Positioning Digital Assets in an Institutional Portfolio

Download as PDF

An investment in digital assets is an investment in the secular change of money, value, ownership, and business structures over the coming decades. This means digital assets should show low correlation vs other asset classes going forward and provide significant diversification benefit to a portfolio. We believe an allocation to digital assets should be a part of many modern portfolios, with such an investment diversified across a broad selection of promising digital assets.

Key Takeaways

Classic portfolio optimization techniques unsurprisingly suggest that adding digital assets to a portfolio may improve performance, but a long-term allocation should be fundamentally driven, and warrants consideration beyond historical performance.
Digital assets are additive to the risk-return profile of a multi-asset portfolio as their performance should show low correlation to equities, fixed income, and alternatives.
Investors should “get off zero” in their allocation to digital assets. An allocation of 1% - 10% of a well-diversified portfolio is a reasonable range within a diversified portfolio. Although an admittedly wide recommendation, investors’ goals and risk tolerances should dictate the precise allocation.

Disclaimer: The discussion contained herein is for informational purposes only, you should not construe any such information or other material as legal, tax, investment, financial, or other advice. Nothing contained herein constitutes a solicitation, recommendation, or endorsement to buy or sell any token. Nothing herein constitutes professional and/or financial advice, nor a comprehensive or complete statement of the matters discussed or the law relating thereto. You alone assume the sole responsibility of evaluating the merits and risks associated with the use of any information or content herein before making any decisions based on such information or other content. Investors should be aware that investing in digital assets involves a high level of risk and should be undertaken only by individuals prepared to endure such risks. Any forward-looking statements made are based on certain assumptions and analyses based on historical trends, current conditions, and expected future developments, as well as other factors that are believed to be appropriate under the circumstances. Such statements are not guarantees of future performance and are subject to certain risks, uncertainties, and assumptions that are difficult to predict. Please see Disclaimers for more information.

Introduction

The rise of a novel asset class is a rare occurrence. With it comes the difficult process of figuring out how to incorporate the new assets into existing portfolios. As we discussed in our piece on Nascent Markets, the early days of new asset classes can be tumultuous. However, significant, uncorrelated returns may be realized by taking on the risks of the new market. In considering a long-term allocation, the justification for taking such a risk must go past historical returns and the assumption that the past will resemble the future.

In this piece we contextualize the takeaways and potential issues with quantitative investment analysis techniques when applied to digital assets, and argue that the rationale for investing in digital assets should be rooted in a strong conviction in the usefulness of the underlying technology, and the uncorrelated determinants of its success vs established asset classes in the long run. Understanding both the quantitative and qualitative narratives surrounding digital assets builds a strong foundation for an investment in the asset class.

Executive Summary

Backtesting traditional portfolios with the addition of digital assets like Bitcoin naturally shows improved portfolio returns, given high realized returns vs traditional financial assets over the last 10 years. However, a long-term allocation to digital assets also warrants consideration of the fundamental drivers of the asset class.
The correlation between digital assets and equity markets has risen in recent years, but this is more a function of unique global macro factors post-COVID than of the asset class conforming to existing markets. We expect digital assets to exhibit low correlations to other markets going forward, given their unique value drivers.
The disruptive, technology-driven potential of digital assets make the asset class more akin to venture capital than to established, liquid markets. In this context, the asset beta of digital assets should show low correlation, in the long run, to most other asset betas in a portfolio.
The most important consideration for a digital assets allocation is the forward-looking, fundamental reasoning as to why a broad range of cryptocurrencies may be expected to continue outperforming other asset classes, and why the returns should be uncorrelated to other asset classes going forward.
Digital assets represent new technology that seeks to improve the capture and transfer of value in wide-ranging parts of the global economy, leveraging principals of decentralization alongside new methods of automation, such as smart contracts, and enabled by innovative incentive mechanisms. We see long-term secular shifts in consumer preferences and expectations concerning trust, accessibility / inclusion, privacy, and equality driving adoption of the asset class.
Investors should consider “getting off zero” in their allocation to digital assets. Digital assets may be additive to many modern portfolios, and the allocation should represent a diversified range of promising projects within the asset class.
Our analysis suggests an allocation of 1% - 10% of a well-diversified portfolio is a reasonable range to consider. We consider hypothetical, mean-variance optimized levels for different types of portfolios in the Appendix to this piece, but conclude that investors’ unique goals and risk tolerances must dictate the precise allocation.

Is Optimization Optimal?

Understanding the Shortcomings of Portfolio Optimization Models

In attempting to size an appropriate allocation to digital assets, research to date has focused on the mean-variance model of modern portfolio theory, as introduced by economist Harry Markowitz in 1952. Despite being an intuitive model (maximize return for a given level of risk), the practical usefulness of simple applications has been debated. A major point of consideration is that the model is highly sensitive to the inputs that represent estimates of expected returns, variances, and correlations of assets. These estimates are generally derived from historical data and can be reasonable measures of probable reward and risk assuming that the distribution of returns is normal and stable. In actuality, distributions of risk and return tend to be skewed and, furthermore, are not stable over time.

These issues are even more pronounced for digital assets. We have previously discussed how the returns of digital assets approximate a power-law type distribution, and that the long right tail of outcomes (i.e. the lack of a normal distribution) is central to the investment opportunity. Furthermore, the relative youth of the asset class means historical data on a broad set of tokens is limited to a handful of years. These years additionally coincided with an extraordinary period in markets from a macro viewpoint, given the negative shock of the COVID pandemic and the countervailing positive shock of the ensuing flood of monetary and fiscal stimulus. With such circumstances, using expected returns and volatility for digital assets in a model alongside stocks and bonds can produce specious results. We say this even though the results of traditional mean-variance optimization make a strong case for large allocations to crypto, given the low correlation and high returns of the asset class.

Exhibit 1: Mean-variance optimization can suggest high allocations to digital assets

Source: Outerlands Capital Research, S&P Global, CoinGecko, Yahoo Finance
Note: Expected return = average monthly return from Jan-2014 to Dec-2023; Expected covariance = sample covariance from Jan-2014 to Dec-2023

For example, we could construct a simple model of an “optimal” portfolio of Bitcoin, US Equities, and US Bonds. We use the most basic estimators (average historical returns for expected return and the sample covariance of the historical returns for the covariance), and sample data back to 2014. Our simple, unconstrained, mean-variance optimization for the highest Sharpe ratio results in a portfolio weighted 15% to BTC, 62% to equities, and 23% to bonds, with a Sharpe of ~1.13 (Exhibit 1).

While we agree with the takeaway (that is: allocate to digital assets), there are numerous points left unconsidered, which together make such an analysis an insufficient basis for an allocation.

Data since 2014 is brief in the context of financial markets, and captures only one major market cycle (the COVID-19 pandemic). On the other hand, Bitcoin (not to mention other digital assets) was an extremely niche asset for much of even this short period, well outside the investible universe of assets for most institutional investors. As such, behavior during the earliest years should likely be discounted.
Investors have access to a wide range of investments, and are unlikely to be optimizing across only three assets. Even small, retail investors now have low-cost, liquid access to a wide range of global markets via mutual funds and ETFs. Sophisticated investors have even more choice, with a deep market of alternative investments to choose from, including hedge funds, private equity, private credit, and venture capital, to name a few. Even within digital assets, using only Bitcoin is insufficient in reflecting the broader opportunity of the space.
While a high allocation to digital assets may well improve a portfolio’s expected risk/return profile, it is prudent to consider the risk contribution of such an allocation as well. For example, the 15% allocation in the model portfolio above is contributing 55% of the total risk of the portfolio (with 43% contributed by stocks, and only 2% by bonds). Although this skew of risk contribution is not unique to digital assets (stocks, for example, in a 60/40 portfolio contribute >90% of risk¹), it is worth considering as a parameter.

The question becomes - if we attempt to address these issues, will we actually end up with a better guide for how much to allocate to digital assets? That is, should the answer for how to position digital assets in a portfolio even come from portfolio optimization models? To this, our answer is: No.

We will go through some of the legwork of addressing the issues noted above and explaining how the results remain unsatisfactory, but will primarily address it in the Appendix to this piece. Instead, we will devote the next section to dissecting and contextualizing how digital assets have behaved vs other assets in recent years, to in turn help inform a forward outlook. Rather than reducing digital assets to one expected return figure, one expected risk figure, and a few cells in a covariance matrix, we look at how and why their behavior has evolved over time. In particular, we examine digital asset correlations in the context of broader trends in asset class correlations, the volatility of the asset class, and also benefits to diversification with digital assets.

This section will focus on the more financialized aspects of digital assets. In some ways treating them as just another widget to add to a portfolio. But this alone would, in a sense, be no better than the optimization exercises. We strongly believe that the most important consideration for a digital assets allocation is the forward-looking, fundamental reasoning as to why a broad range of digital assets could be expected to continue outperforming other asset classes, and why the returns should be uncorrelated to other asset classes going forward. As such, we spend the second part of this piece discussing what makes digital assets a unique, secular growth story. A thesis that, if any part of it has merit, should clearly warrant a non-zero allocation to digital assets.

Part I: Dissecting Trends in Digital Asset Price Behavior

What Has Happened to Digital Asset Correlations?

A core tenet of portfolio construction is to look for assets that have low correlation to other assets. Put enough low- or uncorrelated assets together and you have a portfolio well-equipped to provide returns while weathering the turbulence of markets. Bitcoin has long been heralded as a new breed of uncorrelated, ‘digital gold’, but that narrative has at times been questioned - both for Bitcoin and for other digital assets. With the potential diversification benefit in question, we examine how digital asset correlations have changed over time and why.

A Shift in the Macro Environment Has Impacted All Asset Classes

Discussion of digital asset correlations to other asset classes most commonly focuses on the correlation to the US equity markets. In particular, a popular topic has been how the correlation of digital assets to equities has risen in recent years (compared to the early days of the asset class), making them a less compelling addition to a portfolio. Looking at the numbers, it is indeed clear that the correlation of digital assets to the S&P 500 has generally been higher over the last 3-4 years, with a notable step change amidst the COVID-19 pandemic (Exhibit 2). However, the level of correlation has fallen considerably over the past year, and the story is also more nuanced when considering how a broader range of asset classes have behaved.

In fact, the increase in correlation of Bitcoin and other digital assets to US equities is mirrored in the correlations between many established asset classes. Running pairwise correlations of daily returns across asset classes including US bonds and equities, developed market (DM) ex-US bonds and equities, emerging market (EM) bonds and equities, and even private equity and venture capital (proxied by Refinitiv indices), shows that asset classes in the post-COVID world have decidedly higher correlations to each other (Exhibit 3).

This is important context to consider. It is not as though Bitcoin or digital assets became more correlated to traditional markets due to fundamental changes in the asset class, or a unique shift in investor behavior with regards to the asset class alone. The increase in price co-movement says more about the macroeconomic picture than about the fundamentals of digital assets, or any particular asset class.

The regime shift in correlations can be traced to the profound impact of COVID-19, and in particular the responses of global central banks and governments. Since 2020, the global financial system has experienced significant economic disruption, beginning with unprecedented, coordinated monetary easing and fiscal stimulus from central banks and governments across the world. The sharp uptick in inflation that followed has now spurred one of the most aggressive tightening cycles of monetary policy in decades. The initial flood of liquidity into financial markets via zero-percent interest rates, massive asset purchases, and government stimulus checks drove both amateur and professional investors into a hunt for returns that sent them deep into speculative assets and eye-watering valuations. Unsurprisingly, inflation rose on the back of this dramatic easing in financial conditions, but central banks were seemingly left surprised when it did not come back down per their expectations, prompting a delayed and aggressive tightening of monetary policy.

Exhibit 4: A negative stock-bond correlation has been taken for granted, but has not always been the case, especially when inflation was higher

Source: Outerlands Capital Research, Robert Shiller, Macrotrends, US Bureau of Labor Statistics

The impact of the macroeconomic environment (particularly inflation) on asset prices can be most vividly observed in the correlation between stocks and bonds, which has flipped from negative to positive recently and continues to increase. Looking at a near century-long history of stock/bond correlation (Exhibit 4) shows that while a negative correlation has been the norm since ~2000, for much of the 20th century correlations were positive. This breakdown of asset class relationships that were taken for granted for decades has caused much consternation in the world of asset management – even prompting many to question the appropriateness of the classic 60/40 stock/bond portfolio in this new macroeconomic regime.

An in-depth analysis of monetary policy, inflation expectations, and expected returns across asset classes is out of scope for this piece. However, we stress that the recent behavior of digital assets vs other asset classes must be contextualized in the state of the macro environment and its effect on the behavior of all asset classes.

Correlations Beyond Equities

It is also important, from a portfolio perspective, to look at how digital assets correlate to other asset classes. Modern institutional portfolios comprise much more than just domestic bonds and equities, and many now have significant exposure to global markets and alternative investments like private equity, private credit, real estate, venture capital, and hedge funds. Exhibit 5 explores some of these correlations on a quarterly basis since 2016 - showing generally low correlation of digital assets across the board, particularly vs fixed income, alternatives, and private assets. Notably, over this period, digital assets were relatively uncorrelated to venture capital, which we believe shares the most similarity to digital assets from a fundamental perspective. In comparison, equities show relatively high correlations to corporate bonds, real estate (REITs), and private investments, further supporting the idea that many traditional asset classes have been positively correlated in recent years.

Exhibit 5: Digital assets generally have low correlations to traditional and alternative asset classes

Source: Outerlands Capital Research, CoinGecko, Yahoo Finance, NASDAQ, MSCI, S&P Global, Credit Suisse, Prequin, FTSE, ICE

We caution against reading too much into point-in-time statistics on correlations across an arbitrary time horizon. Relationships since 2016 are not a particularly long estimation window, especially using quarterly data. In our table, these constraints were informed by data availability (with Ethereum, for example, not existing until 2015) and the lower reporting frequency of private asset indices. With these limitations in mind, the analysis does still reasonably suggest that, over the long run, digital assets have not just mirrored equity markets or other established asset classes. This supports the idea that the inclusion of digital assets in a diversified portfolio should improve its expected risk/reward profile.

Exhibit 6: Digital asset correlations are not stable, but have generally been low vs other asset classes

Source: Outerlands Capital Research, CoinGecko, Yahoo Finance, S&P Global

Rolling correlations can help further elucidate these relationships and offer more context than just the snapshot. Exhibit 6 takes a subset of asset classes for which we have higher frequency data and compares the general trend of digital asset co-movement by “stacking” the trailing 12 month correlations of each asset class vs an equal-weight portfolio of the top 100 cryptocurrencies. The value that results from summing the correlations can’t be taken to represent the true correlation (the sum could exceed +1 or -1), but it shows the relative strength of positive vs negative correlations of digital assets to other asset classes. Overall, the average of these correlations has tended to remain positive but low, and have not been stationary. In the context of portfolio optimization, this means that estimates for correlation are quite sensitive to the sample period, and vary over time.

Volatility in Digital Assets - Not What it Once Was

Volatility is another factor top of mind in a nascent asset class, with investors rightly concerned about the additional unpredictability of returns. However, volatility is not what it once was in the crypto markets. One sign of the maturing of the market is the broad decline in overall volatility of the asset class, which at once denotes the presence of a growing investor base and in turn attracts further new investors. Since 2018, the price volatility of digital assets has clearly trended down which is an important sign of increasing stability (Exhibit 7). This decline in volatility is true of major cryptocurrencies like BTC and ETH but is also true of the universe of smaller tokens, reflected in indices such as S&P’s Broad Digital Market ex-Mega Cap index.

Compared to an index of established US equities (a la the S&P 500), digital assets will unsurprisingly still look quite volatile. However, the absolute level of price volatility of BTC and ETH is now on par with some of the most popular tech stocks of the public markets, such as auto-maker Tesla (TSLA) and chip-maker Nvidia (NVDA) (Exhibit 8) – popular portfolio overweights of discretionary equity fund managers.

While the risk profile of digital assets is clearly different from that of traditional public corporations, it should not be a foregone conclusion that digital assets are simply too volatile to invest in. As we discuss in the next section, we additionally see the potential of diversification among digital assets to help lower volatility.

Diversification - Still a Free Lunch

We have written at length about the benefits of diversification in digital assets, but the point bears repeating. Many recommendations for sizing a crypto allocation consider only adding Bitcoin to a portfolio. Bitcoin is a very compelling asset, but it represents only one part of the digital asset narrative. The thesis for Bitcoin is generally focused on its use as a store of value, made unique by the scarcity built into its design and its highly decentralized nature. However, this only touches on some of the points we discussed as secular drivers of growth in digital assets.

Even though Bitcoin is the poster child of crypto, it does not solely provide access to the full thesis of digital assets. An allocation only towards Bitcoin represents a concentrated exposure to a very specific thesis. Just as all of US equities cannot be proxied by an investment into only Apple, the potential of the digital assets markets can’t be proxied by an investment into only Bitcoin.

A common misconception is that different cryptocurrencies provide no diversification benefit in a portfolio. This is not true. Exhibit 9 shows clearly how the volatility profile of even a naive basket of the 100 largest cryptocurrencies over time is significantly lower than the median constituent of said portfolio on its own. We explore this dynamic further by running correlations between monthly returns since 2019 of all pairs of cryptocurrencies that satisfied a minimum trading volume requirement. The distribution of the results clearly shows that, while cryptocurrencies are generally positively correlated, it is a relatively weak relationship. The average correlation between pairs (~55,000 unique pairs analyzed) is ~0.4. This happens to correspond extremely closely with the result of the same analysis on paired constituents of the S&P 500 (~120k pairs), which also averages ~0.4 (Exhibit 10).

The diversification benefit within digital assets is real and should not be ignored. We argue that the best way to gain exposure to the themes of the digital asset space is via a broadly diversified but actively managed portfolio. Such a portfolio could provide exposure to many different use cases and value drivers, while reducing the risk profile compared to concentrated portfolios.

So What?

With these quantifiable considerations under our belt, we turn now to the “So what?”. Is any of the context around investment metrics really enough to justify an allocation to digital assets? It depends on perspective, but we believe there is much more to it. For the already-convinced digital asset enthusiasts, the compelling risk-adjusted returns of crypto are just icing on the cake. However, to the newly interested or skeptical, we think it is critical to understand the fundamental, qualitative thesis that we believe underpins an investment in digital assets.

From our perspective, the most important consideration for investing in digital assets is not found in historical data on correlations or returns. A point from Ray Dalio’s piece “Paradigm Shifts” comes to mind:

““Theories about how to invest changed frequently, usually to explain how the past few years made sense even when it didn’t make sense. These backward-looking theories typically were strongest at the end of the paradigm period and proved to be terrible guides for investing in the next decade, so they were very damaging. That is why it is so important to see the full range of past paradigms and paradigm shifts and to structure one’s investment approach so that it would have worked well through them all. The worst thing one can do, especially late in a paradigm, is to build one’s portfolio based on what would have worked well over the prior 10 years...””

— Ray Dalio

What matters most is the forward-looking, fundamental reasoning as to why a broad range of cryptocurrencies may outperform other asset classes, and why the returns should be uncorrelated to other asset classes going forward. We explore this in the next section.

Disclaimer: Investors should be aware that investing in digital assets involves a high level of risk and should be undertaken only by individuals prepared to endure such risks. Any forward-looking statements made are based on certain assumptions and analyses based on historical trends, current conditions, and expected future developments, as well as other factors that are believed to be appropriate under the circumstances. Such statements are not guarantees of future performance and are subject to certain risks, uncertainties, and assumptions that are difficult to predict. Please see Disclaimers for more information.

Part II: Digital Assets for the Long Run: Uncorrelated Determinants of Success

We have recently written about how digital assets are a nascent market and have gone through similar trials as other asset classes did (albeit decades or centuries earlier). The youth of digital assets as financial instruments is amplified by the fact that the underlying technology itself (to which the market provides exposure) is new and evolving. The principles of Distributed Ledger Technology have been around for decades², but the modern structure and use of public blockchains is decidedly still a developing idea. In particular, the way in which the technology has been built with a focus on disintermediation and trustless interactions is quite novel, as are the many different attempts at creating token-based incentive mechanisms that distribute value more equitably across all stakeholders.

The nascency of the market, the power of the technology, the scope of its ambition, and the relative immaturity of its applications make digital assets a unique, secular growth story. This transcends just plugging Bitcoin into a portfolio as another financial widget. It makes digital assets a unique opportunity to invest in technology that could have a widespread and fundamental impact on the global economy, similar to how the Internet has changed the world in the decades since its invention.

Exhibit 11: World equity markets have grown increasingly correlated

Source: Outerlands Capital Research, MSCI
*DM = Developed Market; EM = Emerging Market; FM = Frontier Market*

Established financial markets today are generally driven by similar macro risk factors, such as economic growth, central bank policy, interest rates, inflation, geopolitical events, and others. In an increasingly global, interconnected economy these risk factors have even blended across country lines, as evidenced by increasing correlation of world equity markets (Exhibit 10).

While digital assets will of course be impacted by macro factors as well, we believe the main fundamental drivers of the asset class are unique and uncorrelated. A thoughtful investment in digital assets is an investment in the secular change of money, value, ownership, and business structures over the coming decades. In this context, growth of the market will be driven by several key themes that can develop independently of global macro. We will explore these next.

Performance Will be Driven by Adoption and Development

The success of digital assets hinges on the adoption of blockchain and token-based products by the broader consumer and business market. This in turn hinges on the continued development and refinement of the technology to meet the aspirations of applications. As with any other product, it must first and foremost be effective in solving a significant problem relevant to a broad demographic in order to attract users. While it is certainly no small challenge to design a product that people or businesses need or want, adoption will ultimately be a function of the creativity and skill of builders in the space, rather than generic macroeconomic conditions.

Fortunately, the pace of development is a distinguishing feature of the blockchain space. Digital assets are frequently linked to projects that develop on an open-source model, based around composability, strong community engagement, and an open market for talent. This development model is flexible and agile, allowing projects to start slowly and cost-effectively but also quickly accelerate and optimize development. A relatively short feedback loop from developer to user will also help projects iterate and achieve product-market-fit.

With regards to adoption, we see several core features of blockchains and cryptocurrencies that can lead to desirable applications with broad usership. While “market” sentiment can shift, these value propositions are fundamental aspects of the technology that should underpin the growth of the market as a whole:

Reduced friction in value transfer and ownership: Bitcoin (token BTC) led the way in introducing a decentralized, state-agnostic currency that functions as a means of payment and store of value. While Bitcoin remains the most valuable cryptocurrency, the evolution of blockchain technology has significantly advanced what is possible in terms of value transfer and ownership, particularly with regards to speed and cost.
Automation of processes through smart contracts and decentralized code execution: A profound and disruptive value proposition of blockchains is their ability to incorporate smart contracts. These contracts can be defined in a myriad of ways and automatically execute when predetermined conditions are met. In turn, the power and permission to act is moved from discretionary, centralized actors, to a systematic, decentralized network that runs on parameters set transparently in code. This facilitates transactions and agreements between parties without the need for intermediaries or establishment of mutual trust, reducing transaction costs and limiting the potential for the agreement to be manipulated by any one party or external actor.
Innovative incentive mechanisms that enable growth and efficient resource allocation: Blockchains, and the applications built on them, have a unique ability to develop micro-economies around cryptocurrencies, or tokens, which can enable network growth and the efficient allocation of resources. At the blockchain or protocol layer, the economics of this generally revolve around the use of native cryptocurrencies to pay for transaction/usage fees on the blockchains, which are paid out to the miners, validators, and stakers that ultimately run and secure the ecosystem. However, similar incentive mechanisms can be structured at all levels of the crypto ecosystem. This is manifested in micro-economies tailor-made to incentivize all participants, from service providers to the end user. Such structures may revolve around tokens used as payment mechanisms, loyalty programs, representation of governance rights, or any number of other utilities.

Furthermore, we expect there to be more direct near-term catalysts that should promote mainstream adoption of blockchain-based projects, including advancements in account abstraction and UI/UX, zero knowledge proofs that can help address privacy concerns, and clearer government oversight and regulation. Ultimately, blockchains will be a success when they are truly invisible – integrated into the way of doing everyday business without any technical barriers to entry. Behind the scenes, zero knowledge cryptography can set the stage for systems that preserve user privacy, while providing enough transparency for regulation. That regulation, in turn, is an opportunity, not a hurdle. The focus of rule-making bodies across the world (including the UK³, EU⁴, Singapore⁵, Hong Kong⁶, Japan⁷, Abu Dhabi⁸, Dubai⁹ and others) on the development of crypto regulations is a valuable sign of acceptance by governments of the technology’s potential.

Even as these catalysts develop, adoption is already underway. Estimates by Crypto.com Research suggest there are 516 million owners of crypto globally, as of June 2023. This represents a 21% increase in the first half of 2023 alone, and ~400% increase since 2020. Notably, this puts crypto ownership at approximately 6% of the world’s population after ~15 years of existence (starting at Bitcoin’s inception in 2008). These figures are not just spurious counts of immaterial ownership. The report uses on-chain interactions with 23 different exchanges to estimate ownership, with addresses limited to the Bitcoin and Ethereum networks. This means that only users who deposit or withdraw funds from another exchange or self-custody wallet on those blockchains are counted. Given that other popular networks exist, and the methodology doesn’t account for users that buy and hold on a single exchange, actual crypto ownership could be significantly higher.

For comparison, the Internet was conceived in 1969, but the breakthrough of TCP/IP came years later and wasn’t live until 1983, which many consider the true birth of the Internet¹¹. Retail adoption did not emerge until the early 1990s and grew slowly at first, meaning global adoption 15 years after the TCP/IP breakthrough was just ~3%¹². This suggests a much more rapid adoption of crypto, which is remarkable given the complexity of the technology, as well as the prolonged bouts of negative press it has experienced. However, crypto has benefited from a population that is more attuned to technological progress as a baseline, particularly in the digital realm, as well as the breadth of its use cases. The technology behind digital assets is also far from finished evolving. Just as the Internet saw a step change in capability with the introduction of TCP/IP, crypto could see major, fundamental technological leaps forward in the coming years as well.

Technology is not the only driving factor of the crypto thesis, however. As we explore next, there are important thematic shifts that also enhance the staying power of the narrative.

Adoption Will be Driven by Secular Shifts in Consumer Preferences and Expectations

We expect the development and adoption of applications based around digital assets will ultimately be driven by secular shifts in consumer preferences and expectations. At the heart of digital assets lies a vision for the Internet (generally referred to as Web3) that fundamentally restructures relationships around critical questions such as:

What is the future of money
How users expect to engage with businesses
What they are willing to pay for services
To whom value creation accrues
How digital identities are managed

And ultimately…

Who (or what) can be trusted?

Many of the answers to these questions center around two prominent concepts espoused by creators and users of solutions built around digital assets: 1) Disintermediation and 2) Trustless interactions.

Disintermediation is based on the idea that economic value can be created by replacing or removing middlemen in a process. Commonly associated with supply chain management, the principle is nonetheless relevant throughout commerce and other services, such as government. Disintermediation has already been a decades long source of value creation by reducing costs and increasing efficiency, but blockchain and smart contract technology has the potential to reinvent its implementation and do so across nearly all sectors of the economy. Such process improvement is inherently secular in nature, as businesses will continually seek to optimize their operations.

Trustless and frictionless interactions are ultimately the goal of these disintermediated systems powered by blockchains and digital assets. In existing practice, disintermediation can reduce reliance on middlemen while still ultimately relying on centralized entities as the arbiters of “truth” in the system. In contrast, decentralized blockchain technology actually allows the network itself, controlled by no single user, to serve as the sole source of truth. This is a powerful concept that could fundamentally alter how people interact and transfer value. Smart contracts are a foundational block in making trustless interactions possible, with logic-driven code, rather than intermediaries, directing processes and making decisions. In this way, digital assets offer exposure to the value-creation potential of disintermediation accomplished in an innovative way that can scale and align with shifting consumer preferences and expectations.

Why Will This Shift Happen?

The pivotal question is of course why this shift in expectations should and will happen. This is a difficult question to answer. Even if an alternative to the status quo is “better”, there are always switching costs to consider, along with general apathy by some towards the perceived problems with present solutions. However, we do see several core fundamental drivers of change that are likely to be exacerbated by shifts in geopolitics and demographics. We summarize these in a few general themes experiencing shifts in public opinion:

Trust: Fundamental questioning of established institutions and whether they have the best interests of all community members, users, and stakeholders in mind. This includes attitudes towards trust in governments, corporations, and other individuals.
Accessibility and Inclusion: Demand for access to a wider array of products and services in a frictionless, cost-effective manner for people across socioeconomic statuses, and independent of geography.
Privacy: A concern for protection of the sovereignty of individuals, their data, their interactions, and their property.
Equality: The notion that inequality in current economic systems has progressed too far; That centralized entities and middlemen have extracted too much value off the backs of their customers or constituents.

We see alignment between the challenges facing humanity and the solutions that digital assets promise. Blockchain technology will not solve all challenges, but there are undoubtedly deep impacts that it can make as part of broader efforts to affect changes around the themes listed above.

These themes can be seen as counter-cyclical – poised to accrue benefits as traditional systems struggle. The success of digital assets could pose real challenges to the beneficiaries of the status quo. In that way, an allocation to digital assets could be seen as a hedge against the impact of secular shifts on established asset classes. This is not to say that the crypto thesis can only see success at the expense of incumbents. There are many ways in which the technology behind digital assets can grow synergistically with existing processes, ultimately expanding the pool of economic benefits rather than cannibalizing it. However, it remains relevant, as investors with patient and prescient perspective, to consider the long-term evolution of the global economy and how digital assets are positioned to benefit.

Digital Assets Have a Global Userbase

The final, perhaps most basic, driver of digital assets’ uncorrelated growth is the fact that it is a truly global asset class. The idea that a digital asset, such as Bitcoin, can be exactly the same, all around the world, is a subtle yet powerful concept. The Internet connected the world and gave rise to the information economy and the platform economy. Digital assets, in turn, are poised to be the underlying infrastructure and economic architecture for digital goods, ushering in the evolution of the ownership economy. Blockchains and digital assets are technologies that enable groups of distinct actors to better achieve their stated purpose – with the ability to do so in a way independent of geography.

The global, open-source, collaborative nature of crypto could be transformative in expanding the availability of globally integrated services that reach much broader demographics. From cost-effective remittances for the millions of immigrants worldwide, to a store of value in unstable economies, digital assets have already proven themselves as effective tools for interaction in a shifting global economy. Such a truly global scope is a rare quality of any vision, and meaningfully differentiates the potential of digital assets vs extant asset classes.

Allocating to Digital Assets: Get Off Zero

As we have stated several times, we view an investment in digital assets as an investment in the secular change of money, value, ownership, and business structures over the coming decades. If any part of this thesis has merit, then a non-zero allocation to digital assets should be part of a well-diversified portfolio.

Sizing that allocation more precisely is a difficult question. As we discussed at the beginning of the piece, using backward looking portfolio optimization techniques should not be the only way to think about sizing a position in crypto. The past has been volatile, “history” is short for digital assets, and the time period for estimating expected returns and covariances was massively skewed by a globally disruptive event (COVID-19). The answer then comes down to investor type, investment goals, and risk appetite. We firmly believe it should not be 0%, while at the same time acknowledging that a 50% concentration would not be prudent for many investors. Realistically, a healthy allocation could be anywhere from 1% - 10%. While admittedly a large range, more precise recommendations assume too much about investor portfolios and preferences. For hypothetical optimization scenarios, see the Appendix.

We can however look to allocation trends across other asset classes for some context. In particular, our allocation recommendation can be supported by allocation trends to venture capital, which has some similarities to digital assets, while also considering the liquidity benefit of the asset class.

Digital Assets in a Venture Capital Context

The digital assets thesis is rooted in the emergence of disruptive, novel technology. This makes it akin to venture capital, with business models built around disruption and scale, alongside an evolving journey to product-market fit and profitability. However, digital assets come with the significant benefit of liquidity - a characteristic uncommon for most investments in early stage projects. This added liquidity of investing in tokens has even given many long-biased digital asset strategies the moniker “liquid venture”.

While there are still structural differences between digital assets and venture capital, venture capital enjoys a much longer history as an asset class. As such, we can look to it for some guidance on how managers have allocated to disruption in the past. Data by Cambridge Associates as of 2019 suggests that, across endowments and foundations, the mean allocation to VC sits at ~6%¹². This proportion has increased from only ~2% in 2001. Notably, the institutions that rank in the top decile of 20-year investment performance have consistently had a much higher allocation to VC, with the mean for this decile coming in at ~15%, up from ~8% in 2001. These figures are supported by other research from SVB/Campden Wealth¹³, which found that family offices in North America have an average VC allocation of 15% (13% for rest-of-world), and that larger family offices allocate more to venture (16% for $500MM+ vs 13% for <$500MM).

Specific data points are difficult to come by, as VC is often rolled up within a broader “private equity” allocation, but there are some anecdotes to consider. CalPERS, the largest public pension scheme in the US, has recently upped its strategic VC exposure from 1% to 6%¹⁴, following an internal review. Other state pensions have also made investments into VC, and even cryptocurrencies¹⁵. The Yale Endowment, one of the largest private school endowments, recently set¹⁶ a target of 23.5% for venture allocation. Various Canadian pension plans also invest actively in the VC space¹⁷, with the eight largest collectively growing their funding for the asset class from ~$4.3bn in 2019 to over $26bn in 2021 (PitchBook).

While allocations vary, the thesis of investing in disruptive technology as an uncorrelated bet (that is, diversifying to a portfolio) has clearly already caught on at many institutions. We argue that an investment in digital assets should also be thought of as an investment in disruptive technology, and as such could constituent a portion of, or an addition to, investors’ venture portfolio.

Liquidity and Transparency

One of the major distinctions between digital assets and venture capital (and many other private assets) is the liquidity provided by direct investment into digital assets. While liquidity is certainly not like public equity markets (outside perhaps BTC and ETH), there is a robust infrastructure of exchanges and over-the-counter trading platforms to move size in the market. Despite the decline in overall volumes in 2023, spot market volume across major crypto exchanges has still averaged north of $10bn/day¹⁸. Comparatively, the market for secondary trades (one in which an investor buys/sells a stake in a fund from/to another investor) in private equity, of which venture capital is a subset, totaled ~$100bn for all of 2022 (<$500mm/day on average, concentrated in large, infrequent transactions). Venture capital made up only 3-5% of that volume¹⁹. Liquidity also provides more immediacy on asset valuations, with market-based pricing for most digital assets available in real time. While this is generally seen as a positive, we do recognize that daily pricing means coming to terms with a more accurate (and generally higher) read on the investments’ volatility vs vehicles priced on a quarterly, or even annual, basis. We stress that this is the result of perception, however.

Realistically, the risk and volatility of venture capital investments in crypto businesses are similar to that of liquid digital assets, especially considering how much of VC bets in crypto are ultimately also tied to tokens. Such exposure could be in the form of Simple Agreement for Future Tokens (SAFTs), locked allocations of tokens, or equity investments in companies whose core value determinant is closely tied to a token. Private, illiquid assets can in effect show smoothed returns and “lower” volatility, but this is ultimately a veil covering the reality of risky investments. Cliff Asness of AQR²⁰ has written recently on this phenomenon of “Volatility Laundering”, and the hazards of understating risk.

Equable Institute also discussed in their State of Pensions 2023 report²¹ that valuation risk is a growing concern for some institutional portfolios, highlighting that: As of 2022, as much as one-third of the $4.8 trillion in assets that pension funds reported owning were based entirely on non-transparent valuation approaches from asset managers (not market-based prices like stocks). Note that this is not a criticism of private investments, venture or otherwise. Illiquidity and valuation risk can be acknowledged and managed. However, it is also a risk factor that is at least partially mitigated by investments into liquid digital assets, particularly if the goal is to gain exposure to venture-like and/or technology assets.

Liquidity is important not only in a downside scenario, but in an upside one as well. The high return potential of alternative asset classes can also lead to situations where an allocation has grown too large relative to the portfolio’s target weights. In such situations, it may be prudent to trim exposure and return the portfolio to target weights. However, this is not straightforward for asset classes that lack liquidity, meaning positions may need to be offered at a discount^22,23, to entice buyers, while also generally paying significant spreads to intermediaries who broker such trades.

Digital assets are traded 24/7, around the globe, on sophisticated infrastructure; offering liquidity and valuation transparency that even many established liquid markets do not.

Conclusion

Digital assets today represent a unique and compelling opportunity to invest in the next major evolution of the Internet and the future of money. Technologies with potential as transformative as blockchain and cryptocurrencies come about very rarely, and even rarer is the ability to directly invest in early-stage innovation without liquidity constraints. An investment in digital assets is an investment in the secular change of money, value, ownership, and business structures over the coming decades. This means digital assets should show uncorrelated performance vs other asset classes going forward and provide significant diversification benefit to a portfolio.

The market for digital assets has been digesting growing pains of recent years, resulting in a period of lower sentiment and valuations. However, the core value propositions of digital assets continue to solidify and offer vast scaling capacity. Decentralized, permissionless networks run on blockchains and incentivized by digital assets can at once enable entirely new business models and operational structures, while remaining flexible enough to build into existing businesses and improve processes and output.

As the underlying infrastructure becomes faster, cheaper, and more powerful, use-cases and applications will continue to grow. Financial and human capital has steadily flowed into the crypto space, while adoption is expanding at established corporations and in broader social relevance. The build-out of regulation and increased government acceptance will also serve to reduce uncertainty and unlock further growth. Jurisdictions pioneering regulation include the UK, the EU bloc, Abu Dhabi, Dubai, Singapore, Japan, and Hong Kong. At the same time, fundamental innovations around privacy, security, and user experience will help crypto push through into the mainstream.

These underlying premises should inspire investors to consider a non-zero allocation to digital assets. Digital assets can be thought of similarly to venture capital, as a bet on emerging, disruptive technology that is uncorrelated to broader macro themes. We believe an allocation of 1% - 10% in a portfolio could be prudent and provide attractive diversification benefits. While broad, the specific allocation should be based on investors’ unique goals and risk tolerances. In assessing an allocation, we encourage investors to look past the performance of recent years, and think critically about their motivations for investment, the potential risks, and also the potential reward. Diversification within digital asset strategy implementations has so far been underutilized, and we see actively managed, broadly diversified portfolios as compelling allocation opportunities.

Outerlands Capital expects significant growth in the scale and value of the crypto market and views the present as an opportune time to invest. Whether at a family office, pension, endowment, or other institution, we believe digital assets should play an important role in many types of portfolios going forward.

Appendix: Toying with the Optimization

In this section we will attempt to address the shortcomings of commonly employed portfolio optimization in relation to digital assets. Although it makes a compelling case for digital assets, we see an analysis rooted in historical data as an interesting, but ultimately insufficient basis for the inclusion of digital assets in a portfolio. However, going through the actions of “improving” the analysis can shed light on further considerations. As we go through the analysis, a common takeaway will be that these issues are more easily identified than dealt with, but it is still worth the effort.

To recap, our major issues with many existing studies is:

Insufficient historical data
Limited overall asset selection, even within digital assets (just Bitcoin)
Unexamined risk contributions

We’ll examine points 1 and 3 first for simple three asset portfolios of stocks/bonds/Bitcoin, and then repeat our analysis first by simply broadening our digital asset allocation past Bitcoin, and then with an expanded menu of established investment options, addressing point 2.

Historical Data

The historical data hurdle applies to any combination of assets we analyze and we see two major challenges:

We would ideally like to use as much historical data as possible to derive our estimates of model inputs (expected return, standard deviation, covariance). For stocks and bonds this data is easy to come by, but Bitcoin did not even exist until 2008, with reliable, continuous price data not available until a few years later.
If we maximize our historical data on Bitcoin (back to say 2010), we have to assess whether early observations on price movement and returns, when Bitcoin was a highly niche asset, are even indicative of future behavior. The handling of this can have a major impact: Bitcoin’s average monthly returns are 15.5% if taken since mid-2010, 6.9% since 2015, and 5.8% since 2020. Volatility over the respective periods was 54%, 22%, and 21%.

That is, we at once want to use more historical data, but at the same time (for Bitcoin and other digital assets), we should probably use less.

The question becomes: At what point does the data for Bitcoin become “representative”? To avoid a completely subjective assessment, we use Bitcoin’s market cap as a proxy for maturity - comparing it to the minimum market cap requirement to be included in the S&P 500. That requirement has risen over time, but Bitcoin’s market cap has continually exceeded it since 2016 (when the requirement was $5.3bn²⁴). As such, we will target using Bitcoin’s data since 2016 in the analysis to follow.

While this captures a reasonable time-frame for Bitcoin, it remains a very short period for stocks and bonds. To address this, we propose using a longer history for stocks and bonds (since 2000), while relying on the shorter data period for Bitcoin’s expected returns, standard deviation, and covariance with stocks and bonds.

With this in mind, we can start looking at the new results of the mean-variance optimization.

Three Asset Portfolio: Bitcoin / Stocks / Bonds

Under the parameters for historical data derived above, we first take a look at a mean-variance optimized portfolio of only stocks and bonds, to establish a baseline and because it reveals interesting considerations in the use of mean-variance optimization.

Exhibit 12: A simple stock/bond portfolio optimization using data since 2000

Source: Outerlands Capital Research, CoinGecko, Yahoo Finance, Bloomberg

With only stocks and bonds to pick from, the result of a mean-variance optimization is a portfolio allocated 10% to stocks and 90% to bonds (Exhibit 12). The skew to bonds may be surprising when compared to the common mental anchor of a 60/40 stock/bond portfolio. However, it is supported by a Sharpe ratio of bonds which was double that of stocks over the sample period (since 2000). Practically speaking though, this heavy allocation to bonds means annual expected returns fall short of even 5%. Recall that the mean-variance optimization is solving for risk-adjusted, not absolute, returns, and portfolio theory suggests employing financial leverage to achieve the desired level of return from the mean-variance optimized portfolio.

In practice, levering a portfolio has its own difficulties and risks, so if an investor has a particular return target hurdle, that has to be added as a constraint to the model. Targeting 7% annual returns (an oft-used base case assuming +5% over a 2% inflation baseline) in a stocks/bonds portfolio gets us back into more familiar territory, with 70% allocated to stocks and 30% to bonds, but with a decidedly lower Sharpe ratio of 0.63. Despite not being the unconstrained, optimal solution, the 70/30 portfolio that achieves a 7% expected return is much more intuitive and practical. As such, we’ll use it as the base case for comparing what the addition of Bitcoin does for risk and performance²⁵.

Starting with the 70/30 portfolio, and allocating just 1% to Bitcoin (taking equal weight away from stocks and bonds), improves the Sharpe ratio from 0.63 to 0.69, while also raising expected returns to 7.9%. Adding on to the Bitcoin allocation (while continuing to take away equal amounts from stocks and bonds), continues to improve the Sharpe ratio of the portfolio, up until a 36% allocation to Bitcoin (with 52% to stocks, 12% to bonds, and a Sharpe ratio of 1.13), after which the Sharpe ratio starts to fall.

Exhibit 13: Risk contribution in model portfolios based around 70/30 stock/bond allocation

That 36% is an untenably high allocation to Bitcoin, but even for smaller allocations, investors should consider not just the capital weight allocated to the asset class, but also how much that allocation is contributing to the overall risk of the portfolio. This relates to point 3 on our list of considerations: risk contribution. As discussed earlier in the piece, even in traditional portfolios the risk contribution will skew towards the risker asset. Equities, for example, contribute 98% of the portfolio’s risk in our quasi-optimized 70/30 portfolio of stocks and bonds (Exhibit 13). If, as above, we added Bitcoin at the cost of equities and bonds until the Sharpe ratio reached a local maximum, that 36% allocation would represent 86% of the portfolio’s risk.

Why does this matter? One might assume that the weight by capital allocated represents a corresponding amount of the portfolio’s risk. In actuality, the allocation is contributing significantly more to the risk of the portfolio - past what an investor may be expecting or be comfortable with. Practically speaking, it means an asset could contribute more to losses than its capital weight suggests²⁶.

To manage the expected risk contribution, an investor may want to place an explicit cap on the risk contribution of Bitcoin in the model. Limiting it to, say, 15%, results in a 4.2% allocation to Bitcoin, assuming the allocation is taken equally from stocks and bonds in our portfolio with the 70/30 starting point. Notably, this portfolio has a significantly higher Sharpe ratio (0.84) and expected return (10.7%) than the base 70/30 stock/bond portfolio, showing that even small allocations to digital assets can significantly improve portfolio risk/reward.

There’s More to Digital Assets Than Bitcoin

Up until now, we have used Bitcoin exclusively in these optimization exercises. This has mainly been to enable comparison between our methods and similar studies. Bitcoin is, of course, the oldest, largest, most liquid, and generally best known of the digital assets, so it is a natural fit for many types of analysis. However, it represents a very narrow thesis relative to the broad potential of digital assets as outlined in Part 2 of the piece. Our thesis on digital assets necessitates a much broader exposure.

Exhibit 14: Adding broad digital assets exposure to a 70/30 stock/bond portfolio

Source: Outerlands Capital Research, CoinGecko, Yahoo Finance, Bloomberg, S&P Global

While we argue that achieving this broader exposure is still the domain of active management, we can still objectively study it by using an index of digital assets such as S&P’s Cryptocurrency Broad Digital Markets index, which is “designed to track the performance of digital assets listed on recognized open digital exchanges that meet minimum liquidity and market capitalization criteria ...”²⁸. Index data is available back to early 2017, and we again use this period for estimating digital assets’ expected return, standard deviation, and covariance with stocks and bonds, while using data back to 2000 to judge the inputs related to stocks and bonds.

We repeat the prior analysis (starting with a 70/30 stock/bond portfolio and adding digital assets), and find an optimal Sharpe ratio after adding a 23% allocation to digital assets, at the equal expense of stocks and bonds.

However, again, risk contribution of digital assets is nearly 80%. Scaling this back, the portfolio sees an optimal Sharpe ratio of 0.78 at a 3.2% allocation to digital assets, while keeping the risk contribution of digital assets to 15%.

Notably, the results skew slightly less in favor of digital assets. Given the slightly lower Sharpe ratio of the S&P Broad Digital Markets index vs Bitcoin over the length of data available, this is not surprising. Furthermore, if the goal of any such exercise is to size an allocation to digital assets in general, not just Bitcoin, it does not make sense to use only Bitcoin as a proxy. It would similarly not make sense to use only Apple stock in deciding an allocation between stocks and bonds. For the purposes of analysis in the next steps, we will continue to employ the S&P BDM as a proxy for a digital asset allocation rather than Bitcoin alone, given its better qualitative alignment to our thesis of a rich, diverse digital asset ecosystem in the future.

Next we consider portfolios with more than just three assets.

Multi-Asset Portfolio

Exhibit 15: Indicative menu of investment options for sophisticated investors

Source: Outerlands Capital Research, CoinGecko, S&P Global, Yahoo Finance, MSCI, Bloomberg, Prequin, Credit Suisse

The modern investor has access to many investment options. As mentioned at the beginning of the piece, even retail investors now have access to a wide array of global markets via low-cost, liquid ETFs. The institutional investor naturally has even more choice, including a variety of hedge fund strategies and private assets. In Exhibit 15 we summarize a hypothetical menu of investments that a sophisticated investor may have in their portfolio, 14 in total. We expand options for public equities and fixed income to encompass the other DM economies, as well as EM economies, and add US corporate high yield bonds to the mix. Within the realm of alternatives, we add gold, macro hedge funds, and a mix of popular private/illiquid assets (equity, debt, real estate, and venture). Note: To include various alternatives, we will use quarterly return data.

With such a line-up of choices, it is worth spending some time considering hypothetical goals and constraints for such a portfolio. Examining the Sharpe ratios of the underlying assets, IG bonds and illiquid alternatives stand out, meaning the model will likely skew to those assets. In the case of IG bonds, a higher allocation will mean strong Sharpe ratios, but lower returns, meaning we may want to consider a minimum return hurdle for the portfolio. With regard to alternatives, their relative illiquidity could be a concern and a cap on overall allocation to those assets may be prudent. Meanwhile, US equities have had a lower Sharpe but are a stalwart allocation in most portfolios, meaning we may want to set a minimum allocation to capture exposure to this market. Finally, as we discussed before, an investor may want to cap the risk contribution of digital assets, given the relatively less well understood nature of their risks.

Exhibit 16: Optimized portfolios with S&P Broad Digital Markets Index

Combined, we will target a portfolio with:

No more than 20% allocation to alternatives
A minimum 7% annual expected return
A minimum 20% allocation to US equities
A maximum risk contribution of 15% from digital assets

Running a mean-variance optimization on the wider menu of assets, without further constraints, produces a portfolio with an attractive Sharpe ratio of 2.14 - notably higher than any results in our optimization with just digital assets, stocks, and bonds. However, as expected, composition-wise, the allocation has a considerable skew to alternatives (38%, mostly illiquid), ballasted by 62% in IG bonds (US and Global), with a portfolio expected return of 6%. With added constraints, the portfolio takes on the obligatory 20% allocation to US equities, ups the digital assets allocation to 1%, adds a bit of EM, and shuffles the alternatives. Overall, the portfolio Sharpe falls significantly, to 1.36, although expected return is higher at 7%.

A Short Foray Into Risk Parity

A drawback of the mean-variance optimization across multiple asset classes is that some may be altogether ignored (given 0% allocation by the model), given subtleties around relative expected returns, volatility, and covariances. In the exercise above, the constrained portfolio gives no allocation to five asset classes, while the unconstrained optimization skips six. This may fall short of investors’ expectations around diversification. One approach to address this is to structure portfolios around risk weighting entirely. That is, taking the risk contribution concerns of digital assets in the prior exercises, and applying the logic to all assets in the portfolio. This is known as a risk-parity portfolio, and the approach promises “true diversification”^29,30, by way of allocation across many assets with weights decided by the riskiness of each of those asset classes. The strategy gained interest in the aftermath of the Great Financial Crisis and has seen widespread application, although the strategy has come under scrutiny in recent years due to its propensity to allocate significantly towards bonds (which have performed poorly in light of the aggressive rate hiking cycle).

In the case of the investment menu presented above, with a number of illiquid alternatives, the application of risk parity may not be practical given such portfolios are often rebalanced monthly. However, it is an interesting thought exercise, and we present the results in Exhibit 17.

Note that allocations in a risk parity model rely only on historical risk and covariance, not on returns. As such, digital assets do not realize the benefit of a higher expected return, and the high level of volatility of the assets makes for a small allocation amidst the more established asset classes - especially when including private assets with very low reported volatility. In a slightly more realistic scenario, using only public, liquid asset classes, digital assets receive a 1.2% weight in the risk parity model (Exhibit 18).

Rather than torturing the data further, we move on to a few takeaways from the optimization exercises.

Optimization Observations and Takeaways

We have only scratched the surface of what is possible in the realm of portfolio optimization, and do not suggest that our analysis in this appendix covers all (or even most) of the challenges one may face in optimizing a portfolio. Furthermore, there are a multitude of more sophisticated analyses and processes by which one can better understand the limitations of data and assumptions going into a model, and more still with which one can add rigor to the model. Our analysis is not a criticism of mean-variance optimization, but rather an exploration of the many moving parts that can inform allocations. That being said, we conclude with a few high-level takeaways that are relevant to any portfolio-building exercise:

The process of portfolio optimization is intricately linked to the estimates you start with. Estimating forward-looking returns is (very) difficult. Historical data can be used as a proxy, but is, by nature, not stable over time, and comes with the assumption that the future will look like the past. Large banks and asset managers spend time and employ a good deal of sophistication formulating in-house expectations for returns over the medium- and long-term (see Footnote 30), but these carry their own assumptions, all of which can in turn be debated.
Theoretically optimal portfolios may not be practical for many investors (“you can’t eat a Sharpe ratio”), particularly given constraints around the use and availability of leverage. As such, models must be calibrated with individual investor goals and constraints in mind. There is no one-size-fits-all in investing, and any guidance that suggests otherwise should be challenged.
Mean-variance optimization naturally favors assets with high Sharpe ratios and low correlations. Many illiquid alternatives exhibit high Sharpe ratios given low reported volatility, but these results may be unnaturally smoothed given the low frequency with which assets are marked-to-market (see Footnote 20).
Risk contributions should be considered as a part of the allocation process, to better understand how assets impact risk measures for a portfolio.

The portfolio optimization exercises above often converge to a 1-2% allocation to digital assets, after limiting risk contribution, which seems reasonable. However, our basic contention remains that an investment into digital assets should be driven by a long-term, fundamental outlook on the disruptive nature of blockchain and digital asset technology.

Footnotes and Sources

Ampersand Portfolio Solutions. 2018. “The Risk Contribution of Stocks.” The Hedge Fund Journal, no. 131 (April). https://thehedgefundjournal.com/the-risk-contribution-of-stocks-2/.
See work by DLT Labs for more history
HM Treasury, “Government Sets out Plan to Make UK a Global Cryptoasset Technology Hub,” GOV.UK, April 4, 2022, https://www.gov.uk/government/news/government-sets-out-plan-to-make-uk-a-global-cryptoasset-technology-hub.
“Markets in Crypto-Assets Regulation (MICA),” 2023, https://www.esma.europa.eu/esmas-activities/digital-finance-and-innovation/markets-crypto-assets-regulation-mica.
“MAS Finalises Stablecoin Regulatory Framework,” August 15, 2023, https://www.mas.gov.sg/news/media-releases/2023/mas-finalises-stablecoin-regulatory-framework.
Georgina Lee, “Hong Kong Regulator to Issue Crypto Licences with Retail Investor Guardrails,” Reuters, May 24, 2023, https://www.reuters.com/technology/hong-kong-regulator-issue-crypto-licences-with-retail-investor-guardrails-2023-05-24/.
Nagase, Takeshi, Takato Fukui, and Keisuke Hatano. “Blockchain & Cryptocurrency Laws and Regulations 2024 | Japan.” GLI - Global Legal Insights - International Legal Business Solutions, October 30, 2023. https://www.globallegalinsights.com/practice-areas/blockchain-laws-and-regulations/japan.
“Virtual Asset Activities,” Abu Dhabi Global Market, accessed November 2023, https://www.adgm.com/setting-up/virtual-asset-activities/overview.
“VARA Rulebook,” Virtual Assets Regulatory Authority (VARA), 2023, https://rulebooks.vara.ae/.
Internet Society, “A Brief History of the Internet - Internet Society,” October 11, 2023, https://www.internetsociety.org/internet/history-internet/brief-history-internet/.
Ritchie, Hannah et al. “Internet.” Our World in Data, April 13, 2023. https://ourworldindata.org/internet#internet-access.
Thurston, David, Maureen Austin, and Cambridge Associates. “Venture Capital Positively Disrupts Intergenerational Investing.” Cambridge Associates, January 2020. https://www.cambridgeassociates.com/en-eu/insight/venture-capital-positively-disrupts-intergenerational-investing/.
“Family Offices Investing in Venture Capital 2022 - Venture Investing in Market Volatility | Campden Wealth,” July 2022. https://www.campdenwealth.com/report/family-offices-investing-venture-capital-2022-venture-investing-market-volatility.
Bradbury, Rosie. “Calpers Ups VC Allocation after ‘Lost Decade.’” PitchBook, June 12, 2023. https://pitchbook.com/news/articles/calpers-venture-asset-class-tiger-lightspeed.
Barry, David G. “Florida SBA Finds Opportunity to Commit to PE Funds.” Markets Group, November 9, 2022. https://www.marketsgroup.org/news/Florida-SBA-PE-Cryptocurrency.
YaleNews. “Investment Return of 6.8% Brings Yale Endowment Value to $31.2 Billion,” September 24, 2020. https://news.yale.edu/2020/09/24/investment-return-68-brings-yale-endowment-value-312-billion.
McIntyre, Catherine. “Pension Funds Eye Venture Capital Buying Opportunities amid Market Downturn.” The Logic, September 27, 2023. https://thelogic.co/news/pension-funds-eye-venture-capital-buying-opportunities-amid-market-downturn/.
The Block. “Daily Exchange Volume.” Data set. The Block, n.d. https://www.theblock.co/data/crypto-markets/spot/total-exchange-volume-daily.
“Venture into the Secondaries,” April 17, 2023. https://www.abrdn.com/en-gb/intermediary/insights-and-research/venture-into-the-secondaries.
Asness, Cliff. “Why Does Private Equity Get to Play Make-Believe with Prices?” Institutional Investor, January 6, 2023. https://www.institutionalinvestor.com/article/2bstqfcskz9o72ospzlds/opinion/why-does-private-equity-get-to-play-make-believe-with-prices.
Equable Institute. “State of Pensions 2023.” Equable Institute, July 21, 2023. https://equable.org/state-of-pensions-2023/.
Saleem, Rehan. “A Sign in the Secondaries - Private Equity News.” Private Equity News (blog), July 7, 2023. https://www.privateequitywire.co.uk/sign-secondaries/.
Private Capital Advisory. “Global Secondary Market Review.” Jefferies, January 2023. https://www.jefferies.com/CMSFiles/Jefferies.com/files/IBBlast/Jefferies-Global_Secondary_Market_Review-January_2023.pdf.
See S&P U.S. Indices Methodology (Appendix A).
A mean-variance optimization with Bitcoin, stocks, and bond provides allocations of 5/5/90, respectively - again an unintuitive starting point.
For further insight into the interpretation of risk contribution, see: Qian, Edward E., On the Financial Interpretation of Risk Contribution: Risk Budgets Do Add Up (February 2005). Available at SSRN: https://ssrn.com/abstract=684221.
For more information, see overview or index factsheet (PDF) on S&P’s website
For more details on risk parity portfolios see works by PanAgora Asset Management and AQR (PDFs). See also Footnote 29.
Maillard, Sébastien and Roncalli, Thierry and Teiletche, Jerome, On the Properties of Equally-Weighted Risk Contributions Portfolios (September 22, 2008). https://doi.org/10.3905/jpm.2010.36.4.060, Available at SSRN: https://ssrn.com/abstract=1271972.
For example, see J.P. Morgan Asset Management Long-Term Capital Market Assumptions (LTCMA), or AQR’s Capital Market Assumptions for Major Asset Classes

Disclaimers

The discussion contained herein is for informational purposes only, you should not construe any such information or other material as legal, tax, investment, financial, or other advice. Nothing contained herein constitutes a solicitation, recommendation, or endorsement to buy or sell any token. Nothing herein constitutes professional and/or financial advice, nor a comprehensive or complete statement of the matters discussed or the law relating thereto. You alone assume the sole responsibility of evaluating the merits and risks associated with the use of any information or content herein before making any decisions based on such information or other content.

Any performance referenced is for informational purposes only and does not represent an actual investment of Outerlands Capital nor does it include any fees related to investing including transaction fees or management fees, among other fees.

As of the date of this publication, Outerlands Capital Management LLC, a Delaware limited liability company, or its affiliates (collectively, “Outerlands Capital”) may hold or advise on long, short, or neutral positions in or related to the companies or digital assets described herein. The information in this report and any materials presented herein (the “Report”) was prepared by Outerlands Capital, is believed by Outerlands Capital to be reliable, and has been obtained from public sources believed to be reliable. Outerlands Capital makes no representation as to the accuracy or completeness of such information. Opinions, estimates and projections in this publication constitute the current judgment of Outerlands Capital and are subject to change without notice. Any projections, forecasts and estimates contained in this publication are necessarily speculative in nature and are based upon certain assumptions. It can be expected that some or all of such assumptions will not materialize or will vary significantly from actual results. Accordingly, any projections are only estimates and actual results will differ and may vary substantially from the projections or estimates shown. This Report is not intended as a recommendation to purchase or sell any commodity or security. Outerlands Capital has no obligation to update, modify or amend this publication or to otherwise notify a reader hereof in the event that any matter stated herein, or any opinion, project on, forecast or estimate set forth herein, changes or subsequently becomes inaccurate. Outerlands Capital may transact in any digital asset or the securities of any company described herein.

This Report is not an offer to sell securities of any investment fund managed by Outerlands Capital or a solicitation of offers to buy any such securities. An investment in any securities or digital assets, including the securities or digital assets described herein, involves a high degree of risk. There is no guarantee that the investment objective will be achieved. Past performance of these strategies is not necessarily indicative of future results. There is the possibility of loss and all investment involves risk including the loss of principal. 
Any forward-looking statements made are based on certain assumptions and analyses based on historical trends, current conditions, and expected future developments, as well as other factors that are believed to be appropriate under the circumstances. Such statements are not guarantees of future performance and are subject to certain risks, uncertainties, and assumptions that are difficult to predict.