An investment strategy modelled strictly on analyst stock ratings would be overwhelmingly bullish and would rarely sell anything.
But that’s not to say that sell-side research doesn’t have value.
A new report from S&P Global Market Intelligence attempts to strip away the pro-management bias inherent in equity recommendations.
Rather than focusing on the recommendations themselves, the report suggests looking at changes in recommendations, which can foreshadow movements.
“It may simply be that analysts, as company and industry experts, can often identify early signs of improvement or deterioration,” the report’s author, Richard Tortoriello, wrote.
Analyst skill, however, can often be diluted by a system designed to generate favourable research.
Equity ratings can be influenced by the drive to support investment banking divisions, for example. Additionally, analysts criticizing a company can sometimes find themselves denied access to management.
“Analysts may issue biased recommendations to market a stock, appease management or avoid frustrating clients who hold shares, or they may simply lack skill,” Mr. Tortoriello said.
When analysts initiate coverage on a stock, they tend to start with a positive rating, or at worst, a hold, according to Brandon Osten, chief executive officer of Toronto-based Venator Capital Management. And rather than downgrade an underperforming stock, an analyst will often simply choose to drop coverage. They don’t typically stop following stocks they rate positively, on the other hand.
All of which contributes to a bullish bias in stock ratings.
“It’s a natural condition that 60 to 65 per cent of the recommendations are going to be buy recommendations. That’s just your starting point,” said Mr. Osten, a former sell-side analyst.
Of all the individual ratings on all of the stocks in the S&P/TSX Composite Index, less than 5 per cent of them are sell recommendations, according to Bloomberg data. Some 35 per cent are holds.
Meanwhile, target prices have their own credibility issues, as they tend to simply follow share price up and down.
“The vast majority of analysts determine their target pricing through momentum in the stock,” Mr. Osten said. “If the stock hits my target, and I don't want to downgrade it, I'll find a reason to increase it.”
Aside from ratings and targets, however, the best analysts tend to be company and/or sector experts, yielding insights that might not be readily apparent to generalist investors.
“Analysts do have an important voice in the marketplace, and analyst sentiment, if examined correctly, can serve as an aid to the investment process,” Mr. Tortoriello wrote.
A strategy based on recommendation changes can generate significant excess returns, according to the study. A portfolio built on betting on positive shifts in analyst sentiment toward U.S. stocks, and shorting downgraded companies, would have outperformed the Russell 3000 Index by an average of 5.7 per cent annually, over roughly the past 20 years.
Target prices can also provide insight into shifts in analyst attitudes, the study suggested. Specifically, the change in the spread between average target price and share price helped predict stock performance over the ensuing three months. Such a trading approach beat the Russell 3000 by 9.4 per cent annually in data dating back to 2001.
Lastly, the S&P report looked at what investing intelligence might be gleaned from analyst revenue projections, specifically how those estimates are dispersed across the analyst community.
By acting as an indicator for corporate quality, stocks with a low range of sales estimates generated excess returns above the Russell 3000 by an annual average of 8.3 per cent.
“We hypothesize that high estimate dispersion indicates that companies are, for one reason or another, difficult to forecast and thus likely to be riskier/lower quality than their lower dispersion peers,” Mr. Tortoriello said.