Smart Beta vs. Alpha + Beta
More commentary from Nicolas Rabener is available at FactorResearch. Sign up for a free email subscription and have the latest research delivered to your inbox.
Simplifying Equity Portfolios
In Buddhist teaching, the primary obstacles that prevent us from ascending to a higher state are ignorance, greed, and anger. This trio of poisons keeps us from evolving not only as people but also as investors.
Most investors still believe we can outperform the market through our stock-picking skills. We remain in this state of ignorance despite ample and available evidence. We also tend to be greedy and to chase performance when selecting securities. And our investment decisions, which should be made as dispassionately as possible, are often influenced by our emotions, anger among them.
This is not a recipe for success.
Just as meditation can focus and clarify the mind, simplification can do the same for most investment portfolios. A single exchange-traded fund (ETF) representing the whole stock market could satisfy the equity allocation of a balanced portfolio. If outperformance is a mandate requirement, then factor investing, among other systematic and research-backed strategies, could be among our options.
But factor investing has its own complexities. Since the financial crisis, a wave of single- and multi-factor smart beta products have become available. Yet these are neither fish nor fowl, just long-only index products with factor tilts.
More recently, mutual funds and ETFs have been launched that give investors direct access to the long–short multi-factor portfolios described in factor investing’s foundational literature. These are the real thing.
So can we create a simple and clean equity portfolio by separating alpha from beta? And what can we do about the innate human weaknesses that sabotage our investment decisions?
Factor Investing in the US Stock Market
While research shows stock picking is mostly a losing game, over time, factor investing has generated positive excess returns.
A portfolio that combines small, cheap, and outperforming stocks is a classic factor strategy. Such a portfolio would have dramatically outpaced the US stock market from 1926 to 2019, according to data from the Kenneth R. French Data Library. Betting on large, expensive, and underperforming stocks, however, would not have ended so well.
We can build the difference between these two strategies through a long–short portfolio with exposures to the Size, Value, and Momentum factors.
Factor Investing in the US Stock Market: CAGRs, 1926–2019
Sources: Kenneth R. French Data Library, FactorResearch
(Though we describe returns from this portfolio as “alpha,” this is not quite correct since the term technically refers to unexplained excess returns. In this case, however, we understand the source of the returns. We might better describe these as “alternative beta returns,” but this could obscure as much as it reveals. So for simplicity’s sake, we’ll use “alpha” to refer to the excess returns of factors.)
Risk and Smart Beta
Based on historical returns, a smart beta portfolio outperformed the US stock market from 1926 to 2019. But both often generated drawdowns greater than 30%, which would have made them hard to hold onto over the long term.
Conceptually, most investors are fine with drawdowns. They are to be expected. But the evidence shows we struggle when our losses mount. We tend to reduce our exposure at or near market bottoms, which leaves us underinvested when the markets recover.
We could improve our smart beta portfolio by including exposure to Quality and Low Volatility stocks, which are thought to offer downside protection. Although this might have led to slightly smaller drawdowns, it would still have been long-only and highly correlated with the stock market. After all, smart beta is still beta. Emotionally, there is little difference between a drawdown of 18% or 20%.
Max Drawdowns of a Smart Beta Portfolio in the US Stock Market
Sources: Kenneth R. French Data Library, FactorResearch
Smart Beta vs. Alpha + Beta
Instead of structuring an equity portfolio as smart beta by selecting stocks ranked by factors, we could allocate to a combination of the stock market and a long–short multi-factor portfolio. We could build an alpha + beta strategy through ETFs or liquid alternative mutual funds. The costs for beta like the S&P 500 are almost zero, and long–short multi-factor ETFs are priced well below 1%, which results in total costs comparable to those of smart beta ETFs.
So what are the key advantages of alpha + beta over smart beta?
- A smart beta portfolio lacks clarity and requires continuous performance attribution analysis to differentiate between beta and factor returns. An alpha + beta portfolio makes it easier to identify whether outperformance is being generated.
- There is a significant difference between smart beta and factor investing in portfolio construction. Allocating to a long–short multi-factor portfolio results in returns more in line with those in factor investing’s foundational academic research.
- Smart beta ETFs have stock market correlations greater than 0.9. By contrast, a long–short multi-factor portfolio has zero correlation with beta. So alpha could be a partial substitute for bonds in a balanced portfolio. In a low interest rate environment, this is an intriguing point. And the alpha’s portfolio weight could be tailored to the investor’s risk preferences: The more risk averse, the less beta exposure.
We created a range of alpha + beta portfolios with exposure to the US stock market and to a long–short multi-factor portfolio with exposure to the Size, Value, and Momentum factors. Though we don’t include transaction costs, the portfolios are rebalanced annually to minimize them.
The smart beta portfolio has generated the highest CAGR, followed by the market (beta), and then the various alpha + beta combinations. This reflects in part the power of compounding returns in a steadily rising stock market.
Whatever the benefits of compounding are in hindsight, investors would have had to stick with the smart beta portfolio. Amid maximum drawdowns of over 80%, that would have been challenging. The alpha + beta portfolios avoided such dramatic declines. Theoretically, investors want to maximize returns, but in reality, we need a smooth ride to stay invested.
CAGRs and Max Drawdowns, 1926–2019
Source: FactorResearch
Shifting the focus to risk-adjusted returns, we can calculate risk–return ratios. These were higher for almost every alpha + beta portfolio compared with smart beta or plain beta. The correlation between beta and alpha returns is close to zero, which generated significant diversification benefits.
Those concerned about lower CAGRs from alpha + beta portfolios could apply more leverage given the higher risk-adjusted returns relative to smart beta or beta, although this would naturally increase cost and complexity.
Risk–Return Ratios, 1926–2019
Source: FactorResearch
Factor Investing in More Challenging Times
Factor investing has yielded attractive returns over the long term. But a long–short multi-factor portfolio with exposure to Size, Value, and Momentum in the US stock market has disappointed over the last decade. The annual excess return was 4.4% from 1926 to 2019, but –0.9% from 2009 to 2019. While other factors like Low Volatility did well over the same period, factor investors might have entertained some doubts or been tempted to cut their losses.
Factor Investing in the US Stock Market (Long–Short)
Source: FactorResearch
So how have the alpha + beta portfolios held up during factor investing’s lost decade? Adding a minor alpha allocation to a beta portfolio increased risk-adjusted returns despite the negative alpha, thanks to the diversification benefits. By comparison, smart beta’s risk–return ratio was lower than that of the markets.
Smart Beta vs. Alpha + Beta: Risk–Return Ratios, 2009–2019
Source: FactorResearch
Given our focus on factor investing’s dark decade, for comparison’s sake, we examined periods of poor stock market returns. Equity markets were flat in the 1940s and 1970s, and recent bear markets have been relatively short. But between the tech bubble in 1999 and the global financial crisis in 2009, stock market returns were almost zero.
But factor returns were positive: Smart beta generated stronger risk-adjusted returns than beta. But the alpha + beta portfolios produced even higher risk–return ratios, even with only minor allocations to the long–short multi-factor portfolio. This is diversification at its best.
Smart Beta vs. Alpha + Beta: Risk–Return Ratios, 1999–2009
Source: FactorResearch
Further Thoughts
Buddhism teaches that the three poisons have their antidotes: generosity, compassion, and wisdom or non-delusion. These can help us on the path to achieving enlightenment. For investors, the last of the three is especially helpful.
Factor investing’s adherents have volumes of academic research to support their approach. Plenty of products provide direct exposure to long–short factors with transparent methodologies, daily liquidity, and low fees. Through them, investors can build simple and clean equity portfolios composed of beta and alpha that should deliver superior risk-adjusted returns.
Yet factor products have underperformed over the last couple of years. It turns out, factor investing has some cyclicality.
The lesson is a simple one: No strategy works all the time. And while this truth might be painful to relearn, acknowledging it can also help clarify the mind. And that can hopefully lead to wisdom.
For more insights from Nicolas Rabener and the FactorResearch team, sign up for their email newsletter.
If you liked this post, don’t forget to subscribe to the Enterprising Investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.
Image credit: ©Getty Images/songwood
Continuing Education for CFA Institute Members
Select articles are eligible for continuing education (CE) credit. Record credits easily using the CFA Institute Members App, available on iOS and Android.
The author mentioned in the Further Thoughts part that there is evidence of past positive “alpha” by long-short multi-factor portfolio. Yes, there are a lot of.
But, could we state there is sufficient evidence for saying that an “alpha” will be delivered by these strategies? Consider how much busy is the multi-factor space now, after all the alpha is a zero sum game.
Hi Simone, thanks for your question. No, I don’t believe we can state with certainty that factor investing will continue to generate excess returns. However, what’s the alternative? Active management is not a solution as most fund managers underperform regardless of the time frame (please see the S&P SPIVA scorecards). For most investors therefore the best choice is simply beta via a cheap ETF, but if an investor is required to generate outperformance, then factor investing remains the best bet.
Great read Nicolas.
What are your top recommended long–short multi-factor ETFs for us to gain exposure to this ‘alpha’ allocation?
Hi Jerome, thanks for your question. We can’t make any recommendations of securities or funds in public, but you can contact us via [email protected] for further details. There are a few liquid alternative mutual funds and ETFs that can be considered. Best regards, Nicolas
Nicoals, thanks for this wonderful article. I am applying this pure alpha + beta approach already successfully for more than 20 years. First with trend following managed futures and then with more and more alternative investment risk factors.
I reap exactly those benefits you described. Much reduced MDDs during the last two crashes without bonds and around +2 percentage points p.a. higher return rate on long-term average over my reference, MSCI ACWI IMI TRN ETFs. This is mainly due to rebalancing alpha by means of the purely systematic 5/25% Swedroe rebalancing rule. My main effort goes into selecting the right alpha instruments and composing them well based on their statistical properties.
But I often cause quite some shaking heads in unbelief or even ridicule ever since with it. I assume the reason is mostly insufficient understanding of the intricate statistics behind trading and portfolio construction, which you all got to get right. But with an engineering background I like statistics and applying it at everyday work as ideal bridge between ideal theory and real practice.
“Consider how much busy is the multi-factor space now, after all the alpha is a zero sum game.”
@Simone, real alpha was, is and will always go far beyond Investing 101. In my experience, each successful trader and investor with sustainable alpha returns masters a unique combination of diverse sciences – math/statistics, behavioural economy, history…, technologies, creative art and even religion with high discipline and decency a far cry away from Wall Street folks.
Thus, in order to have some certainty about alpha delivered in the future, try to find and only invest in those unique masters, who also know, openly admit and respect the limits of their specific alpha source.
One shining example is David Harding, who recently even left the trading art he once piloted, see:
http://www.winton.com/davids-views/July-2018/wintons-david-harding-on-turning-away-from-trend-following-risk-net
You and @Jerome may look for some of these good folks on TopTradersUnplugged.com/start/ by Niels Kaastrup-Larsen, who wants to democratize hedge funds, see TopTradersUnplugged.com/about/
However, one of the best risk factor experts, Lars Jaeger, claims that alpha or rather alternative beta return capacities are much larger than assumed so far if exclusively the futures markets are traded for harvesting it. Because they are from 10 to 500 times larger and more liquid than the equity markets. See his best textbook I know about it: “Alternative Beta Strategies and Hedge Fund Replication”
If the implementation of the HF trading strategies also take the dependency of slippage costs on actual liquidity into account then capacity would be infinite, Lars says. Maybe because: “Only two things are infinite, the universe and human stupidity, and I’m not sure about the former.” – Albert Einstein 😉
I am just in the process of studying this new approach available through low cost ETFs, assumed to revolutionize the hedge funds industry in the futures as indexing, established by Bogle, has done it with the mutual equity fund industry before.
However, I am also still struggling with the issue of alpha or alt beta capacity if the trading strategies are public knowledge. What is your position about it and why, Nicolas? Will alt beta eat itself or be sustainable like traditional beta returns, if implemented well, as Lars claims?
Hi Norbert, thanks for sharing your investment experience and philosophy.
We could replace the long-short multi-factor portfolio with trend following funds (CTAs) in the article and would likely get similar results. Even better, combine both as an alternative allocation, which you seem to be doing.
I haven’t read Lars Jaeger’s book, but should and agree that the alternative beta capacity is large and unlikely to disappear, for a few reasons:
1) Factors are as cyclical as equity markets and are not particularly easy to hold in a portfolio, even if combined into a multi-factor portfolio. Most investors do not have the emotional capacity to stomach the drawdowns, which makes it unlikely that all factors become structurally crowded. The data supports this as factor valuation spreads are diverse, despite record smart beta AUM. Please see this related article: https://blogs.cfainstitute.org/investor/2019/08/19/how-painful-can-factor-investing-get/
2) Plain beta represents negative factor exposure in some cases, eg for the Size factor. The trend towards ETFs, which favours large caps, is making some factors cheaper from a valuation perspective.
3) Factors have different performance drivers and many of them are behavioral, which would only become irrelevant if humans stop trading and AI/algos would trade totally differently (vs being programmed by humans). For example, one reason why Value works is because cheap stocks represent companies in trouble, which makes them emotionally unpleasant to hold and is a market characteristic that is unlikely to disappear.
Best regards, Nicolas
Hi Nicolas, thanks for your appreciated answer. I could not agree more with it, concerning alpha or alternative beta capacities for decent long-term risk-adjusted returns similar to traditional beta but favourably un-correlated.
As long as mostly not so “enlightened” humans are investing, trading or even just deciding about the application of proven AI/algos, this should last. No matter if the algos are programmed by humans or AI, I think. Because “a fool with a tool is still a fool.” 😉