AMERICAN ACCOUNTING ASSOCIATION
CONTACT: Ben Haimowitz (718-398-7642 or 917-903-9287)
Manipulation of benchmarks is a handy way for companies to mislead investors, study shows
"Don't assume that just because it’s hard to read it isn't important"
A company is struggling: Its latest quarterly report is a tale of declines over the past year -- net sales down 6%, unit sales volume down 3%, earnings from continuing operations down 3%, and earnings per share down 1%
Yet, among reasonably sophisticated investors, this weakly performing firm gets a slightly higher rating than one whose key numbers are exactly the opposite -- net sales up 6%, unit sales up 3%, earnings from continuing operations up 3%, earnings per share up 1%.
What to make of this? True, the portions of the press releases where these results are detailed are difficult reading, with their long sentences, lack of paragraph breaks, use of technical terms, and lack of tabular or bullet items -- the kind of presentations that led one recent SEC chairman to declare war on complexity in financial writing. But that is only part of the reason for the peculiar ratings assigned to the two companies, according to a new scholarly paper. The other part is that the paragraphs in which these results are reported are each preceded by a plainly written account focusing on a different set of benchmarks -- namely, the companies' success, or lack of it, in meeting guidance forecasts.
The companies and the press releases are fictional -- parts of an experiment described in the current issue of the American Accounting Association journal The Accounting Review. But the misguided assessments made by the experiment participants constitute important lessons for real life investors, the study's authors believe.
"Because the guidance narrative is more readable than the text detailing actual trends, it becomes more salient to investor-readers," explains Elaine Ying Wang of the University of Massachusetts Amherst, who collaborated on the research with Hun-Tong Tan of Nanyang Technological University Singapore and Bo Zhou of Shanghai University of Finance and Economics. "In the case of these two similar companies, the one with the better results over the past year (which ought to mean more than whether or not earnings beat prior guidance) tends to be underrated because its good results are buried in difficult-to-read narrative, while its guidance shortcomings are spelled out clearly. In contrast, the company with poorer results has buried them in murky prose while its success in surpassing its own guidance emerges in plain language."
Prof Wang adds: "A lot has been made of the opacity of financial writing. But the greatest danger to investors may not be low readability per se but that managers can use it opportunistically to obfuscate unfavorable benchmarks while embracing clarity for favorable ones that are less important. As we found in this study, readers tend to focus on what's easy to understand and to give short shrift to what isn't, and therein lies an opportunity for investor manipulation."
The study cites two instances of press releases in recent years that appeared to be pursuing exactly that. A release issued by United Airlines about its fourth quarter 2013 earnings reports an overall operating loss, but, as the study notes, "highlights the good news in arguably easier-to-read bullet points" -- for example, that it ended the year with $7 billion in unrestricted liquidity and that co-workers earned $119 million in profit sharing. Similarly, a press release from Eli Lilly for its third quarter of 2013 “lists positive facts about the company in easier-to-read bullet points but discusses the decrease in net income and earnings per share using what appears to be more difficult-to-read language.”
The professors tested the effectiveness of such tactics in an experiment involving 131 MBA students at a major U.S. university. Participants averaged 13 years of work experience and had completed a mean of 2.34 finance courses and 3.59 accounting courses. Eighty percent had had experience investing in stocks and about 85% had previously read earnings press releases or company annual reports.
Subjects were presented with background information and historical financial data about a company resembling a firm listed on the New York Stock Exchange and were asked to read a four-paragraph quarterly earnings press release it had issued. In conclusion, they were asked to what extent they agreed that earnings and stock performance would be strong in the near future, on a scale of -5 (strongly disagreed) to +5 (strongly agreed).
Background information and historical financial data presented to all participants were the same, but text dealing with the current quarter varied. The first two paragraphs of this content spelled out in plain language either success or failure in meeting previously issued management guidance, and the fourth paragraph consisted of a pro-forma sentence expressing confidence in the company and its future. The heart of the experiment was the third paragraph, which compared key aspects of financial performance in the current quarter (sales, earnings, etc.) with those of the year-ago quarter. In the version read by half the participants this news was predominantly positive, while in the version read by the other half it was mainly negative. Further, in half the cases these performance highlights, whether positive or negative, were conveyed in easy-to- read charts and bullet points, while in the remaining cases the presentation (adapted from actual earnings releases) was, if not completely opaque, difficult to read..
When both the content pertaining to guidance (paragraphs 1 and 2) and the text reporting key financial results (paragraph 3) were consistent (that is, good news or bad news in both cases), it made little difference whether the third paragraph was hard or easy to read. But where there was inconsistency (for example, the company met analysts' forecasts but reported lower earnings than a year earlier or missed forecasts but reported higher year-on-year earnings) the readability of the key third paragraph became critical
Thus, when the readability of paragraph 3 was high, participants rated the strongly performing company far higher than the weakly performing one (+1.59 versus -1.17). But when readability was low, they rated the two firms about the same; in fact, the weakly performing firm actually did slightly better than the strong performer(+0.17 versus -0.36), although the difference was not statistically significant.
What makes these anomalous results highly pertinent, the paper points out, is that it is commonplace for benchmarks cited in financial reports to conflict. One study, for example, found that in 37% of more than 50,000 reports a company beat analysts' forecasts while earnings declined or missed analysts' forecasts while earnings rose. As the Accounting Review paper notes, "Managers can improve the readability of content related to the benchmark with positive implications and/or obfuscate content of the benchmark with negative implications. In turn such actions can affect how investors react to managers' disclosures."
What is the lesson here for investors? "Allocate more time than you may be inclined to do to the difficult-to-read parts of press releases or earnings reports, particularly if other parts are easy to read," Prof. Wang says. "Don't assume that just because it's hard to read it isn't important. That may very well be what some devious manager is counting on you to think."
Entitled "How Does Readability Influence Investors' Judgments? Consistency of Benchmark Performance Matters," the study is in the January/February issue of The Accounting Review, published every other month by the American Accounting Association, a worldwide organization devoted to excellence in accounting education, research, and practice. Other journals published by the AAA and its specialty sections include Auditing: A Journal of Practice and Theory, Accounting Horizons, Issues in Accounting Education, Behavioral Research in Accounting, Journal of Management Accounting Research, Journal of Information Systems, and The Journal of the American Taxation Association.