Momentum is
a very unique trading style. Cliff Asness defines it as “a phenomenon where
securities that have performed well relative to peers (winners) tend to
outperform. Additionally, securities that have underperformed peers (losers)
tend to continue to underperform” (Fact, Fiction, and Momentum Investing).
Momentum is different
from, and should not be confused with, Trend Following. Momentum ranks securities relative to peer performance, whereas Trend Following focuses on absolute price changes. Moreover, momentum takes no position on a market direction when it
ranks securities by recent performance. Conversely, Trend Following is a useful
tool that can be used to identify the trending behavior of a market.
In the
investment business, momentum is one of the newest strategies. It was officially
discovered in the academic community in 1993. Given this recent discovery by
academics there remains much skepticism around momentum’s efficacy as a viable trading strategy.
Despite the hesitancy
to accept momentum in academia, practitioners have been using profitable
momentum trading strategies for far longer. In the fall of 2014 Cliff Asness
co-authored a paper entitled “Fact, Fiction, and Momentum Investing” that
reviewed academic studies that either supported momentum or disproved the
arguments from its opposition. Below are some of his findings.
Myth 1 –
Momentum Returns are too “Small and Sporadic”
Contrary to
this opinion that momentum does not perform well over long periods of time,
Asness and fellow researchers cite a study that identifies the persistence of
momentum in US equity markets for 212 years (1801 to 2012). Of course,
momentum, like any investing style, will have its periods of underperformance.
However, in their findings out-of-sample data further confirm the existence of
momentum persistence in financial markets. Critics may also note that momentum
is more volatile or sporadic than other investment factors. Even if a period of
higher volatility does temporarily impact momentum investing adversely, over
the long run it should provide better relative and risk-adjusted returns than
other factors such as value, equity risk premium, and size. Further, Asness
states that “momentum returns are large, large after risk-adjustment, and
larger than other factors, even those occasionally being promoted by the same
crowd calling momentum ‘small and sporadic’.”
Myth 2 –
Momentum Cannot be Captured by Long-Only Investors
If this myth
were true, then all the documented positive returns for momentum would come
from being short the losers. Asness and his team find that “there is little
difference between the long and short sides of momentum. Historically, almost
half of the ‘up-minus-down’ premium came from ‘up’.” In other words, the long side is just as profitable as the short side. After looking at the evidence,
this is one of the weakest arguments against momentum but, oddly, one of the
most pervasive misconceptions in the investment community.
Myth 3 –
Momentum is Much Stronger in Small Cap Stocks than Large Caps
Researchers
have found little to know evidence that momentum performance is correlated to
the size factor. Data shows that the momentum factor outperforms value in both
small and large cap categories. Additionally, Asness notes that “momentum,
unlike value, is far more robust among large versus small stocks.” In other
words, despite the arguments from the value investing community, evidence
suggests that there is virtually no premium from large value investing.
Myth 4 –
Momentum Does Not Survive Trading Costs
While
momentum does have higher turnover than other strategies, evidence suggests
that this assertion is false. For example, Frazzini, Israel, and Moskowitz
found that the trading costs per dollar for momentum are actually fairly low.
As a result, even with higher turnover, momentum is able to survive its
transaction costs.
Myth 5 –
Momentum Does Not Work for a Taxable Investor
After
looking at the data from past performance this statement simply does not hold up.
When compared to value, despite having nearly five to six times more turnover
momentum actually has a similar tax burden. There are two reasons why this is
so. First, momentum investing is favorable for the taxable investor since losers are sold quickly and winners are held for longer time periods. In other
words, the strategy generates mostly short term losses and the majority of its
gains are realized as long-term capital gains. Second, since momentum stocks
typically do not pay a dividend there is less taxable income for this strategy.
Conversely, value strategies often rely on dividends to make up a significant
portion of their returns and can increase the tax burden on a taxable investor.
Myth 6 –
Investors should be worried about Momentum’s Returns Disappearing
There could
be a multitude of biases and misunderstanding in the investment community that
are creating this illusion. As mentioned earlier, momentum’s positive
performance was found to be persistent over a 212 year period. Since momentum
is a newer factor than value or size, academics may have a difficult time
reconciling its validity in actual practice. In order to test this assertion,
Israel and Moskowitz looked at several out-of-sample periods for momentum to
see if momentum’s returns decreased. Based on their findings, Asness asserts
“there is no evidence that momentum has weakened since it has become well known
and used by many institutional investors.” Moreover, momentum’s positive
performance persistence offers diversification benefits to other factors. Due
to momentum’s negative correlation to these other factors, even if its expected
returns dwindled to zero an investor would still want to weight some of a
portfolio toward momentum. That is, momentum would act as a hedge when another
factor is losing.
Myth 7 –
Different measures of momentum can give different results over a given period
This
assertion is not only true for momentum but any strategy. For example measuring
value by price-to-earnings or price-to-cash flow will give different results
over any random time period but are effective over the long term just like
momentum’s measurements. For momentum, using the past 12-month return is the
most common metric. In trading, it is good to build simple, effective strategies that follow the principle of Occam’s
razor. Asness suggests “to guard against data mining, choosing the simplest
measure or taking an average of all reasonable measures tends to yield better
portfolios.” If the average of all these different measures produces similar
results it should be a sign of robustness, not weakness, for a strategy.
Myth 8 –
There is no Theory Behind Momentum
Commonly,
theories behind momentum are categorized as either risk-based or behavioral.
The behavioral argument suggests that momentum represents an over- or under-reaction by market participants. For example, investors under-react due to
the slow transmission of information, displaying the disposition effect (i.e.
holding losers too long and selling winners too soon), or being too
conservative. Conversely, when investors overreact they chase returns due to recency
bias and provide a feedback mechanism that moves prices higher. Secondly, as
with other factor premiums, momentum premium is assumed to be compensation for
risk. Assness notes that “as long as risk appetite is sustained, the premium
will remain stable and long-lived. Likewise, so long as the behavioral biases persist so will the momentum premium.”
Instead of
perpetuating unfounded allegations about momentum investing, practitioners who
dispute the efficacy of momentum for the aforementioned reasons should take
time to look at the body of evidence. In doing so, they will not only become
well-informed on momentum but also be able to add a valuable return stream to their existing strategy allocation.
As always,
please feel free to contact me with any comments or questions. Thanks for
reading.
John
Comments
Post a Comment