This is Part 5 in a series about smart beta ETFs. See below for links to other posts in the series. In this installment, we look at the momentum factor: the idea that stocks that have recently risen or fallen in price will continue that trend over the medium term.
Like value, momentum in the stock market is an old idea, but the academic evidence goes back only to the early 1990s. It was first documented in a 1993 paper suggesting you could generate excess returns by buying US stocks that performed well over the previous three to 12 months and selling those that performed poorly over the same period. Later studies found that momentum also exists in international markets.
In 1999, Mark Carhart published a now famous paper arguing that outperforming mutual funds cannot be reliably identified in advance. Carhart suggested fund managers who seem to have a “hot hand” are often just the lucky beneficiaries of momentum in stock returns. He eventually added momentum to the factors identified by Fama and French to create the Carhart Four-Factor Model.
How you define momentum matters a lot. Stock prices tend to revert to the mean over periods of several years—in other words, momentum isn’t a long-term phenomenon. But the research also suggests it takes time for a trend to gather steam, so very recent performance isn’t meaningful either. That means there’s a “sweet spot” in the intermediate term. Most academic definitions of momentum today include stock prices over the previous two to 12 months. But as we’ll see, ETFs that screen for momentum stocks frequently use other criteria as well, such as earnings surprises.
There seems little doubt that momentum is real. In an entertaining article from 2014, Cliff Asness and his colleagues at AQR Capital Management argue there’s been a momentum premium in US equities dating back to 1801, and in the UK “dating back to the Victorian age.” There’s plenty of evidence it exists in other asset classes, too, they write: “Some of this evidence predates academic research in financial economics, suggesting that the momentum premium has been a part of markets for as long as there have been markets.”
Moreover, unlike the size premium, momentum hasn’t diminished since it was first identified. In a 2012 paper reviewing the evidence for this factor, Robert Novy-Marx declared that “strategies based on intermediate horizon past performance have performed consistently over time and, if anything, have been more profitable over the last 40 years.”
Why have momentum stocks outperformed?
As with the other factors, there’s ongoing debate about why the momentum premium persists. Some argue that momentum stocks are more likely to see sharp reversals as surging prices plunge back to earth. If this is true, then investors are simply being compensated for accepting this additional risk.
The behavioral explanations are probably more convincing. The idea is that investors either overreact or underreact to news such as a company’s earnings announcements. In the short-term they may be slow to respond to good news: the market price of the stock does not immediately go up, leading to a correction several months later, so the stock delivers a higher-than-average return over the medium term. At the other end, following a series of positive announcements, investors become overly optimistic and expect subsequent announcements to be just as rosy. Their enthusiasm sends stock prices soaring in the medium term, but when future news turns out to be more modest, lower returns eventually follow.
What the critics say
The momentum factor is one of the most challenging for investors to capture. By its nature, momentum is fleeting, disappearing after about a year. So any strategy targeting momentum stocks will likely see a lot of turnover. Buying and selling stocks frequently increases transaction costs and taxes, both of which can take a significant bite out of returns.
As always, investors also need a lot of discipline to profit from momentum. Most people prefer to hang on to losing stocks while selling their winners, a tendency behavioural economists call the disposition effect. To be a momentum investor you need to do the opposite, routinely holding high-momentum winners and dumping the losers.
How to access the momentum factor with ETFs
Inventors looking to focus on momentum have a lot of choice. As always, it’s important to look at the methodology each fund uses to select its stocks.
As part of its family of smart beta ETFs, the iShares Edge MSCI USA Momentum Factor ETF (MTUM) includes about 120 large and mid-cap companies in the MSCI USA Momentum Index, which ranks stocks according to price changes over the previous six and 12 months. The iShares Edge MSCI International Momentum Factor ETF (IMTM) uses a similar methodology and holds about 250 stocks in developed markets outside the US, including Canada.
In the US, PowerShares has the oldest family of momentum-based ETFs covering US equities (one for large and mid-caps, another for small-caps) and international equities (both developed and emerging markets). The PowerShares DWA Momentum Portfolio (PDP) holds about 100 US companies, with no individual stock making up more than 3% of the fund. Its benchmark, the Dorsey Wright Technical Leaders Index, is based on relative strength, a measure of how an individual stock’s price has recently moved in relation to the overall market.
In March, the Canadian arm of PowerShares launched a tactical fund that uses these US-listed ETFs as its underlying holdings. The PowerShares DWA Global Momentum (DWG) can hold a variable combination of US, developed or emerging markets ETFs depending on their relative strength, and when that strength is flagging, it holds Treasury bills instead. As of September 13, the fund held about 53% in PDP and 47% in T-bills.
First Asset in Canada also offers a family of momentum ETFs tracking more transparent Morningstar indexes that select stocks based on several factors, including return on equity and earnings revisions and surprises, as well as recent price changes. The First Asset Morningstar Canada Momentum (WXM) uses this methodology to select 30 companies, which are then equally weighted. Similar ETFs cover the US and international markets with 50 and 200 equal-weighted holdings, respectively.
Earlier this year the firm launched the First Asset Global Momentum Class ETF (FGL), which holds a combination of the above three funds. But unlike its PowerShares counterpart, FGL does not do this tactically: it simply holds the three regions in proportion to their market cap, so it’s about 59% US, 35% international and 5% Canada.
Vanguard Canada’s new family of smart beta ETFs includes the Vanguard Global Momentum Factor ETF (VMO), which can hold stocks from all developed countries. This is an actively managed fund, however: it does not track an index, and there is no publicly available information on the methodology.
Other momentum ETFs are based on the recent price movements of sectors or asset classes, rather than individual stocks. The First Trust Dorsey Wright Focus 5 ETF (FV), for example, is built from five US sector ETFs weighted according to their relative strength. The Cambria Global Momentum ETF (GMOM) goes a step further and builds a portfolio of ETFs covering broad asset classes from a universe that includes equities, bonds, real estate, gold, commodities and currencies. The Canadian-listed Horizons Managed Multi-Asset Momentum ETF (HMA) uses a related strategy.
Other posts in this series:
Smart Beta ETFs: Your Complete Guide
Understanding the Value Factor
I really wonder how a porftolio of these Smart Beta ETFs would work if you attempt to emulate the asset allocation of the CCP portolios. With the likely higher MERs I think it would be tough to match the returns of the low fee CCP portolios. I assume you will discuss such in the series finale? I guess there could be a potential for higher returns, at the cost of more management by yourself and in fees..
I’m surprised you didn’t mention Vanguard Canada’s Global Momentum Factor ETF (VMO) in your rundown of available momentum factor ETFs. Too new?
@Todd: Thanks for mentioning that. VMO (as with all the other Vanguard factor ETFs) is actively managed, so it’s impossible to say anything about the methodology. But I should have included it in the list, and I have updated the post accordingly.
Is there an ETF / index fund that tracks MSCI Canada IMI Momentum Index? What’s the use of an index if it cannot be invested in somehow? Or am I missing something? Thanks