There are few scapegoats more overloaded with blame for all that ails the U.S. economy (at least when we’re not on the brink of defaulting on our loans) than securities class action lawyers and the Securities and Exchange Commission. You know the rap. Class action lawyers are accused of accomplishing nothing more than transferring money out the pockets of corporate shareholders and into their own wallets; the SEC, meanwhile, is derided for failing to detect flagrant fraudsters like Bernard Madoff and letting the true perpetrators of the mortgage crisis off the hook. (Reuters, incidentally, has a great story today about the SEC’s new hotline for fraud tips, so the next Bernie Madoff won’t get away with deceiving investors.) There’s precious little hard data to measure the deterrent effect of SEC enforcement or securities class actions — how can you count averted frauds? — so it’s all too easy to assume securities litigation and SEC enforcement don’t stop corporations from misbehaving.

But a new working paper by a trio of economics number-crunchers concludes not only that SEC enforcement and class action litigation are both associated with “significant deterrence,” but also that “effective deterrence requires sustained SEC activity and litigation in the industry.” In other words, corporations are likeliest to stay out of trouble when both regulators and plaintiffs lawyers are policing their industry. “We were quite surprised by that,” said study co-author Simi Kedia, an economics professor at Rutgers Business School. “In academics, class actions aren’t very well regarded.”

Kedia wrote the paper, “The Deterrence Effects of SEC Enforcement and Class Action Litigation,” with Emory accounting professor Shivaram Rajgopal and University of Washington accounting Ph.D. candidate Jared Jennings. (I heard about it at the Harvard Corporate Governance website.) She told me it’s a working paper they’ve presented at a couple of conferences but are still planning to refine before submitting for publication.

Nevertheless, it’s a provocative starting point for discussion. The authors made a momentous methodological leap in order to derive meaningful conclusions about deterrence. Because it’s impossible to measure averted fraud, Kedia explained, they decided to look at “discretionary accrual” accounting as a proxy. Accruals, as Kedia explained them to me (an admitted accounting ignoramus), represent the difference between cash and profits. A company that books sales early, for instance, may report high accruals. Accruals can be perfectly legal, Kedia said, but they’re generally an indicator of “creative” or “aggressive” accounting. The higher a company’s accruals, according to Kedia, the more “aggressive” its accounting is, particularly when its accruals exceed what she and her co-authors calculated to be normal accruals by industry.

“We can’t directly see when [corporations] stop engaging in fraud,” Kedia said. “Measuring accruals is an oblique, tangential way to make the point.” (She emphasized that accruals are not illegal, but reducing reported accruals signals that a company has become more conservative in its accounting, and, presumably, in the corporate behavior that underlies that accounting.)