Reversion To The Mean: Market Rewards Improvement And Persistence

Market is efficient

Michael Mauboussin and Dan Callahan, CFA from Credit Suisse studied the implications of patterns of reversion to the mean for shareholder returns. Mauboussin and Callahan looked at how changes in cash flow return on investment (CFROI®) correlate with total shareholder returns (TSR) for more than 1,000 companies. TSR includes price appreciation and dividends paid.

Credit Suisse’s analysis is based on a sample of 1,355 companies from the Credit Suisse HOLT® database. Data for CFROI® is measured from 2002 to 2012, and companies without these data points such as financial companies and regulated utilities are excluded. Mauboussin and Callahan sorted their sample into quintiles based on 2002 CFROI® data, creating 5 portfolios with each stock having an equal weight. The first quintile is 20% of stocks with the highest CFROI® for 2002 and the last quintile is 20% of stocks with the lowest CFROI® for 2002. Credit Suisse’s analyst team held portfolios constant through 2012 and tracked TSR for each.

also see: Mauboussin On Economic Returns, Reversion to the Mean

Market Rewards

Source: Credit Suisse HOLT® and FactSet

From the bar chart above, it seems that it is profitable to buy the lowest CFROI® quintile companies. However, the Sharpe ratio, which measures the ratio between return relative to a risk free investment and variability, tells a different story. Higher numbers are better than lower numbers in terms of measuring the reward/variability relationship. The first quintile has the highest Sharpe ratio at 0.45, followed by the third and fourth quintile at 0.42 and 0.41, respectively. The last quintile has the worst Sharpe ratio of 0.29 and the second quintile comes in at 0.34.

Mauboussin and Callahan highlight that Sharpe ratios are closely clustered, which suggests that there is no easy way to generate excess returns. Also, the bar chart suggests that market participants placed a high value to companies that generated high CFROI ® and a lower value to companies in the last CFROI ® quintile. In Credit Suisse’s view, the Sharpe ratio distribution and the bar chart suggest that markets are efficient.

also see: A Better Way To Do SRI – Socially Responsible Investing

Market rewards improvement of CFROI® and punishes deterioration

Mauboussin and Callahan studied the relationship between changes in quintile ranking from 2002 through 2012 and TSR. The table below shows the possible combinations of quintiles in 2002 relative to 2012. For example, the first row shows companies that were in the first quintile in 2002 and the corresponding columns show the ending quintile for 2012. The 2002 Q1 row and the 2012 Q2 column show companies that were on the first quintile (Q1) in 2002 and ended on the second quintile (Q2) in 2012. The TSR for those companies is 12.9% and the standard deviation is 27.4%.

Results show that the market punishes deterioration of CFROI® and rewards improvement. For example, the companies that started 2002 in Q1 and ended in the last quintile (Q5) in 2012 had an average TSR of 6.3%. Meanwhile, companies that started in the fourth (Q4) and last quintile of CFROI® and ended in Q1 in 2012 enjoyed TSRs of 29% and 27.2%, respectively.

Market Rewards

Market rewards the power of persistence

Mauboussin and Callahan’s analysis also shows that companies that stay in top CFROI® quintiles can have favorable risk/reward ratios and excess returns. For example, companies that stayed in Q1 for every year of the last decade returned 20.6% with a standard deviation of 21.8%. A Sharpe ratio of 0.70 suggests that variability was rewarded with returns.

Meanwhile, companies that remained stuck in Q5, which represented about 2% of the total population, delivered a TSR of just 5.5% with a standard deviation of 55.3%. The Sharpe ratio for these firms was essentially zero, as these stocks delivered little return for the amount of variability incurred.

also see: SocGen on The value Risk Premium Debate