The Search for Structural Alpha

Welcome to this week’s edition of Cape May Wealth Weekly.. If you’re new here, subscribe to ensure you receive my next piece in your inbox. If you want to read more of my posts, check out my archive!

My personal investment strategy (and thus, the strategy I’ve implemented in my recent family office job) has been heavily influenced by a two-step process:

First, making the right choices on where you can generate Alpha - and where a Beta-oriented investment might be the better choice. As I wrote in What’s Your Alpha?, investors should question which asset classes provide an excess return on their time, money, and skillset. 

Second, if an asset class does not offer excess returns, they should look for a Beta-oriented investment.  Said investment should offer them cheap, simple and tax-efficient access to the long-term market return of the respective asset class.

While I deem myself capable of delivering Alpha in some asset classes, I am more focused on being an expert in beta-oriented investments - and more importantly, ways to further optimize them. 

That has put me on the ongoing hunt for something that I like to call Structural Alpha.

What is Structural Alpha?

As I had explained in What’s Your Alpha?, I define a ‘Beta return’ as the market return of a given market or benchmark. For example, if you invest in public equities, your “Beta return” is typically what you would get from a simple market-based investment, such as an index fund. 

Alpha, in return, is the excess return of an asset over the market Beta. Take public equities: Here, Alpha is generated through some form of stock picking, i.e. overweighting certain stocks, sectors or regions over another, with the goal of trying to pick the stocks which you think can outperform the market (or by underweighting those who think will underperform).

‘Traditional’ Alpha generation can be challenging. ‘Traditional’ Alpha is not constant, but volatile over different market cycles: While the best stock picker might outperform over longer periods of time, their return might deviate substantially from the Beta return of the underlying index. And of course, there is no guarantee that any Alpha-oriented investment will outperform over the long term.

Hence, I try to hunt for what I call Structural Alpha: 

Ways to generate excess returns (Alpha) over our long-term market return, without taking on the return volatility of traditional, Alpha-oriented investments.

Opportunities with Structural Alpha should aim to offer constantly positive outperformance as long as the underlying opportunity exists. However, given that benefit, Structural Alpha also tends to be small, frequently in the low single digit excess return per year. However, it’s worth noting that there is no guarantee for a positive return in every given period - typically, Structural Alpha is just an excess return over an index or benchmark, so if said index or benchmark is negative, so is your ‘advantaged’ investment. In this regard, it is different from ‘traditional’ sources of Alpha which might especially generate positive returns when its reference index is down.

Lastly, and as the sub-title to this newsletter might’ve already suggested, Structural Alpha arises from investments subject to what I call the Complexity Premium: Opportunities that offer said outperformance because they are either harder to understand than ‘conventional’ investments, and/or because they offer excess returns by taking on non-traditional sources of risk.

So which investment opportunities offer us Structural Alpha?

The Joys of Swap Outperformance

Investors in ETFs and index funds typically have the choice between two methods of replication. The most common one is physical replication, meaning that the ETF’s manager will try to replicate the index by buying the individual instruments (i.e. stocks or bonds) making up the respective index. Take the German DAX 40, where a physical ETF would buy each of the underlying 40 companies in their corresponding weight.

The other option is so-called synthetic replication. A synthetic ETF doesn’t buy the underlying instruments directly, but insteads enters into a derivative contract with a counterparty (such as a bank or asset manager), which guarantees the ETF the return of the corresponding index. Most synthetic ETFs don’t offer structural advantages: If anything, they might be more expensive than ETFs using physical replication, which benefit from additional sources of income such as securities lending.

But there are a number of exceptions. One of them is in the world’s biggest equity market: The USA.

In the US, foreign investors investing in the US are subject to withholding tax on dividends. As a result, they pay 30% tax on dividends paid by US-listed companies. With the MSCI USA’s dividend yield of 1,30% (as of August 2024), this results in a tax drag of almost 0,40% per year. One way to mitigate this is to pick an Irish-domiciled, physical ETF. Under a double taxation treaty between Ireland and the US, the withholding tax stands lower at 15%, meaning investors would be subject to only half of the aforementioned tax.

But synthetic ETFs go even further: Under the so-called US HIRE Act 871m, certain index derivatives are exempted from withholding taxes, meaning investors in the ETF using derivatives for their replication of the underlying index don’t have to pay those taxes at all. In other words, a hypothetical swap ETF is expected to generate a ‘Structural Alpha’ of 30% of the MSCI USA’s dividend yield, minus the cost of the ETF and swap.

Invesco, one of the largest managers of US swap ETFs, has a nifty chart in a whitepaper demonstrating these benefits. We can see the benefit of the average physical (often Ireland-domiciled), showing a 1-year ‘outperformance’ of almost 0,20%. For the swap-based ETF, this benefit is even more pronounced, ranging between 0,30% and 0,40%:

This outperformance seems minor, but of course, we have to consider compounding returns: Assuming an annual excess return of 0,30% p.a., the 10-year annualized return of the MSCI USA would go up from 12,9% p.a. (Source: MSCI) to 13,2% p.a., which over the 10-year-period compounds to a 10% total return difference. Not too bad!

But more importantly, is this potential for Structural Alpha when you view it in the context of a diversified equity portfolio. We touched upon this to some degree in last week’s newsletter: As of August 2024, MSCI USA made up over 64% of the MSCI ACWI. If we multiply this 64% weight with our expected Structural Alpha of 0,30% p.a., we would expect a benefit at the level of our equity portfolio of 0,21% p.a. And here, the magic of our search for Structural Alpha kicks in: Thanks to the continued drop in index fund and ETF management fees, it has become possible to build a portfolio that tracks the non-US parts of MSCI ACWI for less than the 0,21% in “fee budget” we have available from the Structural Alpha we can generate in our US allocation.

Or in other words: Thanks to the swap outperformance, we are able to build a purely passive portfolio that after swap benefits and costs might have a return equal to (or perhaps even in excess) of our costless benchmark. 

From personal, professional experience, I can indeed confirm that it is possible. But nevertheless, you might rightfully wonder: Where’s the catch?

Structural Alpha, Uncommon Risks

Experienced investors know that there is no free lunch - and the same thing is the case for investments offering Structural Alpha. From my experience, they tend to be subject to non-traditional risks, to which investors should pay special attention.

Let’s take another look at our synthetic US equity ETF. Fundamentally, our Structural Alpha comes from the aforementioned quirk of US tax law. If that quirk was to go away, we’d lose our Structural Alpha. Is that likely? Admittedly, I can not assess it, but it’s something to be considered.

Another example of a synthetic ETF that has historically provided investors with Structural Alpha is China. As the Financial Times wrote in 2022, hedge funds and other market participants were looking to short Chinese markets, which do not offer the short selling liquidity known from US and other regions. Hence, they looked to go short through derivatives, and the ETF providers offering the other side to these derivative trades were paid handsomely, with the swap outperformance (i.e. our Structural Alpha) going as high as 15 percent (!). However, the enhancement has fluctuated significantly, at times even turning negative, meaning that there might be times where we would actually underperform a non-advantaged ETF.

One should also highlight the general complexity of synthetic ETFs. Here, the risks lies in our counterparties: If they were to be unable to fulfill their duties of providing us with the return of the underlying index, we’d face the risk of tracking error, which stems from the difference in our desired index (MSCI USA, for example) and the underlying securities held by the ETF. While I personally find the default of a counterparty to be unlikely (given that they are some of the largest investment banks in the world), it is still possible, and once again, investors should know what can go wrong, and how it would affect their portfolio. I know family offices that are unwilling to take this risk, instead investing in ETFs using physical replication. (For more information,  Invesco’s whitepaper touches on some of the risks.)

Overall, the case for synthetic ETFs is a great example for the nuances of the Complexity Premium. If you want to conduct proper risk management for your synthetic ETF, you might be less concerned about the ups and downs of the market, or the risk of underperformance relative to the index. Instead, you’re concerned about potential changes to US tax law, or second-order effects to choppy markets, such as your swap counterparty becoming unable to fulfill its commitments. Not what you’d usually be looking out for.

Other Cases of Structural Alpha

Readers might wonder what other things I’ve found that offer Structural Alpha. And unfortunately I have to admit that my list, so far, is short. Besides China und US swap-based ETFs, some of the ideas I’ve considered and heard from fellow investors are as follows:

(Disclaimer: The following are all investments that I yet have to research as properly as I have synthetic ETFs. Any ideas below should definitely not be considered investment advice.)

CLOs: CLOs are a way to securitize a portfolio of high-yield loans. The portfolio is financed through a number of credit notes of varying credit rating - the higher the rating, the lower the return, but the earlier the repayment. While many family officers I talk to are most interested in the riskier parts of CLOs, I am intrigued by senior (AAA) tranches: From what I’ve seen and read, they offer similar returns, yet lower default risk relative to investment grade bonds of similar debt as investors are affected by portfolio-level default risk rather than single-company defaults.

MBS: Similar to CLOs, Mortgage-Backed Securities are securitizations of individual home loans. What I am intrigued by are so-called Agency MBS: Securitizations guaranteed by US government agencies, meaning that an investor might benefit from similar duration and credit quality over US treasuries at a slightly higher yield.

Credit-Linked Notes / Structured Products: Structured Products are essentially trades bilaterally negotiated with a bank or asset manager, in which they build you a customized financial product for a fee. Since they are issued by a bank or asset manager, you also take on the credit (i.e. default) risk of the counterparty - even if your trade goes the right way, if your counterparty defaults, you might lose your money. However, since this also allows the banks to (partially) refinance themselves, they typically offer an uptick in return, yield, etc. somewhat in correspondence to the counterparty’s credit rating. Some investors making use of this to take directional trades under benefit of Structural Alpha: For example, one told me that they had a bank build them a structured product delivering the return of a certain fixed income index at return greater than Delta >1 (i.e. Delta 1,2x, meaning they receive 120% of the upside or downside of the reference index) thanks to the credit component. In effect, they are able to boost their return by taking on the (otherwise unrelated) counterparty risk of a bank.

What do all of those ideas, as well as our swap ETF, have in common? They are perfect examples of our Complexity Premium: By taking on complexity (i.e. having to understand the many loans underlying a CLO or ETF), non-traditional risk (i.e. the swap ETF’s tax risk or the default risk of a bank even though the underlying trade isn’t subject to it), we are able lift our return without being dependent on the direction of the market.

Alpha … In the Right Structure 

Especially in asset classes where I don’t expect to deliver Alpha, aiming for a simple Beta exposure (i.e. through an index fund) or at least an over-diversified source of Alpha with the goal of receiving the market return should be the way to go for any investor.

Structural Alpha is the most natural evolution of this pursuit. In that regard, I always like to refer to my favorite Howard Marks memo, in which he tells the story of a pension fund manager who reached the top 10% of pension fund managers simply by aiming for returns that hit the second quartile (so between average and slightly above average) every year. Pursuing that strategy by sourcing Structural Alpha across your portfolio should hopefully let a long-term investor see a similar long-term ‘ranking’ among their fellow investors.

But Structural Alpha is typically the last optimization measure an investor should focus on. Too often do I experience that investors are looking for ways to beat the market through smart, complex trades like the ones I mentioned above (see Simple Investments, Complex Investments). But even the best, risk-free, complex trade doesn’t help if your asset allocation does not fit your Investment Objectives,or if your investment setup has some other avoidable yet impactful mistake. Or in other cases, if your tax setup results in you leaving avoidable euros (or dollars) on the table.

So please, look out for Structural Alpha in your boring investments (and please reach out to me if you have other ideas!) - but don’t miss the forest for the trees.

Liked what you read? If you enjoyed this piece, make sure to subscribe by adding your email below. I write about topics covering the world of family offices, asset allocation, and alternative investments.