The Elephant
Is there everything wrong with this statement? If you want to create wealth, you need to let your winners run and you have to cut your losers short. Unfortunately this statement is also an "elephant" modern finance is trying to comprehend.
A few questions that can challenge this adage; what qualifies as a winner or a loser? How long should one hold the winner, a few days, a few years? What happens if instead of cutting losers, we were selecting and accumulating losers? Are today's winners tomorrow's losers or vice versa? Is buying winners about running after growth? And, selling losers about risk management or spurning value? Should all maxims be intuitive?
Inverting the elephant
We know how it sounds if we would have inverted the adage and said "running losers, cutting winners". This would become another school of thought which would suggest, "buy in weakness and sell in strength". So, you understand our thesis now, the financial community is actually left figuring out what to do, buy or sell the winner. One may say, this is how markets work, one man's loser is another man's potential winner. Well, the debate is more than academic and not just a matter of speculative dismissal.
Academic conflict
DeBondt and Thaler [1985] using overreaction showcased that a stock experiencing a poor performance over a three- to five-year period subsequently tends to outperform. This implies that, on average, stocks which are 'losers' in terms of returns subsequently become 'winners' and vice versa. This confirms that running losers and cutting winners could deliver over a 36-month period. In an SSRN paper on relative strength and portfolio management, John Lewis explains how working with relative momentum (buying strong above average performers) strategy is good for a three to six-year period but not good for one month and larger periods, above nine months. For longer periods, buying above average performers (outperformers) fails.
The annual cycle
Sam Stovall confirms that when it comes to an annual cycle, on average, winners beat the losers two to one margin; seven out of every 10 years (since 1970). While industry losers barely beat the S&P 500 by 200 basis points, the winners outperform the benchmark by 600 basis points at 16 per cent annualised and a risk-adjusted return (returns divided by standard deviation) at 0.69.
All about time
Though the annual cycle is important, getting in and out every January is a disciplined approach. Even if we assume investors can follow it, can we conclude that a bi-annual or multi-year holding is not profitable?
The more we dig into performance, we see that relative performance (buying winners, cutting losers) and extreme reversion (running losers and cutting winners) does cross paths frequently. So much so that it's hard to label "running winners, cutting losers" as a universal market behaviour.
Polarity and case study
When exactly does the switch-over from relative to reversion happen; after six, 18 or 36 months? Is relative performance clearly better than reversion? Is relative performance clearly active? Is reversion clearly passive? Is relative performance momentum investing connected to value investing? We ran our relative performance, extreme reversion and worst style models for the Sensex 30, Nifty 50 and CNX 100 from July 2006. The relative performance and worst style model was rebalanced quarterly; the extreme reversion model was rebalanced bi-annually.
ORMI Worst 20 (allocation in the worst 20 from CNX 100) and ORMI Relative Performance Sensex 30 (weighting according to winner rankings) delivered similar risk-adjusted returns (annualised returns vs standard deviation) and as we changed the group from Sensex 30 to Nifty 50 the margin of outperformance between relative performance and extreme reversion (weighing according to loser rankings) became marginal. Hence, dismissing losers was indeed a bad idea.
Is there everything wrong with this statement? If you want to create wealth, you need to let your winners run and you have to cut your losers short. Unfortunately this statement is also an "elephant" modern finance is trying to comprehend.
A few questions that can challenge this adage; what qualifies as a winner or a loser? How long should one hold the winner, a few days, a few years? What happens if instead of cutting losers, we were selecting and accumulating losers? Are today's winners tomorrow's losers or vice versa? Is buying winners about running after growth? And, selling losers about risk management or spurning value? Should all maxims be intuitive?
Inverting the elephant
We know how it sounds if we would have inverted the adage and said "running losers, cutting winners". This would become another school of thought which would suggest, "buy in weakness and sell in strength". So, you understand our thesis now, the financial community is actually left figuring out what to do, buy or sell the winner. One may say, this is how markets work, one man's loser is another man's potential winner. Well, the debate is more than academic and not just a matter of speculative dismissal.
Academic conflict
DeBondt and Thaler [1985] using overreaction showcased that a stock experiencing a poor performance over a three- to five-year period subsequently tends to outperform. This implies that, on average, stocks which are 'losers' in terms of returns subsequently become 'winners' and vice versa. This confirms that running losers and cutting winners could deliver over a 36-month period. In an SSRN paper on relative strength and portfolio management, John Lewis explains how working with relative momentum (buying strong above average performers) strategy is good for a three to six-year period but not good for one month and larger periods, above nine months. For longer periods, buying above average performers (outperformers) fails.
The annual cycle
Sam Stovall confirms that when it comes to an annual cycle, on average, winners beat the losers two to one margin; seven out of every 10 years (since 1970). While industry losers barely beat the S&P 500 by 200 basis points, the winners outperform the benchmark by 600 basis points at 16 per cent annualised and a risk-adjusted return (returns divided by standard deviation) at 0.69.
All about time
Though the annual cycle is important, getting in and out every January is a disciplined approach. Even if we assume investors can follow it, can we conclude that a bi-annual or multi-year holding is not profitable?
The more we dig into performance, we see that relative performance (buying winners, cutting losers) and extreme reversion (running losers and cutting winners) does cross paths frequently. So much so that it's hard to label "running winners, cutting losers" as a universal market behaviour.
Polarity and case study
When exactly does the switch-over from relative to reversion happen; after six, 18 or 36 months? Is relative performance clearly better than reversion? Is relative performance clearly active? Is reversion clearly passive? Is relative performance momentum investing connected to value investing? We ran our relative performance, extreme reversion and worst style models for the Sensex 30, Nifty 50 and CNX 100 from July 2006. The relative performance and worst style model was rebalanced quarterly; the extreme reversion model was rebalanced bi-annually.
ORMI Worst 20 (allocation in the worst 20 from CNX 100) and ORMI Relative Performance Sensex 30 (weighting according to winner rankings) delivered similar risk-adjusted returns (annualised returns vs standard deviation) and as we changed the group from Sensex 30 to Nifty 50 the margin of outperformance between relative performance and extreme reversion (weighing according to loser rankings) became marginal. Hence, dismissing losers was indeed a bad idea.
The author is CMT, and founder, Orpheus CAPITALS, a global alternative research firm