Splitting the investment atom: Asset allocation through a risk factor lens

25 Jun 2024

Written by LGT Crestone Senior Asset Allocation Specialist Matthew Tan

Curiosity and a desire to better understand how our world works are fundamental to the human experience. This drive, in large part, inspired the innumerable scientific discoveries that have allowed us to better conceptualise our world in terms of elements, molecules, atoms, and sub-atomic particles, as we continue to search for the ultimate underlying forces that drive our universe. 

The task of constructing a diversified portfolio of assets to achieve long-term investment goals might appear worlds removed from the domain of particle physics, but the brightest minds in investment have nevertheless embarked on a similar multi-decade journey of attempting to isolate and better understand the underlying drivers that determine how asset classes behave and interact with each other.

In this Observations piece, we outline some of the main steps that have been taken so far along this journey of investment discovery. We introduce a state-of-the-art fundamental risk factor framework that underpins our asset class modelling and strategic asset allocation process. We also present several key benefits of applying this approach to construct robust and diversified multi-asset portfolios.

For decades, the investment world’s brightest minds have striven to discover a ‘theory of everything’ that explains how asset classes behave for the purpose of helping investors build better portfolios.

From Markowitz to Fama and French—a brief history of the investment world’s pioneers

The investment world’s journey to construct a framework to better understand asset class returns picked up in earnest in 1952 when Harry Markowitz introduced the concept of Modern Portfolio Theory (MPT). This theory applied the mathematics of mean-variance optimisation to allow investors to construct risk-efficient portfolios by taking into account the expected risk, return, and interlinkages between asset classes. Markowitz won a Nobel Prize in Economics for his work, and the MPT methodology is still a key foundation of investment strategy and portfolio construction to this day.

A vital component of the MPT framework revolves around developing robust and high-quality expectations for the risk, return, and interlinkages (or correlations) between various asset classes. Together, these forward-looking expectations are known as capital market assumptions, and they are the key driver of the quality (or lack thereof) of any MPT analysis. 

Given this importance, a major question that investors pondered was how to ensure their capital market assumptions were indeed robust and could be constructed in a disciplined, systematic manner, rather than being reliant on heuristics or on the whims or biases of an individual. While a major theoretical breakthrough had been made with MPT, practitioners lacked for some time the empirical tools and technology to truly maximise its potential.

In the 1960s, the Capital Asset Pricing Model (CAPM) was introduced as a tool to aid in this quest. Developed independently by Jack Treynor, William F Sharpe, John Lintner, and Jan Mossin, CAPM seeks to model an individual asset by determining how much of its behaviour is driven by the broader market. To do this, they used the power of mathematics, employing regression analysis based on historical experience to calculate the sensitivity, or beta, of an asset to the broad market.

CAPM proved to be an intuitive and powerful companion to MPT and allowed practitioners to differentiate various investments. However, it did have caveats—particularly, the focus on a single measure of sensitivity that viewed all asset classes as various shades of the one colour (the market).

Arbitrage Pricing Theory (APT) was introduced in the 1970s by economist Stephen Ross as an enhancement to CAPM. Rather than just focussing on the market as a single driving factor, APT allowed investors to consider and analyse asset classes against multiple underlying factors, such as the economy, inflation, and others, as illustrated in the chart below.

Advances in both theory and practice over the years have led to the development of a plethora of new and advanced asset class and portfolio construction models.

CAPM and APT allowed investors to increasingly decompose asset class returns into fundamental factors

Source: LGT Crestone. Size of components are illustrative only.

The twin tools of CAPM and APT, as well as increasingly powerful computers (and statistical methods) and an ever-expanding international investment universe, have supported significant further study into the composition and forecasting of asset class returns. These include the 1992 Fama-French three-factor model and 2002 work by Fung and Hsieh, which involved developing risk factors to decompose hedge fund returns.

Nevertheless, a set of underlying fundamental principles underpins these models and forms the basis of LGT Crestone’s proprietary risk factor approach.

Today’s state-of-the-art: Principal component analysis, data mining, and fundamental risk factors

Today’s landscape of factor and multi-asset investing is a varied one, encompassing developments from principal component analysis (PCA) to machine learning and data mining. However, while a plethora of modern risk factor models exists, the fundamental principles remain broadly similar and are based on the following primary beliefs:

  • Asset class returns can be explained by a set of underlying systematic factors.
  • The sensitivities of any individual asset class to these underlying factors can be determined through a combination of empirical analysis and qualitative review.
  • Once determined, these sensitivities can be used on a forward-looking basis to forecast asset class returns in an internally and intertemporally consistent manner. 

The basic thesis is analogous to the concepts of particle science. By deconstructing seemingly dissimilar asset classes into their constituent components (like a scientist might deconstruct molecules into atoms and atoms into protons and electrons), investors can tease out underlying similarities that were previously masked and better isolate true sources of diversifying risk. Rather than requiring a multi-billion dollar particle accelerator to conduct this analysis, reliable historical returns data, a solid understanding of the risk factor approach, and statistical modelling software are all the tools the astute investor needs to begin down this path.

To illustrate this approach in practice, one might consider the characteristics of a venture capital investment, or an infrastructure asset. Intuitively, one might expect the former to have some underlying linkage to broader equity markets (as it is effectively purchasing equity in a business), and we might expect the latter to have some linkage to broad economic growth and inflation (think of a toll road where through-traffic tends to be linked to economic activity and toll rates can be raised with inflation).

Various statistical studies, including LGT Crestone’s own proprietary risk factor approach, have indeed validated these hypotheses and also revealed additional underlying drivers. For venture capital, there is a linkage to small-cap equities (reflecting the relatively small size of the companies that are invested in). For infrastructure, there are additional linkages to equity markets (reflecting the equity-type nature of the investment) and interest rate risk (reflecting the stable, bond-like nature of infrastructure cashflows). 

As the chart below shows, there are still components of these asset classes that can’t be fully explained, and this can be attributable to idiosyncratic risk or manager skill. But on the whole, the analysis still provides us with a more informed understanding of the asset classes and how they might be expected to behave looking forward. Armed with our improved understanding, we might now expect a venture capital investment to be sensitive to broad equity market weakness, and we might expect an infrastructure asset to be vulnerable to a period of rising interest rates.

This fundamental risk factor approach decomposes asset class returns into intuitive, tractable macro-economic drivers…

Analysing asset classes through a risk factor lens helps us better understand the underlying drivers of return

Source: LGT Crestone. The sizes of individual risk factor components are illustrative only.

… which forms the cornerstone of our asset allocation and investment strategy process.

There are multiple applications of these findings. Some investors use them to construct liquid market proxies of private asset classes, such as hedge fund replication strategies, while others might use this framework to enrich discussions with their underlying managers (for example, querying why a venture capital investment is underperforming when the broader share market is rallying). 

A robust multi-asset risk factor framework allows investors to build more resilient portfolios

At LGT Crestone, we leverage this framework in the analysis, design and construction of robust, long-term multi-asset portfolios. Indeed, it forms the cornerstone of our asset allocation and investment strategy framework and underpins how we derive our proprietary capital market assumptions and construct investment portfolios for clients. Rather than viewing asset classes as indivisible and independent ‘atoms’, we instead leverage the power of our proprietary risk factor framework and view them through the lens of eight fundamental macro risk factors, alongside the idiosyncratic risk premia derived from manager skill and illiquidity. These factors are laid out in the table below:

LGT Crestone’s eight fundamental macro risk factors

Source: LGT Crestone.

While they may not encompass the totality of possible risk factors (there are hundreds, if not thousands, with more being discovered by financial academia every year), our firm belief is that this suite of risk factors provides a set of intuitive and forecastable risk factors. These risk factors can adequately explain the bulk of the risk and return characteristics across most asset classes, and provide us with a versatile set of tools (our own investment particle accelerator, to stretch the analogy) to analyse and consider new and emerging asset classes as they arise.

This more complete lens gives us much greater insight into the underlying drivers of asset class behaviour…

Benefits of our risk factor approach

The substantive benefits of this approach are borne out in the table below, which disaggregates a range of asset classes according to our assessments of their sensitivity to the various risk factors.

Sensitivity of asset classes to various risk factors

Source: LGT Crestone. Values are indicative and for illustrative purposes only.

… allowing us to build more nuanced, thoughtful, and resilient portfolios.

We gain a deeper understanding of the key drivers of each asset class

Many of these drivers should be intuitive to most investors (e.g., government bonds have a high sensitivity to interest rate risk), while others might be instructive (e.g., hedge funds, which invest across multiple asset classes, tend to have some exposure across the risk factors). Further, some drivers might be unexpected (e.g., high yield credit displays some sensitivity to small-cap equities, likely reflecting the smaller average company size of high yield issuers).

We can develop forward-looking asset class assumptions that are consistent and scalable

Having identified our eight fundamental macro risk factors, we can develop consistent and robust forward-looking assumptions for almost any asset class. These are based on our expectations for the forward-looking behaviour of the risk factors, as well as the sensitivity of that asset class is to each risk factor. This combination greatly enhances the breadth, depth, and quality of our capital market assumptions. 

We can assess the relative attractiveness of different asset classes based on risk exposures

An important advantage of the risk factor approach is that it sets a consistent and level playing field in considering where to deploy scarce capital. This allows us to make deliberate and nuanced investment decisions. For example, if we believe that interest rates have peaked and are likely to decrease, we can look to target investments in asset classes that have strong exposure to interest rate risk (government bonds, credit, property, or infrastructure). To the extent that some of these might be more attractive on a relative basis, we can express our interest rate view at a total portfolio level and access an undervalued asset in one fell swoop. 

The benefits of a risk factor approach to asset allocation are manifold… 

We gain more visibility and control over risk exposures at a total portfolio level

Having a robust multi-asset risk factor framework gives investors the tools to decompose their total portfolio of investments across the underlying risk factors. As an example, the framework allows us to derive a reasonable estimate of a total portfolio’s sensitivity to equity markets, or even its sensitivity to interest rates. The former is a key total portfolio risk management tool employed by some of the largest institutional investors, including the Future Fund. The latter lens might allow investors to better monitor and manage their portfolios’ exposure to the risk of rising interest rates. 

The power to decompose and understand the underlying drivers of risk and return becomes even more valuable when managing a sophisticated portfolio of alternative assets, which can encompass private markets, real assets, hedge funds, and more esoteric investments like royalty streams or even insurance-linked securities. The idiosyncratic nature and heterogeneous nature of alternative assets can be challenging to navigate, but a sound risk factor framework can arm investors with a strong tool to chart their path.

… and can help investors to  build, monitor, and manage robust portfolios for the long term.

We can build robust scenarios to stress test portfolios

A valuable aspect of the risk factor approach is that it allows us to stress test portfolios by enabling the formulation of robust, consistent scenarios across asset classes. For example, we can consider the impact of a potential economic scenario (such as a recession) on equities, interest rates, credit, and the other risk factors. We can also model the potential returns across traditional and alternative assets under such a scenario in an internally consistent manner. This provides a powerful enhancement to the risk management of a multi-asset portfolio.

In summary…

The charge of constructing a robust and diversified multi-asset portfolio has challenged investors for decades. This has sparked a journey of investment discovery that has greatly improved how we understand the underlying drivers that determine how asset classes behave and interact with each other. In consequence, this has given greater insight to help investors improve investment strategy and asset allocation.

In this Observations piece, we pay homage to some of the key pioneers of this unceasing quest, and introduce the fundamental risk factor framework that underpins our asset class modelling and strategic asset allocation process. It provides us with a powerful lens to ‘split’ the investment atom and understand the underlying drivers of asset class and portfolio behaviour, enabling us to construct resilient portfolios in a thoughtful and deliberate manner. In an increasingly uncertain investment environment, we think this is a compelling addition to any astute investor’s toolkit.  


This document has been prepared by LGT Crestone Wealth Management Limited (ABN 50 005 311 937, AFS Licence No. 231127) (LGT Crestone Wealth Management). The information contained in this document is of a general nature and is provided for information purposes only. It is not intended to constitute advice, nor to influence a person in making a decision in relation to any financial product. To the extent that advice is provided in this document, it is general advice only and has been prepared without taking into account your objectives, financial situation or needs (your Personal Circumstances). Before acting on any such general advice, we recommend that you obtain professional advice and consider the appropriateness of the advice having regard to your Personal Circumstances. If the advice relates to the acquisition, or possible acquisition of a financial product, you should obtain and consider a Product Disclosure Statement (PDS) or other disclosure document relating to the financial product before making any decision about whether to acquire it.

Although the information and opinions contained in this document are based on sources we believe to be reliable, to the extent permitted by law, LGT Crestone Wealth Management and its associated entities do not warrant, represent or guarantee, expressly or impliedly, that the information contained in this document is accurate, complete, reliable or current. The information is subject to change without notice and we are under no obligation to update it. Past performance is not a reliable indicator of future performance. If you intend to rely on the information, you should independently verify and assess the accuracy and completeness and obtain professional advice regarding its suitability for your Personal Circumstances.

LGT Crestone Wealth Management, its associated entities, and any of its or their officers, employees and agents (LGT Crestone Group) may receive commissions and distribution fees relating to any financial products referred to in this document. The LGT Crestone Group may also hold, or have held, interests in any such financial products and may at any time make purchases or sales in them as principal or agent. The LGT Crestone Group may have, or may have had in the past, a relationship with the issuers of financial products referred to in this document. To the extent possible, the LGT Crestone Group accepts no liability for any loss or damage relating to any use or reliance on the information in this document.

This document has been authorised for distribution in Australia only. It is intended for the use of LGT Crestone Wealth Management clients and may not be distributed or reproduced without consent. © LGT Crestone Wealth Management Limited 2024.

Subscribe to insights and observations

Please provide your first name.
Please provide a valid email address.
Please provide a phone number.
By subscribing to insights and observations you acknowledge you have read and agree to our privacy statement