This is the first in a three-part series on the Total Portfolio Approach (TPA). In this paper, we discuss some of the key components and objectives of TPA and explore using high-frequency private markets data to document risk factor exposures, dynamic correlations, co-tail dependency, and hurdle rates. Future papers will examine several case studies of TPA by large investors and make use of factor exposures calculated with privateMetrics and infraMetrics indices to design a multi-asset portfolio.

TPA with Private Assets

The Total Portfolio Approach aims to find a common risk language across the portfolio, including measuring portfolio level factor exposures, dynamic co-movements, tail-risk, and other risk measures. Implementing TPA can seem difficult with private assets since it requires high frequency, market based private markets pricing and risk metrics. Appraisal-based private asset proxies, with their lagged and stale valuations, materially understate marginal risk contributions, making it impossible to compare investment opportunities on an apples-to-apples basis. In effect, risk measurement errors are likely to lead to excessive allocations to private assets, defeating the point of TPA. Using listed proxies is even more clearly self-defeating since risk factor exposures between public assets and the private asset proxy would be the same by design.

The TPA Framework

Several institutions employ a “reference portfolio” as a governance mechanism to set a risk tolerance level, or as a quasi-benchmark (sometimes both). This reference portfolio is typically comprised of listed equities and bonds and used in part to validate a risk budget with the Board. With the TPA framework, the reference portfolio can both set the risk budget and opportunity cost for active risk: active risk comes in the form of deviations from the reference portfolio, including through the inclusion of private asset classes. Moving from the reference portfolio to a diversified portfolio including public and private assets is expected to improve risk adjusted returns (Sharpe ratio). Unlike alpha which is manager-specific, TPA targets improved diversification and the harvesting of risk premia available across imperfectly correlated risk exposures.

privateMetrics and infraMetrics

Using private market indices like SIPA’s is key to implementing TPA with private assets:

1. Higher frequency private assets data can be used to quantify both macro and style factor exposures, allowing for a linkage with listed securities. This can inform on risk at total portfolio level and may facilitate rebalancing using factor exposures.
2. Dynamic correlations between private and listed assets, detailing how the relationship evolves in different regimes and stress periods.
3. Dynamic risk premia (produced monthly) allowing for comparison of market priced risk premia across private and listed assets.
4. More realistic risk and return data that can be used to establish sensible hurdle rates for adding private equity, private infrastructure equity (or debt) to the portfolio.