Equity Factors: The Lego Blocks of Portfolio Management

By Michael Hunstad, Ph.D., Head of Quantitative Research, Northern Trust Asset Management 
mrh12@ntrs.com

As an asset owner your job is to manage the return and risk of your portfolio. To do so, you call on your years of experience and professional insight to guide the portfolio toward a specific desired outcome. At your command are a handful of tools you skillfully employ to steer performance, often with a nudge, sometimes with a jolt, but always with precision. As the owner of the portfolio I ask you a simple question: would you prefer more or fewer tools?

In the fixed income world we have many design options at our disposal: duration, convexity, option adjusted spreads, yield, liquidity, etc. The portfolios we create show clear signs of intelligent construction: liability driven investment structures with matched key rate durations, defined benefit plans that mirror lump-sum-versus-annuity convexity using mortgage-backed securities, property and casualty portfolios that are sensitized to medical inflation – the list goes on and on. In this space we can become creative in crafting outcomes. In a way, these options are like LEGO® blocks. They are the building blocks that allow our imaginations to run wild.

What building blocks do we have in the equity world? After some consideration we may give a reluctant response like style boxes or geography, neither of which is particularly inspiring. It’s somewhat ironic that despite equities being traditionally the largest component of our portfolios, they have the fewest levers to influence outcomes. As a result, most asset owners are resigned to simple cap-weighted benchmarks, a prospect almost unthinkable in fixed income. Some employ active managers to generate excess returns, but this addresses only one of a multitude of possible objectives. Where are the LEGO® blocks? Where is the creativity? 

History has not been our friend in this regard. We’ve been fed a steady diet of CAPM and Sharpe ratios that allow a singular lever in managing equity outcomes, i.e. risk. A few academics in the 1960s and 1970s led us to believe market beta was king and the market portfolio, which we assumed could be proxied by cap-weighted benchmarks, was the most risk-efficient equity allocation. Money flooded into bulk-beta index products and by the 1980s, the single building block approach was entrenched dogma. We don’t know about you, but we’ve never seen a kid have much fun with just one LEGO® block. It certainly stifles creativity.  

The irony is these LEGO® blocks do exist in equity markets; they have just been clouded by marketing hype known as ‘smart beta’. Since the very beginning, dissenting researchers have challenged the cap-weighted precept, providing mounting evidence that alternative factors such as size, value, momentum, quality, volatility and dividend yield produce returns that are independent of market beta. These are the new LEGO® blocks and provide an unprecedented opportunity to completely rethink equities and their role in our portfolios. We now have more blocks to play with − more creativity and, quite frankly, more fun. We are just starting to build and the results are already astonishing!

For example, we recently worked with a severely underfunded pension plan in need of bolstering equity returns without increasing risk. We crafted a portfolio targeting size and value factors in emerging and developed international markets, where they felt higher returns would be generated, and offset the higher risk and beta with low-volatility factor exposure in domestic U.S. equities. The net result was a portfolio with less risk than their global benchmark but with upside participation greater than 100% and downside participation significantly less than 100%. For the 12 months ending June 2016, this asymmetry allowed the portfolio to outperform its MSCI ACWI benchmark by almost 900 basis points, leading to a measured improvement in funded status.  

In a similar manner, today’s forward-thinking pension plans are dynamically varying factor exposure with funded status to better manage surplus risk and alleviate deficits. Sovereign wealth funds are re-engineering core passive portfolios with multi-factor allocations, insurance companies are using factors to hit complex income and risk-based capital targets – and that’s just the beginning. Prescient asset owners have developed “shock resistant” portfolios with a combination of low-volatility, dividend yield and quality factors, which have held up very well during recent stress periods, particularly in the aftermath of Brexit. Others are strategically altering value and size exposures across the globe to express views on growth and diverging monetary policy. Some are even using factors to replicate their favorite fundamental active managers to create high-liquidity sleeves without compromising performance. The possibilities appear endless.

Important to remember is that while the applications of equity factors are plentiful, the factors themselves are not. In previous blogs, we noted most academics and asset managers alike have accepted just seven factors as alternative levers: beta, size, value, momentum, dividend yield, low-volatility and quality. Studies by Carhart (1997), Novy-Marx (2013) and Grinblatt, et. al. (2016) confirmed more than 95% of equity returns can be attributed to these seven factors. While literally hundreds of alternative factors have been proposed, virtually all have been shown to be nothing more substantive than haphazard combinations of these seven “elemental” LEGO® blocks.

The same holds true outside of public equities. Stafford (2015) showed the returns of private equity benchmarks were built almost exclusively from public equity factors, namely size and value. Likewise, numerous papers have linked hedge fund returns to public equity factor building blocks. For example, Mitchell (2013) draws the connection between size, value, quality and momentum and hedge fund index returns. Similar results were obtained by Ang, Goetzmann and Schaefer (2009). Israel, Kang and Richardson (2015) even find value, momentum and low-volatility corollaries in the fixed income space. It seems our LEGO® blocks are very versatile with complex portfolios of active equity, passive equity, hedge funds, private equity and even fixed income composed of a handful of fundamental building blocks snapped together in different ways. 
While to us it is clear that factors are the future of investing, we recognize certain challenges must be addressed if they are to live up to their full potential. First, the entire asset management industry hangs under the specter of ‘smart beta’ and all its negative connotations, waylaying the recognition of factors as independent tools. While the reaction is inevitable, it is also regrettable. We feel it is a tragedy that overzealous marketing has obscured the ultimate utility of equity factors, a beautiful thing lost in a sea of rhetoric.

Second, available factor-based products seem remarkably ill-suited for the goal of precision equity management. The first ‘smart beta’ products were a reaction against passive benchmarks whereby the natural inclination was to weight holdings using something other than market capitalization. So were born a slew of indexes under the rubrics of “alternatively-weighted,” “fundamentally-weighted” and “equally-weighted.” Their proponents found weighting stocks on criteria such as sales, cash flow, book value, dividends or even number of employees (i.e. just about anything), would generate higher returns than a capitalization-weighted benchmark. However, they were well aware the source of these excess returns was factor exposures, typically size and value, which were haphazardly acquired as a by-product of the weighting scheme.

While there is nothing inherently objectionable in these approaches, they bear the obvious question: why beat around the bush? If factors are the building blocks of precision portfolio management, why would we expose ourselves to them in an ad-hoc manner? These weighting schemes produce factor exposures, to be sure, but these exposures are unstable and change materially though time. 

Recognizing many owners wished to take control, managers and index providers began a global arms race churning out targeted “factor-based” products at breakneck speed. The proliferation has been nothing short of massive, with providers like ERI Scientific Beta alone now sporting 4,274 choices . Needless to say, this has left asset owners confused and somewhat annoyed. Adding to their befuddlement are the never-ending salvos of new product introductions. If one line of factor-based products doesn’t sell we’ll see a “high efficiency” line introduced in short order. All of this has caused asset owners to lose faith. After all, where is the conviction? Do all 4,274 choices reflect “best thinking”? Are we not trying to capture the premia of a handful of factors? If so, why do we need thousands of products? We have heard asset owners accuse providers of creating products just to “see what sticks” and based on the evidence we can’t disagree. 

Despite all of these challenges, we are not dissuaded. Through the fetor of ‘smart beta’ our confidence in factors as the future of precision equity management remains unshaken. After all, ours is ultimately a creative industry and the desire to build is indomitable. However, while we are optimistic we are also cautious. A few tips bear stating:

1. There are a finite number of LEGO® blocks from which active equity, passive equity, hedge fund, private equity and even fixed income strategies are constructed. Even if you have 50 managers in your portfolio, whether you know it or not they represent nothing more than varied combinations of these same elemental building blocks. However, in many cases you have a manager building upwards as another is tearing down. The instance of cross-cancellation of exposure can be very high. We believe it far better for asset owners to be in control of their LEGO® blocks rather than leave these important decisions to managers.
2. The best factor-based equity strategies capture the purest factor premia. Many factor-based strategies use the finite set of building blocks in a very confused manner. We are suspicious of products that obtain exposures indirectly through alternative weighting schemes and those that cloud the simplicity of factors with complex construction mechanisms such as maximum de-correlation and maximum de-concentration. To build innovative portfolios you need the purest factor exposures, plain and simple. There are only seven factors, not 4,000.   
3. Arguing over the alpha-generating capability of factors is pointless. While many factors have historically outperformed cap-weighted benchmarks, this is not their raison d’etre. Factors each play a specific role in the portfolio and are valued for more than just excess return, just as bonds are valued despite their tendency to underperform equities. We believe too much time is spent analyzing the nuances of individual managers and not enough time thinking big-picture construction about how factors can work together.
4. Start from the basics. Recognize there are a handful of LEGO® blocks to work with and build the simplest portfolio that works for you. Stop over-engineering with complex manager line-ups. 

For more information on how to assemble your portfolio around factors, please see our white paper Improving Active Risk Budgeting or contact us at GlobalEquityStrategy@ntrs.com and let our experts guide you through the “construction” process.

REFERENCES

1. Ang, A., W. Goetzmann, and S. Schaefer. “Evaluation of Active Management of the Norwegian Pension Fund – Global.” A special report prepared at the request of the Ministry of Finance. Oslo, Norway (2009).
2. Carhart, Mark M. "On persistence in mutual fund performance." The Journal of finance 52.1 (1997): 57-82.
3. Grinblatt, Mark, et al. "Style and Skill: Hedge Funds, Mutual Funds, and Momentum." Mutual Funds, and Momentum (January 6, 2016) (2016).
4. Israel, Ronen and Palhares, Diogo and Richardson, Scott A., Common Factors in Corporate Bond and Bond Fund Returns (April 28, 2016). Available at SSRN: http://ssrn.com/abstract=2576784 or http://dx.doi.org/10.2139/ssrn.2576784
5. Mitchell, Jeffrey H. "Hedge Fund Replication: Traditional Beta, Alternative Beta, and Alpha." NEPC: www. nepc. com/writable/research_articles/file/2013_04_ hedge_fund_replication-traditional_beta_alternative_ beta_and_alpha. pdf (2013).
6. Novy-Marx, Robert. "The other side of value: The gross profitability premium." Journal of Financial Economics 108.1 (2013): 1-28.
7. Stafford, Erik. "Replicating Private Equity with Value Investing, Homemade Leverage, and Hold-to-Maturity Accounting." Homemade Leverage, and Hold-to-Maturity Accounting (December 20, 2015) (2015).