Pinpointing which firms fudge earnings numbers and why

It’s widely assumed that executives are less likely to inflate their earnings when they work at high profile companies that operate under a good deal of regulatory oversight.

Yet it’s also widely known that managers in high-profile companies face tremendous pressure from investors to meet or beat Wall Street estimates every quarter, which incentivizes them to overstate their company’s performance. 

How should policymakers looking to curb accounting fraud reconcile these countervailing forces? My colleagues —Daniel J. Taylor and Robert E. Verrecchia, both at the University of Pennsylvania’s Wharton School — and I set out to answer that question.

{mosads}When it comes to embellishing corporate earnings, managers weigh the costs and benefits: mainly the probability and consequences of getting caught in an accounting scandal against the potential for a higher stock price. (Since many managers’ salaries and bonuses are based on performance measures such as earnings, they are incentivized to inflate their company’s performance.)

So, what keeps a manager honest? We argue that one piece of the puzzle lies in the quality of the information environment surrounding the company. 

To test our hypothesis, we used three primary measures to gauge the quality of the information available about the firm: the number of institutional investors that own the company’s stock, the number of industry analysts that follow the company and the number of articles about the company that appeared in the mainstream media over the course of a year. 

Our dataset included more than 300 restatements resulting from fraud or an SEC investigation. These intentional misrepresentations took place between 2004 and 2012 — well after the introduction of the Public Company Accounting Oversight Board (PCAOB), the agency created by the Sarbanes-Oxley law to oversee auditors.

Using a combination of modeling and regression analyses, we found that as the quality of the information environment improves, misreporting first increases, reaches an inflection point and then decreases. Misreporting is greatest in a medium-quality environment and is least in both high- and low-quality environments.

When executives operate in low-profile companies — meaning there’s not much information or media coverage about what the leadership team is doing and why — we found that they are not particularly inclined to exaggerate earnings.

In this type of environment, the weight that investors place on earnings in valuing the company tends to be low, and so embellishing earnings is not that beneficial — even though no one is watching.

However, as the information environment improves and the company’s profile grows, investors and analysts pay more attention to its earnings. The benefits of reporting embellished earnings grow and managers become more inclined to overstate performance.

At a certain threshold — which equates to about 17 media articles per year and a dedicated analyst following of a little over six — executives become less likely to exaggerate their numbers. The company is so high profile that any further investor attention increases the cost of misstatement more than it benefits the potential for higher compensation.

In other words, as a company garners investor interest, analyst coverage and media attention, the probability of misstatement increases. The market is beginning to pay more attention to the company’s financials.

However, once the company has established a sufficiently high profile — with greater interest from investors, analysts and media — the probability of misstatement decreases.

From a research perspective, what’s most interesting is that middle ground. It’s only when the firm has reached a certain level of prominence that a manager’s incentive to overstate performance decreases.

Policymakers have looked at ways to curtail misreporting for some time. And for good reason: Accounting fraud is incredibly costly. Multibillion financial accounting scandals — like the ones that felled Enron, Fannie Mae and WorldCom in the early 2000s — are prime examples.

But the fact is that regulation aimed at curbing accounting fraud can often have unintended consequences. For companies with little transparency, additional transparency increases the incentive for misreporting. However, for companies that are already very transparent, more transparency decreases the incentives for misreporting.

The association between transparency and misreporting is not necessarily negative. Consequently, regulations designed to increase transparency will not necessarily reduce accounting fraud.

Delphine Samuels is an assistant professor of accounting at the MIT Sloan School of Management.