CEOs who are approaching retirement tend to inflate their earnings forecasts, according to a new study.

The study, which appears in the current issue of The Accounting Review, published by the American Accounting Association, found that, on average, the predictions of chief executives err far more on the side of optimism in their final year than in prior years, producing an upward bias two and a half times greater.

“Relative to CEOs who will continue with their firms, retiring CEOs face strong incentives to engage in opportunistic terminal-year forecasting behavior in an attempt to inflate stock prices,” said the study by Shawn X. Huang of Arizona State University, Cory A. Cassell of the University of Arkansas, and Juan Manuel Sanchez of Texas Tech University.

The incentives tend to grow with the size of the CEO’s holdings of stock or stock options, which would greatly appreciate in value with share-price increases.

The study noted that “because SEC trading rules related to CEOs’ post-retirement security transactions are less stringent than those in effect during their tenure with the firm, post-retirement transactions can be made before the earnings associated with the opportunistic forecast are realized and with reduced regulatory scrutiny.”

Under current regulations, chief executives have to register trades of company stock with the SEC during their firm tenure. After they retire, the rule continues to apply for up to six months following an opposite transaction. For example, if a CEO makes a large stock purchase four months prior to departing the firm, he or she would have to file a record of sale with the SEC for only two months after departure but not subsequently.

“Given our understanding of this rule, the SEC may want to consider extending the period for post-retirement filings of transactions,” said Huang. “In any event, our results should be of value to investors, since the great majority of retirements are announced well before departure. In our sample of 862 retirements, this was the case about 70 percent of the time, with the announcement occurring an average of about 80 days before the CEO's departure. In other words, this knowledge is generally available, and investors may want to take it into account in assessing CEO earnings forecasts, particularly since there is some evidence that early announcements are associated with forecast inflation.”

The new findings came from an analysis of forecasts issued from 1997 through 2009 by retiring chief executives relating to earnings that were not realized until the CEOs left office. Retirement, which occurred on average at age 61, was defined as a turnover event in which the CEO did not secure a position with another public company by 2011. Earning predictions in the final year were compared with those made in the years immediately prior (two to four prior years, depending on the length of CEO tenure).

The professors found CEOs much more likely to issue earnings predictions in their terminal year than beforehand. Approximately 35 percent made at least one forecast in their final year, compared to only about 20 percent in the preceding years, an increase that the professors attributed to the fact that “retiring CEOs who have not previously issued forecasts but would like to issue good news or biased forecasts in their terminal year must change their behavior.”

Consistent with this, the tenor of forecasts became a lot more optimistic in the final year. About two-thirds of the terminal-year forecasts sparked an above-market return in the company’s stock in the three days surrounding the CEO’s prediction, while only about half did so in preceding years. In addition, the upward bias of forecasts increased by about two and a half times, with the average overestimate of earnings rising from 0.2 percent of share price to 0.5 percent. In other words, in a company with a $10 stock price, the terminal-year CEO forecast would exceed actual earnings by about five cents per share compared to about two cents in prior years.

Having found a propensity toward inflated earnings forecasts in the CEOs’ final year, the researchers sought to identify the factors that might encourage or discourage it. A major contributing factor, they found, was the extent to which the CEOs’ compensation depended on their companies’ stock price, through such mechanisms as stock or stock-option grants. In cases where this dependency was above average, the upward bias of their terminal-year forecasts was about 70 percent greater than it was for chiefs for whom it was below average. Another tip-off to inflated forecasts, the professors found, was a sharp reduction in capital expenditures and R&D, policy choices that artificially boost current-period income and lend apparent support to opportunistic predictions.

In contrast, a number of factors appeared to constrain the CEOs’ terminal-year forecast inflation. They tended to be less pronounced than otherwise when a firm hired a successor CEO internally, when a CEOs retained a position on the firm’s board of directors, and when the company had high institutional ownership and a high degree of board independence.

The study, “Forecasting without Consequence? Evidence on the Properties of Retiring CEOs’ Forecasts of Future Earnings,” appears in the November/December issue of The Accounting Review, published six times a year by the American Accounting Association, a worldwide organization devoted to excellence in accounting education, research and practice.

Register or login for access to this item and much more

All Accounting Today content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access