The Sound of Alpha
In recent years, asset managers have unveiled many new investment products designed to improve portfolio returns for their clients
Amid the rapid innovation, the Magnetar team has taken a time-out to revisit a fundamental question:
“What drives investment performance?”
To answer this question, we leverage existing academic frameworks and extend them with our own findings. The goal of this piece is to share a simple explanation of the various components that we believe drive investment performance. To better illustrate the components and how they fit together, we draw relationships between investment management and the more accessible world of music.
Rewinding to Finance 101, the most fundamental concept of investing is arguably the risk/return tradeoff, in which investors demand higher returns for taking on more risk.
An investor starting from scratch has two options to generate returns:
1. Risk-free assets that pay a risk-free rate: Treasuries
2. Risk assets that pay a rate of return above the risk-free rate: Stocks, Corporate Bonds, etc.
Say an investor chooses to purchase an index representing the universe of stocks instead of treasuries. In compensation for assuming the risk of stock market swings, the investor is compensated in the form of an equity risk premium above and beyond the risk-free rate (i.e., the investor’s total return is the risk-free rate plus the equity risk premium). Equity risk premium is a type of “bulk beta,” meaning that it can be easily accessed by tracking a market index.
By definition, active portfolio managers must deviate from market indices to achieve market-beating returns. These deviations are known as “alpha,” and they can be positive or negative. In the 1960s William Sharpe and others combined these concepts into the Capital Asset Pricing Model (CAPM), decomposing stock returns as follows:
Return = Risk-Free Rate + Beta (Equity Risk Premium) + Alpha
Sometimes CAPM is shown without the alpha term as it is assumed that alpha is, on average, zero—i.e., manager skill adds no net value. The implication of that assumption is that an investor should simply invest in passive indices to access bulk beta return streams.
Just as a portfolio manager allocates to asset classes that result in a particular profile of risk and return, a musician employs instruments that result in a particular frequency profile. Going one step further, it would be appropriate to analogize bulk betas—the most fundamental form of risk premia—to the most fundamental instrument humans possess, our voices.
Fortunately, investment management and music have developed components beyond bulk betas and our voices.
Fast forward to the 1990s, when finance professors Eugene Fama and Kenneth French began to question the basic version of CAPM. They found that sub-sectors of the stock market exhibited persistent aberrations misaligned with CAPM forecasts, which catalyzed their search for a more comprehensive model. Eventually Fama and French concluded that distinct risk premia emerge when isolating for value stocks and small capitalization stocks, resulting in a value and small cap style factors, respectively.
Put simply, style factors are risk premia that emerge when systematic strategies target stocks with specific characteristics . While many have rushed to propose new factors, few have proven robust enough to reach the same acceptance as value and small cap factors. One such exception is the momentum factor. A stock whose daily performance over some period of time tended to be in the same direction, either up or down, could be characterized as a momentum stock. A stock’s momentum factor might be the number of days during the period that the stock moved in the dominant direction. The performance of the momentum factor could then be derived from the historical performance of stocks with high momentum factors.
The discovery of style factors meant that we moved from a single factor to a multi-factor risk-return model:
Return = Risk-Free Rate + Beta1 Factor1 + Beta2 Factor2 + Beta3 Factor3 + … +Alpha
There’s no denying the appeal of style factor investing. Not all stocks are the same, so a more discerning approach might identify benefits associated with certain categories of stocks. Stocks representing a particular factor might deliver outsized expected returns for a given unit of risk. Stocks driven by other factors might deliver risks offsetting risks elsewhere in the portfolio, effectively functioning as a hedge, or perform differently depending on the regime.
Plug those characteristics into your portfolio management tool and you may find an optimal portfolio that is even more optimal than you expected. The classic efficient frontier of superior risk-return combinations is replaced by a more efficient frontier, which we believe is reflected in the graph below:
Notably, that superior risk-return profile is not alpha. It’s all beta—or, better said, betas, plural, measured relative to the multiple factors. Outperformance is not driven by a manager’s selection of superior investments discovered beyond the boundaries of the factor definitions. It is the result of a portfolio constructed from risks that pay investors a risk premium but are not highly correlated with one another.
While style factors may represent a sophisticated approach, there is certainly an act of faith involved. Defining style factors requires assumptions about measurement periods, methodologies, weighting schemes, etc. For example, should a momentum strategy be determined based on stock price movements over the last 1, 10, or 30 days? Each manager will have their own answer. All these inputs are judgment calls and given their subjectivity introduce room for error and confusion.
In the end, style factors are abstractions. They do not occur naturally in the wild—you can’t buy or sell a “momentum.” Instead, you need a model that creates and defines momentum, as well as a process to implement it and make it consumable for investors.
Let’s return to our music analogy. The introduction of style factors into a portfolio of bulk beta, as additional sources of risk premia, is akin to a musician supplementing the vocals with wind instruments. The latter brings to the musical composition a new textural dimension and set of frequencies. Witness Frank Sinatra’s regular inclusion of saxophones and trumpets in his many hits, and how much that adds to the music as a whole.
Musicians can optimize their compositions by using an equalizer to isolate specific frequencies—dialing up or down its contribution to the target state. In the illustration below, the bands on the equalizer range from very low sounds (32 hertz) to very high sounds (16,000 hertz).
A factor-driven portfolio manager also has an “equalizer”—a model (e.g., Barra) that helps measure and control how much of which risk premia (e.g., equity risk premium, value, small, cap, etc.) to include in their portfolio mix.
For both music production and investing, this higher degree of control should ultimately deliver a superior product because it is designed to produce a better mix of components.
We also believe specific risk premia also emerge when investors are systematically exposed to specific hedge fund strategies.
Take risk arbitrage. This strategy involves buying the stock of a company being acquired in a merger, and neutralizing exposure to the acquiring company’s stock price by shorting its stock. At their core, all risk arbitrage investments are similar. If you are long the target, a completed deal is a win, whereas a terminated deal is a loss. The design is to capture, on a net basis, wins that outpace losses.
We have found that, over time, systematic exposure to the universe of risk arbitrage transactions (i.e., owning stocks of M&A target companies, appropriately hedged) has resulted in a net positive risk premium—a Risk Arbitrage Risk Premium. This risk premium exists for the same reason why re-insurance premium exists—it is the price of transferring of risk. Fundamental investors owning a target company that received a recent windfall in the form of an announced acquisition do not want to underwrite the risk of a deal breaking, or to own the skewed downside-focused return profile. So, they often prefer to sell their stock to risk arbitrageurs who are comfortable owning it at what they believe is the right price.
Prior to deal announcement, an M&A target company’s stock price may be significantly impacted by style factors (e.g., a company with low book-to-market value would be categorized as a value stock). However, after the company enters the M&A arena, the deal-specific factors typically overwhelm the stock-price impact from style factors and much of the broader stock market. Intuitively, this decoupling results in favorable diversification benefits for investors holding a traditional portfolio exposed to bulk betas and style factors.
Extending this framework beyond risk arbitrage to other systematic implementations of hedge fund strategies opens an entirely new source of systematic return streams that come with varying return potential. Importantly, these strategy factors which target the returns from pursuing hedge fund strategies rather than attempt to capture exposures to the market or value, small-capitalization, or momentum style factors (although they may share some overlapping drivers). Designed to be different, we believe that strategy factors can offer portfolio managers a tool with significant diversification potential.
The suite of hedge fund strategy factors (which we also refer to as Hedge Fund Risk Premia—HRP) now available to portfolios can be analogized to augmenting our existing combination of vocals (bulk beta strategies) and wind instruments (style factor strategies) with a set of stringed instruments. These stringed instruments, from violins to cellos, can further expand the sound frequencies possible adding more options for the equalizer. Yes, the frequencies of these three groups of instruments may have some overlap, but they each have different textures and timbre. The three groups of instruments covering different frequency ranges and coming from quite different sources, enhance the richness of the musical combination.
Taking a step back, what this means is that a portion of returns derived from hedge fund strategies is not actually alpha but is rather systematic risk premia. Which brings us to the final driver of portfolio returns—what is alpha?
Alpha—the final building block of returns—capitalizes on idiosyncratic risks. Commonly referred to as “stock-picker skill”, alpha represents the returns that an active portfolio manager should be able to deliver above and beyond a systematic return stream derived from an index.
Yes. it’s true that a more evolved understanding of beta—encompassing, traditional, style and now HRP factors—now seems to explain a greater share of where investment returns come from. What’s left over is alpha, and it is only logical that if betas (including style factors and hedge fund risk premia) explain more and more, alpha explains less and less.
Still, “less and less” is not “nothing.” To the contrary, specific managers and specific investment processes can demonstrate consistent performance above what their strategies’ betas might imply.
We believe alpha from portfolio managers comes primarily from a few areas:
* Filters (quantitative and qualitative)
* Asset allocation
For portfolio managers focused on public markets, their task is like starting with an index made up of the entire universe. Their first logical step is to filter that universe based on their perception of what securities are likely to outperform or underperform. After the universe of opportunities is narrowed, asset allocation skills are necessary to capitalize at the right time in the right proportion.
The mixture of specialized knowledge and deep experience needed to construct successful filters may be scarce, and the persistent ability to time opportunities scarcer still. However, when skillfully and properly implemented, the alpha from an “index-plus-tilt” approach can be significant.
We believe that portfolio managers focused on private investments can extract alpha on multiple fronts beyond filtering and asset allocation. For example, portfolio managers may have a sourcing edge: industry relationships can serve as a differentiator in identifying and securing the most attractive opportunities. Structuring can also play a role: thoughtful, creative ways to approach investments across the capital structure can help develop the most forgiving and robust risk-return profiles. Active operational involvement with a portfolio company—hands-on industry expertise in which the manager serves as a strategic partner and advisor—can also accelerate growth and value creation. accelerate growth and value creation.
Every manager has their distinct approach to alpha. Consistent alpha generation makes for star managers, just as moving performances makes for star musicians. While anyone can sing the same lyrics or play the same tune on a piano, the musicians who sell out stadiums and have multi-decade long careers don’t simply sing; they are the ones who take it a step further with unique vocal delivery, pacing, energy, charisma, costumes, staging, lighting, equipment, and beyond. This is the case if it’s Eric Clapton’s guitar riffs generating straight ahead rock ‘n roll alpha, or the New York Philharmonic’s dense orchestration producing a classical multi-strategy version of alpha all its own.
Ultimately, we hope it has become clear that portfolio returns are derived from a multitude of return building blocks. Investors have for years had ready access to bulk beta return streams within their 60/40 portfolios, and have more recently gained sophistication about style factors. We believe the next frontier for investors will be understanding and accessing hedge fund strategy factors. Yet even as systematic products harnessing these factors improve, there will always be a place in investor portfolios for the next Beyoncé, or for a portfolio manager who can truly deliver that newfound oh-so-special brand of alpha.
Get updates on new Magnetar Perspectives
Please fill out this form and we will connect to provide you with access to: The Sound of Alpha.