Benchmarking Multi-Asset Portfolios: The Global Capital Stock
This article was coauthored by Gregory Gadzinski, PhD, of the INSEEC Research Center at the International University of Monaco, and Andrea Vacchino of Panthera Solutions.
Multi-asset investing is nothing new.
William Sharpe first postulated that the market portfolio was the natural starting point in portfolio construction in 1964. The balanced portfolio, which held fixed proportions of bonds and equities, was another antecedent to today’s multi-asset portfolio.
With increased global access to various asset classes, multi-asset investing has entered the mainstream for certain second generation asset allocation methods — for multi-factor/multi-period models.
Constructing optimally diversified multi-asset portfolios is all the more critical for investors who follow the principles of such third generation asset allocation methods as adaptive asset allocation or adaptive risk management techniques.
Multi-asset investing’s growing popularity reflects these preferences: More than 14,000 multi-asset funds are offered in Europe alone.
Yet after all these years, the core question remains: How do you know whether a multi-asset portfolio is worth the investment?
Observation vs. Opinion
Many institutions have policy guidelines based on their investment managers’ preferences and expectations about the risks and returns associated with each asset class.
Other institutional investors run peer-group comparisons with similar multi-asset managers or measure their portfolios against broadly defined total return indices.
But these are only work-arounds. No policy portfolio benchmark exists against which investors can measure their multi-asset investment strategy.
A hypothetical global index purist could simply buy all of the outstanding assets in the world. The resulting global market portfolio would include all risky assets in proportion to their market capitalization. The purist could then use this as a benchmark to compare strategies.
While this approach has an appeal, not all risky assets are investable or measurable. With publicly available financial databases, we can now compute a global investable market portfolio. Researchers at McKinsey Global Institute report built a “map” of global financial assets. However, their data incorporated only traditional financial assets, omitting alternative assets.
So other researchers went a step further and added real estate and private equity to their broader asset universe. But they only included investable assets, with their portfolio weights based on their assets under management (AUM).
One Step Beyond
Our calculations are the next logical stage in this evolution. Our springboard is the definition of “capital.”
Capital is “non-financial assets having a dual role in an economy, being both a source of capital services in production and a storage of wealth,” according to the United Nations System of National Accounts (UNSNA).
Thomas Piketty used a broader definition: He extended the concept of “storage of wealth” to financial assets and assumed “capital” to mean the stock of all assets held by private individuals, corporations, and governments that can be traded in the market regardless of whether these assets are being used or not.
By merging these two definitions, we developed a measure of capital stock that consists of quantifiable financial and non-financial assets.
As a selection rule, we include assets whose price information is public or can be calculated from global institutional databases. We exclude assets whose valuations depend on partial, non-transparent sources, that are not observable, or are based on non-replicable methodologies.
More precisely, we calculated the global market value of the following 11 asset classes from 2005 to 2015 as our proxy for the global market portfolio:
- Public equities
- Private businesses
- Government securities
- Financial institution bonds
- Non-financial corporate bonds
- Cash equivalents
- Cash
- Securitized loans
- Non-securitized loans
- Real estate
- Land
Global Capital Stock 2005–2015
The following charts depict global capital stock per asset class during the decade in question in trillions of US dollars and as a percentage.
Global Capital Stock per Asset Class: In Trillions of US Dollars
Global Capital Stock per Asset Class by Percentage
The world’s global capital stock reached $512 trillion in 2015, down from $517 trillion in 2014. The largest component? Debt, at $194 trillion, which grew without interruption until 2014. Non-financial corporate bonds showed the largest increase in 10 years.
Private businesses were worth $100 trillion in 2015, while public debt stood at $95 trillion after five years of relative stability due to the appreciating US dollar rather than a decrease in public borrowing. Finally, equity holdings rose to an all-time high in 2015, maintaining a stable weight over the preceding three years.
Putting This Data to Use
So how can our findings be applied? On the buy-side, they can be used to:
- Monitor multi-asset managers by benchmarking their holdings against the global capital stock components, and then discovering appreciated or unrecognized drifts.
- Achieve a comprehensive understanding of the performance attribution of multi-asset strategies.
- Construct portfolios with the global capital stock as the initial point of reference.
On the sell-side, they can be applied to:
- Investment decision making (factor allocation)
- Risk management (style drift, concentration risk)
- Marketing and branding
Of course, our research does have limitations. Given the lack of data for non-financial assets, there is an intrinsic margin of error in our calculations of the global capital stock. But we are optimistic that further research will help close the data gap and reduce this margin of error.
Our research in “The Global Capital Stock: A Proxy for the Unobservable Global Market Portfolio” is an important step toward defining a reliable benchmark for multi-asset portfolios. We will continue our efforts to develop more accurate measures. In the meantime, we invite you to explore our methods in more detail.
If you liked this post, don’t forget to subscribe to the Enterprising Investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.
Image credit: ©Getty Images/erhui1979
The white paper “Satisfying the Prudent Man” by Scott Juds provides insight with a quantitative definition of portfolio risk. The paper also outlines reducing draw-down and developing a portfolio of single position strategies. I am a subscriber to SectorSurfer.
Good to read this – Its in line with what we believe at Redstone, especially around the Real Assets.
This is what we think about this area http://www.redstoneap.com/single-post/2016/10/19/redstone-believe-that-asset-managers-should-be-allocating-more-to-real-asset-investments
Why do you think Target Risk indexes are not more widely adopted as benchmarks? These have been developed by Dow Jones, S&P and Morningstar and were designed as portfolio benchmarks ranging from Conservative to Aggressive (each has 4 or 5 risk levels for their benchmarks). Why does the industry seem to ignore these?
Good point. I think it’s a lot to do with benchmarks being other funds. Many investors still thinking that S&P 500 index investing implies some degree of asset safety as it delivers the return of the index more reliably than many active managers. But now the tech is there and we’ve had it clearly demonstrated that traditional asset allocation methods were imperfect to a much greater degree than we already supposed it clear that risk-based allocation should be where we are at. We think about the market, not from a nominal basis but from a putative risk or return per unit of risk. Simplifies things a lot and leads us much more toward real assets than we would otherwise be.
I’d just add that of course established fee structures in asset management is the core reason. We are a private family office and aren’t constrained in that way.