Why The Total Portfolio Approach May Be Unsuitable for Mid-Market Plans

By: Timothy Atwill, Tacoma Employees’ Retirement System, and Jason Malinowski, Seattle City Employees’ Retirement System

This article argues that the Total Portfolio Approach (TPA) is ill-suited for mid-market public pension plans because it weakens governance, is unlikely to improve performance and introduces additional costs.

Recently, there has been a flurry of articles and thought pieces endorsing the Total Portfolio Approach (TPA) for managing a pension’s asset portfolio. TPA does away with a strategic asset allocation (SAA), in favor of setting a reference benchmark (say, 80% equities / 20% bonds). The CIO and investment staff then build a portfolio which has a similar risk profile as the reference benchmark but eliminates formal asset class limits which exist in the SAA approach. Such a system has been pursued by several mega-funds, including Australia’s Future Fund, New Zealand’s Super Fund, and, most recently, CalPERS.

As CIOs of mid-market public pension plans, we see several challenges with TPA that are papered over or simply not addressed in most discussions. These issues are especially true for smaller pools of assets such as our plans, which make up the majority of public pensions in the US. It is certainly not evident to us that TPA is an improvement upon the time-tested SAA approach, especially for mid-market plans.

Our concerns fall into three main categories:

1: Governance is Weakened

Under the SAA model, the investment staff is given a set of asset classes and target weights. Accompanying these target weights is a set of rebalancing ranges within which the staff are expected to operate. On a periodic basis, the board validates that portfolio allocations are within the given ranges. This makes monitoring compliance relatively straightforward for the boards of mid-market plans, which are typically composed of stakeholders with limited investment experience.

Compare this to the TPA approach, where a board must be familiar with a set of relatively obscure risk factors and then interpret if the relative exposures in a plan’s portfolio are compliant. This is leaving aside for a moment the numerous issues estimating risk factors for private market assets or even setting an appropriate risk tolerance.

TPA is intended to give wide discretion to the CIO and investment staff. But without active monitoring, success is premised on the staff both possessing talent and good intentions. This may not always be the case. Given the potential information asymmetry between the investment staff and the board, it’s not hard to imagine scenarios where staff loads up on risks not modeled well by risk factors in pursuit of higher returns.

2: Return Improvement is Unlikely

The proponents of TPA claim that it can outperform an SAA approach by 0.5% to 1% per annum and that early adopters have outperformed by up to 1.8% annualized1. This outperformance is attributed to allowing staff to invest dynamically based on their forward-looking expected returns without being constrained by asset class targets and ranges.

But this contradicts a long-track record of disappointing results from such unconstrained methods of investing. For example, tactical asset allocation (TAA) funds have consistently struggled since their invention in the 1990’s despite employing a similarly dynamic strategy. In fact, Morningstar recently found that the average TAA fund underperformed a comparable allocation fund with a static SAA by 1.4% per annum from 2005 to 2024². Other recent academic studies come to much the same conclusion.³

Given that TPA has been employed by a small and diverse set of mega-funds during a decade-long market rally, its claims of return improvement are difficult to verify and may not persist in more challenging markets.

3: Additional Costs Not Considered

TPA prides itself on the ability to “go anywhere”, typically quoting a list of niche and exotic strategies which do not fit cleanly into the asset classes of SAA.

The early adopters of TPA have been mega-funds, who already have in place large amounts of staff and systems to research, monitor, and risk-manage this broader opportunity set. However, mid-market plans are much more resource constrained, with headcounts ranging from no dedicated staff to perhaps a dozen, and limited IT resources and risk systems.

TPA appears to be a solution to leverage existing resources at these mega-funds, while a TPA implementation for most plans would represent a substantial level of additional fixed costs. It is not clear the potential incremental returns would justify such an investment.

Given these troubling aspects of TPA, we believe a high level of caution is warranted for any plan looking to transition from SAA to TPA. For mid-market plans, we view the proof presented so far as inadequate to motivate a re-tooling of existing processes for the possibility of minimal gains and the introduction of unforeseen risks. While consultants and mega-plans will continue to promote TPA, most plans are better off sticking with SAA.

About the authors: Timothy Atwill, CFA, PhD, FCAS, is the Chief Investment Officer for the Tacoma Employees’ Retirement System (TERS). Prior to joining TERS in 2021, Timothy was an investment professional at Parametric and Russell Investments.

Jason Malinowski, CFA, is the Chief Investment Officer for the Seattle City Employees’ Retirement System (SCERS). Prior to joining SCERS in 2014, Jason was an investment professional at BlackRock.

Endnotes:
^1. Thinking Ahead Institute
^2. Why Tactical-Allocation Funds Failed—Again | Morningstar
^3. Static Indexing Beats Tactical Asset Allocation; Journal of Index Investing