As large North American companies release their first-quarter results, investors should be aware that these updates contain much more information than meets the eye.
Share price fluctuations occur almost instantaneously upon the publication of quarterly figures, with computer algorithms reacting to earnings and revenues faster than any human could hope to digest and act upon the information. But that doesn't mean there are no profitable opportunities available for investors willing to wait and listen to the company conference calls that follow these earnings announcements.
"Academic research has often argued that investors focus excessively on whether the reported numbers beat or miss expectations," wrote Macquarie Research's quantitative analysts in a nt note. "If this is true, then we can expect the soft information which emerges from the call to be incorporated slowly in prices, which would lead to return predictability."
Quantitative analysts at Macquarie undertook a thorough examination of conference calls from 2001 to 2014 and future stock performance from companies representing roughly 70 per cent of its coverage universe to get a sense of whether this "soft information" is useful for investors.
Take the example of two companies that both beat on earnings, but experienced diverging trends in sentiment compared to their previous conference calls.
"Our analysis finds that, on average, the company with improved conference call tone tends to outperform over a period of three calendar months after the call has taken place," they wrote. "The effect is robust and significant both statistically and economically."
The methodology used by Macquarie is to use a "bag of words" approach, counting up how many positive and negative words appear in the text of conference calls.
"In practice, by using simple algorithms we try to mimic a human reader who forms an impression on whether the general tone of the call was bullish or bearish," they wrote.
Macquarie draws upon a number of specialized dictionaries, such as the one developed by the University of Notre Dame's Tim Loughran and Bill McDonald, to distinguish between "positive" and "negative" words.
Macquarie advises that investors can use textual analysis as a screening tool, along with fundamental analysis, in the stock selection process.