Alpha-Beta Separation – Why and How?

Partners Capital is a global investment management firm serving institutions and family offices across seven offices in North America, Europe and Asia Pacific. In this newsletter, we write about markets and investments. If you would like to subscribe please click this link .

In discussions about long-term investing, it is rare that analysts spend much time about execution. Macro and market overviews tend to dominate the debate. It is (mostly) a mistake – high quality execution can make the difference between theoretical returns and a successful investment. Alpha-beta separation is an example of such an execution issue. In the current environment of expected lower beta returns, where generating some outperformance becomes more important, the approach merits another look, and we thought it was worth writing one of our newsletters about it.

Put simply, alpha-beta separation consists of separating capital allocation decisions in two components: an alpha component where capital is allocated to active strategies designed to deliver excess returns and a beta component designed to gain market exposure passively in areas where capital is not expected to generate alpha. This presents a number of benefits from optimising the fee wallet (i.e., paying fees only where alpha is generated) to improving expected returns for a given risk exposure level.

Alpha-beta separation is not a new technique: it has been around for about 20 years and has been tested through market cycles, including the global financial crisis or covid 19 and associated periods of elevated market volatility. We have however recently had many discussions with sophisticated investors that show that it is worth stepping back and reflect on why alpha-beta separation can be useful in a portfolio, how it is implemented, how using leverage is important in delivering and how the inherent risks involved can be both managed and mitigated.

The main point of this newsletter is that alpha-beta separation as a portfolio construction technique is well worth exploring if well implemented and managed. If you indeed believe that we are indeed facing an area of low expected returns for traditional portfolios, this not so new approach is well worth a new look.

Limitations of the traditional approach to portfolio construction

Let’s start with the process to deploy a traditional portfolio as illustrated on Chart 1. Step 1 starts with setting a risk exposure level; in this case, it is set to a similar equity risk as a traditional 60-40 equity-bond portfolio. At Step 2, the manager uses for example Mean Variance Optimisation to decide on an asset allocation mix that optimises the risk-return profile of the portfolio. In Step 3, the manager allocates capital within the asset classes using active managers or passive instruments (e.g., ETFs) to fill the overall (beta) risk budget.

Chart 1 – Traditional portfolio construction

This portfolio is an illustration and not a recommendation

There is nothing fundamentally wrong with the approach we just described, and it is how most portfolios are designed and implemented. There are however several issues with this approach. To name a few:

• Alpha potential: First, this approach, by construct, limits the maximum alpha potential available. Once risk capital has been allocated to deliver beta, there is typically not much left to target alpha, and whatever capital is then allocated to alpha may come at the expenses of tracking error as discussed below.
• Limited manager universe: Second, because they start by setting a rigid asset allocation, investors are then limited in their choice of managers in order to fill these allocations. Indeed, the simplest way to implement this approach is to use a core-satellite approach where the core allocation tracks the benchmark passively and where satellite allocations are with active managers targeting outperformance against the same benchmark. This leaves little room to innovation and leads to potentially missing attractive opportunities, in particular amongst absolute return strategies (e.g., macro, convertible arbitrage, M&A arbitrage, CTA, etc.).
• Limited control of fee wallet: Third, investors using this approach are unlikely to optimise their fee wallet and end up paying fees for parts of the portfolio that could and should be delivered passively and cheaply. It is worth spending a bit of time on this last point. To start with, most investors cannot separate ‘alpha’ from ’beta’. Therefore, when investing in an active manager, they end up paying fees, performance fees in particular, on both alpha and beta returns (and sometimes on cash returns). Without rigorous analysis, it is difficult or impossible to tell whether a manager is indeed creating value or solely tapping beta from different factors[1] that could have been replicated cheaply. Let’s take the example of an active equities fund that has returned significant upside over its benchmark – say the S&P500 index. How much of that performance resulted from idiosyncratic stock picking vs. from a broader positioning of the portfolio against factors, such as growth or value? These factors have shown sustained periods of strong performance (positive or negative) and have driven the returns of many active portfolios. They can (and should) however mostly be replicated cheaply through passive strategies, ETFs for example.
• Managing tracking error: Fourth, investors using this approach need to evaluate a delicate trade-off between tracking error[2] and alpha potential. Indeed, when investing in strategies that deliver both beta and alpha, active managers generating the highest alpha also tend to exhibit the highest tracking error. This makes it difficult to reliably deliver the target beta.
• Portfolio rebalancing: Finally, reallocating the beta portfolio – either to rebalance the portfolio towards its target allocation or to target a different asset allocation – can be complicated as it requires reallocating between potentially complex strategies and also potentially reallocating between active managers who may have limited liquidity.

How alpha-beta separation works

Investors started exploring and implementing approaches to address the issues above about 20 years ago. Chart 2 illustrates a fundamentally different approach to building a portfolio where the steps are essentially reversed compared with the traditional approach described above.

Chart 2 – Portfolio construction using alpha-beta separation

This portfolio is an illustration and not a recommendation

The process is inherently iterative, but a simplified version starts with deciding which managers you believe are the most likely to deliver outperformance. The objective is to evaluate managers based on their skills, irrelevant of the asset class they invest in or of their strategy as long as their net exposure to the relevant benchmark for the overall portfolio is low. This may sound counterintuitive but, alpha being inherently harder to find than beta, we might as well start from the hardest problem to crack. This creates opportunities to invest materially behind absolute return strategies, which only made up a small portion of the more traditional portfolio for the reasons we discussed above. Because absolute return strategies contribute little to the target beta exposure, the portfolio manager has a “freer hand” to allocate to these strategies. Finally, because absolute return strategies typically have the highest potential for alpha and typically higher Sharpe ratios and Information Ratios[3], these allocations materially increase the potential and expected performance of the portfolio. Step 2 completes the portfolio with diversifying asset classes, such as property or inflation link bonds.

Finally, in Step 3, investors will use wherever possible capital-efficient derivatives to “fill” the beta portfolio. Utilising passive exposure instruments this way allows access to the deepest and most liquid markets at very low costs. Instead of committing – say $100 – in capital to an ETF tracking an index, the investor would buy a derivative contract with notional of the same $100 amount, post a cash margin (say $10 assuming a margin requirement of 10%) and have the remaining $90 of capital in cash, which can then be invested in alpha generating strategies. This is visualised in Chart 3 below where an investor using this approach has $90 (90%) of their original cash capital leftover (after posting margin) that can be reinvested into other alpha generating assets/opportunities. Note that while index funds or ETFs could be used for this part of the allocation, we believe that it is in investors’ interest to use derivatives to achieve a more capital efficient portfolio because they can achieve equivalent economic exposure with less cash. On the flipside, not using derivatives would leave a much smaller capital available to invest in the strategies reviewed in Step 1 above and would severely limit the overall alpha potential of the portfolio.

Chart 3 – Example of how ‘replication’ frees up capital for investment

How alpha-beta separation using replication can enhance risk-adjusted performance

Let’s take the example of a manager that is investing solely in public equities. A traditional portfolio would earn cash return plus equity risk premium plus potential manager outperformance (i.e., alpha) – these three components being comingled within each investment in the portfolio. Similarly, a portfolio leveraging alpha-beta separation would also earn equity risk premium (through investments in derivatives) plus cash return plus potential manager outperformance (with the last two coming mostly from allocations to absolute return). Outperformance potential would be materially larger for this portfolio because more capital would be deployed behind active strategies with higher alpha potential (i.e., the increased absolute return allocation). Chart 4 illustrates this point quantitatively with two simplified portfolios with the same Equity-Like Risk[4] of 100%. In this case, the chart shows how alpha-beta separation using replication can enhance returns within the same equity risk budget; in the same way, the same technique could be used to target 100% of equities potential returns but with a risk profile similar to that of a 60-40 equity-bond portfolio.

Chart 4 – Illustrative return comparison between traditional and alpha-beta separation portfolios

Simplified portfolio assuming 150% gross exposure and 50% allocation to absolute return strategy

Chart 5 below synthesises this argument by comparing how the resulting portfolios differ conceptually. In summary, the traditional portfolio targets an asset allocation and provides limited opportunities for alpha within each asset class whereas the alpha-beta separation portfolio allocates to the best available alpha opportunities first and then tops up the portfolio to achieve a set risk level.

Chart 5 – Conceptual representation of the traditional- and alpha-beta separation portfolios

What to watch out for?

Free lunches are very rare and there are issues that investors need to watch out for when investing in or when implementing alpha-beta separation.

Managing leverage in a prudent manner

As mentioned previously, the use of derivatives introduces leverage. It is important to note that this leverage (i.e., use of derivatives which increases gross exposure) is critical to achieving a cash efficient portfolio. Indeed, maximising alpha without leverage would imply accepting a decrease in the share of beta returns available. Unfortunately, beta returns have tended to be greater than alpha returns through investment cycles. They remain the main engine of return for many portfolios with alpha being a second engine and the “cream on top”. So, a portfolio using alpha beta separation built on top of ETFs for beta allocations would deliver materially lower returns than that leveraging derivatives.

One strong mitigating factor is that the use of leverage is not meant to increase market exposure for the portfolio it is meant to help achieve a chosen level of exposure in a cash efficient manner. Put simply, it is different from leveraging a portfolio to increase its returns; the cash freed up by the approach described previously is not used to buy additional directional exposure – i.e., in the example on Chart 4, it is not used to buy more equity exposure and there is the same amount of equity market risk in both approaches: exposure does not increase. Finally, gross leverage measures the absolute value of positions and does not consider the correlations of assets or the netting of short positions.

This being said, leverage creates risks (if not from additional exposure) that need to be managed – in particular a)?liquidity risk and collateral management to match potential margin calls and b)?counterparty risk. If markets decline, holders of index futures will be required to post additional collateral. Posting that collateral requires liquidity and deciding how much liquidity to hold for such events is critical. This means holding a prudent level of capital. What that level of capital is depends on the risk appetite of the portfolio manager, but we tend to think about it in a conservative manner: can the portfolio liquidity withstand a 3σ move in market exposure? Would the portfolio liquidity have been sufficient during observed past events of extreme stress in markets?

Counterparty risk is another of risk of executing a portfolio of derivatives that goes way beyond the scope of this newsletter but that requires attention and care.

Managing alpha risk and alpha volatility

Alpha can be elusive – most managers do not deliver any beyond the own fees – and alpha can be volatile even for the best managers. It means that managing alpha risk is a key task to execute a portfolio using alpha-beta separation effectively. It is stating the obvious, but it starts with deep skills in manager selection without which this discussion is essentially moot. Second, it requires sizing managers carefully within the portfolio to ensure that alpha volatility from any of them does not affect the performance of the portfolio beyond a set level. This means that, in addition to looking at capital allocation to different strategies and managers, the allocation sizing needs to take into account their contribution to alpha volatility at the portfolio level.

Managing the fee wallet

Adding alpha also means adding fees – here again there are no free lunches. By choosing this approach, investors will shift their portfolios from low fee beta investments to higher fee alpha investments, absolute returns hedge funds in particular. This will increase the fee wallet without doubt. We do not believe that fees and fee levels per se are the right area to focus on. Return on fees or stated differently excess returns (alpha) after fees is really where investors should focus in managing the fee wallet. This is not only true for portfolios using alpha-beta separation, but because a large share of capital at risk, including leverage, is allocated against fee paying strategies, investors should have an even deeper focus in this area.

***?

As we mentioned at the start, portfolio execution is one of the elements that can take a portfolio from mediocrity to outperformance and it is worth allocating time and resources to getting it right. If you believe that forward looking expected beta returns, on equities in particular, are likely to be subdued compared to the past 10 years and that alpha has become more important, alpha-beta separation offers an opportunity to materially optimise portfolios. It is not new, but it is worth a new look.

[1] Factors are broad and persistent drivers of returns that explain how an asset class is performing. Typical investment style factors include for example size, value, quality, or momentum. Different factors will perform differently under different macro-economic assumptions. Most factors can be replicated passively and cheaply.

[2] The tracking error is the difference between the realised fund performance and that of the corresponding benchmark. More active management typically leads to a higher tracking error.

[3] The Sharpe Ratio measures the performance of a portfolio beyond the risk-free rate compared to the volatility of those returns. The Information Ratio measures the performance of a portfolio beyond a benchmark compared with the volatility.

[4] % Equity-like Risk is the risk of the portfolio expressed as its risk relative to developed market equities risk. For example, a portfolio with a risk level of 50% is expected to have half the risk of developed market equities.

[5] Hypothetical return expectations are based on simulations with forward looking assumptions, which have inherent limitations. Such forecasts are not a reliable indicator of future performance.

Important Information

Investment returns will fluctuate with market conditions and every investment has the potential for loss as well as profit. The value of investments may fall as well as rise and investors may not get back the amount invested. Past performance is not a reliable indicator of future performance.

This material is for information only, and we are not soliciting any action based upon it. This material is not an offer to sell or the solicitation of an offer to buy any investment. It is based upon information that we believe to be reliable, but we do not represent that it is accurate or complete, and it should not be relied upon as such. Opinions expressed are our current opinions as of the date appearing on this material only. We do not undertake to update the information discussed in this material.

Any projections, market outlooks or estimates in this material are forward–looking statements and are based upon assumptions Partners Capital believe to be reasonable. Due to various risks and uncertainties, actual market events, opportunities or results or strategies may differ significantly and materially from those reflected in or contemplated by such forward-looking statements. There is no assurance or guarantee that any such projections, outlooks or assumptions will occur.

Any reference to tax treatment will depend on individual circumstances and is subject to change. You should consult your own tax advisors to understand the tax treatment of a product or investment.

Whilst every effort is made to ensure that the information provided to clients is accurate and up to date, some of the information may be rendered inaccurate by changes in applicable laws and regulations. Partners Capital may have relied on information obtained from third parties and makes no warranty as to the completeness or accuracy of information obtained from such third parties, nor can it accept responsibility for errors of such third parties, appearing in this material.

Copyright ? 2024, Partners Capital Investment Group LLP

Within the United Kingdom, this material has been issued by Partners Capital LLP, which is authorised and regulated by the Financial Conduct Authority of the United Kingdom (the “FCA”). Within Hong Kong, this material has been issued by Partners Capital Asia Limited, which is licensed by the Securities and Futures Commission in Hong Kong (the “SFC”) to provide Types 1, 4 and 9 services to professional investors only. Within Singapore, this material has been issued by Partners Capital Investment Group (Asia) Pte Ltd, which is regulated by the Monetary Authority of Singapore as a holder of a Capital Markets Services licence for Fund Management under the Securities and Futures Act and as an exempt financial adviser. Within France, this material has been issued by Partners Capital Europe SAS, which is regulated by the Autorité des Marchés Financiers (the “AMF”).

For all other locations, this material has been issued by Partners Capital Investment Group, LLP which is registered as an Investment Adviser with the US Securities and Exchange Commission (the “SEC”) and as a commodity trading adviser and commodity pool operator with the Commodity Futures Trading Commission (“CFTC”) and is a member of the National Future’s Association (the “NFA”).