresentations uncovered by auditors, as revealed by securities class-action lawsuits; accounting-related enforcement actions by the U.S. Securities and Exchange Commission; SEC securities fraud cases; and restatements of financials that did not involve clerical errors, but were instead related to judgment by management.

The final choice of counting restatements may be the most controversial. The authors acknowledge that including them pushes fraud numbers above 13 per cent of companies in most years, while limiting the incidences to the other measures of misconduct produces rates between 1 per cent and 4 per cent.

The authors acknowledge that a fraud involves misrepresentation, materiality (the error must be large enough to be important to an investor’s decisions) and intent. “All our measures contain an element of misrepresentation ... where we cannot deliver is on intent. Intent can only be proven in court, and all these alleged fraud cases are settled before they reach the final verdict, since directors’ and officers’ insurance does not indemnify executives if they are convicted in court.”

Mr. Dyck’s work was publicized Monday in The New York Times, which described the study as having “become a fascination among general counsels, corporate leaders and investors” since its early-January publication. However, the Times also quoted Joseph Grundfest, a Stanford Law School professor, and former SEC commissioner, criticizing the study by saying “events they call fraudulent include alleged frauds that weren’t frauds, honest mistakes and differences of opinion about accounting treatment.”

Mr. Dyck told The Globe he thinks fraud is the right term. “You could call this the benign term, ‘misrepresentations.’ But we think ‘fraud’ captures the experience of an investor. There was a material misrepresentation which should have been revealed to them.”

The authors note five previous academic studies since 2010 that estimate the prevalence of fraud, fraudulent intentions or intentional manipulation of financial results at between 10 per cent of 18 per cent of companies.

To arrive at the US$830-billion annual cost of fraud, the authors used other published estimates of the costs of disclosed and undisclosed frauds, multiplied them by their estimates of incidents, and applied the loss of 1.6 per cent of market value to the total capitalization of the U.S. equity market.

Mr. Dyck teaches a program for directors of public corporations here in Canada. “This is one of those things that I talk to directors about on the very first day: Fraud is not something that’s small. If you think it’s small just because you look at the observed numbers, you’re missing a lot of what’s going on.”

In the U.S., Mr. Dyck says, “we’ve got pretty active set of enforcers,” considering litigious class-action lawyers and the SEC.

“In Canada, we have less so. If you just take a look at detected frauds in the Canadian example, they’re much lower frequency as a percentage of the underlying [companies]. I think that the problems are at least as bad here as they are in the U.S., and that would say that the undetected fraud is probably higher in Canada than it is in the U.S., at least for the non-cross-listed [companies].”

Follow related authors and topics

Authors and topics you follow will be added to your personal news feed in Following.

Interact with The Globe