Global Head of Real Estate Applied Research, MSCI
MSCI’s Will Robson on private infrastructure’s role in a multi-asset-class portfolio
Will Robson, at MSCI, discusses some of the key points from a recent blog, Assessing Private Infrastructure in a Multi-Asset-Class Portfolio, such as misconceptions surrounding private infrastructure assets and how having the right data can help investors understand return characteristics and allocate capital.
About Will Robson
Will Robson, Global Head of Real Estate Solutions Research at MSCI, discussed the firm's recent launch of a real estate debt fund index in a February 2025 interview. He stated that the index helps to "institutionalize" real estate debt and predicted continued growth in debt flowing through alternative lending routes. Robson noted that lenders and equity investors providing "rescue capital" are using MSCI's loan-level data to identify distressed assets, particularly in markets such as U.S. CBD offices that have seen significant price corrections. In earlier appearances, Robson addressed the use of data in multi-asset-class portfolios. In a 2022 interview, he argued that investors often rely on proxy data for private infrastructure, which can lead to "sub-optimal allocations" by misrepresenting correlations and volatilities. In a 2021 presentation on climate risk, Robson distinguished between backward-looking carbon footprinting and forward-looking measures like climate value at risk, which he described as assessing the potential financial implications of future climate hazards on portfolios.
Transcript (6 segments)
W
Will Robson0:17
Hi, I'm Will Robson. I'm Global Head of Real Estate Solutions Research at MSCI.
The role that private infrastructure plays in the multi-asset portfolio depends on what an investor is invested in and why. So most commonly, it's about inflation hedging or high and stable income, low volatility and correlations for diversification. But it could be about seeking high returns in sectors that they're looking to see grow, or maybe it's about impact investing, investing in climate projects or social infrastructure.
Allocation to private infrastructure is commonly justified on the return characteristics, the correlation, the return levels and expectations. For any given level of return expectation, lower volatility and correlations to other asset classes typically increases an optimal allocation. However, investors often use proxy data for infrastructure to determine those characteristics. Often they use data from bond markets or listed infrastructure, or perhaps from valuation-based data from the private markets themselves. But these data sources and proxies can often under or overestimate the correlations and volatilities, and then you can end up with sub-optimal allocations, either higher or lower.
There's many misconceptions around private infrastructure assets. I think one of the most common is around the heterogeneity of the asset class. Typically, you hear people talk about the asset class as if it's this homogenous number of assets, but it's very diverse. So you have some investments that are very equity-like with high growth exposure and less diversification benefits, but then you have other assets that are much more bond-like and do provide those kind of more stable returns but with lower returns, more secure cash flows. Our private infrastructure data illustrates significant differences between, for example, contracted assets and non-contracted assets. Or for example, during 2020, within transport, airports performed very differently to transport. So there's a big mix, and so there's lots of opportunities to invest in what you want and get the characteristics that you want.
Private infrastructure investments can perform quite differently from listed infrastructure investments, and so you get a concept of the private risk premium or reporting bias around valuation data can have an influence on the allocation approach. We know from comparing our private infrastructure data with the listed infrastructure data that there are significant differences in performance. We also know that looking at our private valuation data, that this kind of data is smoothed and lagged to some extent. And when you use that kind of data in a multi-asset allocation model, it can dampen down the volatility and lower the correlations with other asset classes. So by combining the public and the private data together in a modeling framework that accounts for these nuances, you can get the best of both worlds and come up with an estimate that is more accurate to the real true characteristics of private infrastructure and get a more optimal allocation.
We've seen private infrastructure allocations grow substantially over the last few years, and many expect that to carry on into the future. But as infrastructure becomes a bigger part of the overall portfolio, having the right data and modeling approach to account for all these nuances we've talked about is going to be increasingly important. And as investors get more in their portfolios, they're going to understand the return characteristics and what it's actually doing to their portfolio in more detail. And not least, all the data, the thirst for data around climate and the climate transition is really accelerating the general thirst for transparency. And when people are asking for data on emissions and physical risk, they're going to start asking basic questions about return characteristics and how they can reflect that in their portfolio analysis.