2015 news toogoodtobetrue 883x432 1

Too good to be true?

8 December 2015

The article at a glance

There is a clear connection between earnings manipulation and a string of consecutive “beats” of analyst expectations, finds study co-authored at University …

There is a clear connection between earnings manipulation and a string of consecutive “beats” of analyst expectations, finds study co-authored at University of Cambridge Judge Business School.

Graph with question mark
If a company “beats” analyst expectations for a series of consecutive reporting periods, some sceptics might wonder if the numbers are simply too good to be true. A new study co-authored at University of Cambridge Judge Business School finds that they often are.

There is a statistically clear connection between earnings manipulation and a string of consecutive positive earnings “surprises” linked to higher company valuations, found the study of US companies between 1985 and 2010. It found that a significantly high proportion of firms that have manipulated earnings as indicated by Accounting and Auditing Enforcement Releases (AAER) by the Securities and Exchange Commission (SEC) have positive “earnings strings” of four to eight quarters that exceed analysts’ forecasts.

The research shows further that managers at “AAER firms” manipulate earnings to report such a string of positive surprises in efforts to maintain rather than earn high valuation premiums. In other words, it finds that managers at so-called market darlings face an escalating commitment to meet the high expectations that would justify the lofty valuations, and this escalation eventually causes them to manipulate earnings.

Dr Jenny Chu
Dr Jenny Chu

The study – “The valuation premium for a string of positive earnings surprises: the role of earnings manipulation” – was co-authored by Jenny Chu of University of Cambridge Judge Business School, Patricia Dechow and Annika Yu Wang of Haas School of Business at University of California, Berkeley, and Kai Wai Hui of Hong Kong University of Science and Technology.

“A significantly greater proportion of manipulating firms report positive earnings strings” than both a general population of US firms not subject to SEC actions and a selected sample of propensity-matched firms with similar characteristics to the AAER firms, the study found.
Specifically, during manipulation years 53 per cent of manipulating firms have a string of four quarters exceeding expectations (compared to 43 per cent for the general population of other firms) and 42 per cent of manipulating firms have positive strings for eight quarters (compared to 32 per cent for other firms).

“This study focuses on a clearly defined manipulation period identified by the SEC in order to provide insight into the connection between positive earnings strings, manipulation and market overvaluation,” says co-author Jenny Chu, University Lecturer in Accounting at Cambridge Judge Business School. “This allowed us to contrast valuation multiples before, during and after the manipulation period, and our findings have important implications for understanding the motivation of managers in trying to fool the markets.”

Specifically, the study found that manipulating firms with positive earnings strings already enjoy high valuations prior to misstatement, so these firms seek to manipulate earnings to maintain such valuations. Prior to the manipulation period, AAER firms had median forward price-to-earnings ratio of 43, which declined to 35 during the manipulation period and dropped further to 22 after the manipulation was discovered. In contrast, the ratio for the general population of firms was around 20.

Similar results were found when adjusting for firm characteristics in terms of factors such as company size, leverage and book-to-market ratio.

Using Compustat, FirstCall and other data sources, the final sample used during the study period included hundreds of firm-year observations for the AAER firms and tens of thousands of firm-year observations for firms in the general population.

Further analysis in the study found that while some managers are able to manipulate earnings for two to three years, they “find it increasingly difficult to use earnings management to beat benchmarks for longer periods” – so only a few AAER firms can achieve benchmarks for more than eight to 12 quarters.

The study also found that manipulating firms shy away from providing earnings guidance during manipulation years: only 22 per cent of AAER firms provided guidance, and a third of these firms stop guidance during the manipulation period – with half of these firms providing guidance for only one quarter of the year. This suggests that managers stop providing voluntary guidance when they don’t anticipate future good news.