Friday, 14 December 2012

Using the "Wisdom of Crowds" of Analysts to Find Safe, Profitable Canadian Stocks

James Surowiecki's best selling book The Wisdom of Crowds explains how combining the opinions groups of people can produce better decisions and numerical estimates than those of individual experts in a field.

Along the same vein, academic finance researchers some years ago discovered that it is worth looking at the opinions of professional stock analysts collectively instead of only individually (e.g. see Differences of Opinion and the Cross-Section of Stock Returns by Karl Diether, Christopher Malloy and Anna Scherbina and most recently Analyst Forecast Dispersion and Aggregate Stock Returns by Guang Ma, which reviews and summarizes other papers up to 2011). What the researchers specifically found was that the dispersion or range of earnings estimates could predict to some degree future returns. The lower the range of estimates, the better the future returns or conversely, the greater the range of estimates, the lower the future returns. It seems the issue is mainly about true uncertainty as to where a company is heading, which makes intuitive sense - if analysts cannot agree, widely diverging futures are more likely. For an investor that is useful information about risk, or conversely, safety.

Collecting the list of Canadian stocks to analyze
We decided to have a look at bigger companies trading on the TSX to test the idea. The more opinions / estimates the better should be the results and bigger companies tend to have more analysts following them and forecasting earnings. We merged the list of holdings from three mainstream Canadian equity ETFs to come up with 110 stocks - the cap-weighted iShares S&P/TSX 60 Index Fund (TSX symbol: XIU); the accounting factor weighted PowerShares FTSE RAFI Canadian Fundamental Index ETF (PXC) and; the low beta BMO Low Volatility Canadian Equity ETF (ZLB).

We collected most of the base data by entering the stock symbols in Globe Investor's My WatchList, then downloaded the spreadsheet to our own PC to add the key column (outlined in blue) where we compare the spread between the highest and the lowest analyst estimate of next year's (2013) earnings per share (EPS). Most of the EPS figures come from Yahoo Finance, where a Quote page for any stock has an Analyst Estimates link on the left side of the page (e.g. for Royal Bank here). Unfortunately, Yahoo does not have coverage for all the stocks in our list, so we turned to our own discount broker BMO Investorline (other brokers may have this data too) to get the missing ones, shown by the cells in yellow in the comparison table. (That filled in every stock except for Onex Corporation for some reason - do no analysts cover the company? Another company Superior Plus we decided to put aside as only two analysts cover it.)

Result - A definite pattern: low dispersion equals better past returns
Sorting the stocks from lowest to highest dispersion of EPS estimates in our series of comparison tables below seems to align on average with better performing stocks, though of course there are exceptions.

The lowest dispersion stocks display multiple desirable features:
  • more consistent profitability - positive earnings, no losses, they are all making profits
  • more stability - smaller recent earnings and sales surprises; none of the stocks' beta is above the market average of 1.0, most are well below (beta is a measure of the volatility of a stock's price relative to the TSX market average - a beta of 0.5 means that if the market moves down, or up, 2.0% the stock will move only 0.5 x 2.0% = 1% down, or up)
  • positive compound 5-year returns (observe the very few red negative numbers near the top of the list). 
We should point out that we have here compared past results with current future-looking estimates, whereas the research looked at the future earnings estimates against future subsequent returns. However, if the research papers are right, for most of these companies, the past will be like the future and shareholders will continue to smile.
Slow and steady stocks seem to win the race on average
Though the trailing 5-year returns are positive, most are single digit and the first stock with extremely high annual 22% compound 5-year returns is Alimentation Couche-Tard in 32nd place for low analyst dispersion. The highest 30+% per year stocks are mixed in amongst the big losers.

No surprise - Industries like banks, real estate, insurance companies, utilities, consumer staples, telecomms have less dispersion
The image of stability of companies in these sectors gets borne out by the data.

No surprise - Low volatility, low beta stocks have tight earnings estimates
The same stability that results in analysts getting quite close to each other's estimates along with smaller profits and sales surprises also results in a stable stock price and low beta. ZLB naturally contains more of these stocks than other ETFs as it selects its holdings based on low beta.

The winners - More Reward than Risk
We divided all the stocks into three approximately equal-sized groups. This one looks quite attractive.

In between - Potential Big Reward but Appreciable Risk Too

Unattractive - More Risk than Reward

Bottom line: This data should not be a stock selection decider on its own as other factors like accounting ratio analysis should enter the decision process - a good company may not be an attractive buy at current prices. However, the dispersion of analyst estimates is an additional useful tool in the investor's assessment kit bag.

Disclaimer: this post is my opinion only and should not be construed as investment advice. Readers should be aware that the above comparisons are not an investment recommendation. They rest on other sources, whose accuracy is not guaranteed and the article may not interpret such results correctly. Do your homework before making any decisions and consider consulting a professional advisor.

No comments: