Structural Alpha: Why the Diversification Premium Isn’t Going Anywhere

Factor investing has become commonplace in the world of investment management ever since Eugene Fama and Kenneth French published their three-factor asset pricing model three decades ago. Since then, a lot of dedicated work has gone into uncovering new factors and sources of excess return unexplained by the factors that came before them, launching a revolution in quantitative and systematic investing. The diversification factor is similar in that it is a source of excess return and isn’t explained by other factor models explicitly.[1] However, unlike most other factors that seem to either weaken significantly or disappear altogether when the mainstream is made aware of their existence, the diversification factor, which has been mainstream knowledge since the early 1950s, has not lost its impact. While this may seem too good to be true, there is a reason that this alpha source not only exists but persists. It is structural in nature. The structure of markets, and how investors participate in markets, makes it such that diversification is available to harvest, and a premium is available for those who are willing to step outside of the convention and do things differently. There are three main drivers for the structural nature of this factor validated by both Viewpoint’s internal as well as third-party research.

LEVERAGE AVERSION

To harvest the premium offered by the diversification factor, an investor must be willing and able to utilize leverage. Investors have systematically demonstrated an aversion to the use of leverage in their portfolios due to either perceived risks inherent with using it, a lack of knowledge about how to properly utilize it, or a lack of efficient access to it. The concept of leverage aversion is well documented in financial literature and is the key driver of the low-volatility and low-Beta factors in equity markets. Generally, as investors seek more return, instead of adding leverage to a more efficient and diversified basket of stocks, they tend to increase concentration in higher risk equities, eschewing low volatility stocks and allowing those who are willing to allocate to them to capture greater upside than would be predicted by other pricing models. The same effect is exacerbated at the asset class level and more specifically the multi-asset portfolio level. Lower volatility portfolio construction is avoided by those who require higher returns and are unwilling to utilize leverage, thereby reducing expected risk-adjusted returns on higher risk asset classes relative to lower risk asset classes. Portfolios that tend to allocate more weight to these unwanted asset classes than the overall market for the sake of diversification are able to capture higher risk adjusted returns, meaning they realize a higher Sharpe ratio. Using leverage, investors, regardless of their return preference, can feast on this Sharpe ratio and capture a premium over conventional portfolios.

While leverage itself does increase risk, the context is important. There is a huge difference between adding leverage to an already risky investment and adding liquid, prudent leverage to a diversified, low-volatility portfolio in order to target a level of risk that is more suitable for an investor.

INDUSTRY & MARKET STRUCTURE

If markets are truly efficient as is theorized in the Efficient Market Hypothesis (EMH), excess return opportunities should not exist. While we don’t subscribe to the EMH outright, as it is just a theory to contextualize pricing anomalies, there is a case to be made for varying degrees of market efficiency across the investible universe. The reason that factor premiums in equity markets become harder to capture as more people find out about them exists because the demand for assets that exhibit attributes that drive excess returns increases, thereby increasing their prices and diminishing the size of future opportunity. We agree with Samuelson’s Dictum that markets are mostly efficient on the “micro” scale but largely inefficient at the “macro” scale. This means that mispricing and fundamental changes are captured very quickly when looking within asset classes, such as at individual stocks. However, at the asset class level, things tend to remain detached from fundamentals for long periods and mispricing can persist, creating larger opportunities if properly identified. Why does this occur?The structure of the investment industry, and the constraints that exist because of it, make it difficult for capital to flow from one asset class to another. Institutions and professional managers oversee the investment of the vast majority of society’s capital. The way the management of this capital is structured is based on relatively tight asset class, geographical, and often, risk limits making it difficult for capital to quickly flow at the macro scale. For example, it is relatively easy for an equity manager to respond to a perceived mispricing between two stocks by buying the underpriced one and selling the overpriced one. However, he has no ability within his mandate to express a view that the equity market is overpriced relative to the bond market; he is a buyer of equities and equities alone. Further, he might be constrained to buying equities that are listed only on North American exchanges and can’t even express a view based on the relative attractiveness between European equities and North American equities. These silos exist across the investment industry. The effect of this, which has been empirically proven in a recent paper that puts forward “The Inelastic Markets Hypothesis,” is that the price of equity investments does little to change the demand for these markets. In a world of forced buyers, opportunities for emergent factors such as the diversification factor to capture excess returns over long periods of time can continue to persist. Speaking about the diversification factor specifically, these inefficiencies lead to increased sources of idiosyncratic risks across asset classes and therefore more opportunity for true diversification to have an impact by mitigating these risks.

SHORT-TERMISM & BENCHMARK OBSESSION

Somewhat related to the structure of the investment industry is the way that benchmarks are built and reported against. The industry is somewhat ingrained in conventional portfolio construction and therefore tends to compare everything to these conventionally built portfolios. This isn’t inherently a bad thing, and we don’t advocate for anything that reduces transparency and comparability, but the key issue is the short-term nature of reporting in financial markets and the incentives that come from it. As John Maynard Keynes famously said, “Worldly wisdom teaches that it is better to fail conventionally than to succeed unconventionally.” This statement couldn’t be more true when it comes to reporting returns on an unconventional strategy against what is considered to be the norm and the scrutiny that comes with it. Diversified portfolios, by definition, are never the best performers in the short term. The reason diversification works is because some assets perform well, while others perform poorly. And so, in the short term, there is a good chance that a better diversified strategy will underperform against a given concentrated portfolio, but over the long term, well-diversified portfolios tend to outperform.

For this reason, no matter how compelling the trade-off between leverage risk in place of concentration risk becomes, the industry is likely to be slow to adapt. To manage career risk, managers may continue to be fine failing conventionally and, therefore, those that are willing to embrace long termism and allow their investments to deviate somewhat in the short term should be paid a premium for withstanding that pain.

The diversification premium offers a clear value proposition for investors who are willing to apply an unconventional style of portfolio construction that has been well thought out and proven over many decades. Being purposeful about diversification while embracing the benefits of a prudent application of leverage and being long-term oriented can give investors an edge that is structural in nature and therefore unlikely to disappear any time soon.

FOOTNOTE

[1] The Capital Asset Pricing Model (CAPM) suggests that returns are only commensurate based on their level of undiversifiable risk relative to the market. However, in practice, portfolios that are overweight lower risk assets compared to the market portfolio tend to have higher risk-adjusted returns over long periods of time. This tells us that the market portfolio isn’t the optimally diversified portfolio and that something that weights assets based on risk contribution likely does a better job.