Bots took the CFA test,
And aced it with AI finesse.
While humans still cram,
Bots breezed Level III’s exam,
In minutes, no coffee, no stress.
ChatGPT and Claude might be able to pass the CFA exams, but they’ll still need to meet the 3-year work requirement to become ChatGPT, CFA and Claude, CFA.
Modern endowment investing seemed to originate with Swensen’s “Yale Model,” but in the past couple decades there have been experiments in the broader allocator world (e.g., TPA portfolio management). Which of these portfolio allocation strategies do you think will hold up for the next 10 years, and what further innovations do you think the industry might take?
I’ve admittedly been sitting on this question for a while. For one thing, I have only an outsider’s perspective on the total portfolio approach (TPA), whereas I’m intimately familiar with endowment investing, which is generally associated with a strategic asset allocation approach (SAA). For another, TPA has generally not been widely adopted by U.S. institutional investors.
Now that latter point could be changing, as CalPERS (California Public Employees’ Retirement System), the largest U.S. public pension plan, is considering a move to TPA1. Before I go any further, let me provide some background. And also a warning: this is about to get pretty dry. While I think discussions on portfolio management philosophies are interesting, I’m deeply aware that not everyone would agree! For everyone who didn’t ask this question, feel free to stay or go, I won’t be offended.
Ok, some basic outlines:
Strategic asset allocation (SAA) is the portfolio construction approach used by most U.S. institutional investors. SAA involves setting long-term target allocations across asset classes (e.g., public equities, fixed income, private equity, real assets, etc.), typically revisited every few years.
Total portfolio approach (TPA) is a portfolio construction approach that focuses on a portfolio’s total exposure to risk factors (e.g. equity beta, interest rates, liquidity, etc). TPA is more widely used among sovereign wealth funds, including Singapore’s GIC, Australia’s Future Fund, and New Zealand’s Superannuation Fund.
TPA has been gaining support and traction because of its commitment to flexibility. In theory, each and every new investment is evaluated not in an asset class context, but instead in terms of how the investment impacts the overall portfolio. Its advantages are obvious when considering potentially compelling investment opportunities that don’t neatly fit into asset class categories – for instance debt and equity hybrid strategies, like capital solutions. These in-between strategies might not have an obvious home within a traditional SAA framework.
In theory, SAA is more inflexible, with explicit targets to pre-specified investments. In practice, the gap between TPA and SAA is not necessarily that wide. The criticism I’ve seen and heard regarding SAA tends to fall into a few categories:
Rigid asset class categories, excluding investments that don’t “fit neatly”
Siloed and/or territorial asset class decision making that is suboptimal from a big picture, total portfolio perspective
Investing to “fill buckets” and potentially missing out on the larger universe of available opportunities
I hear these criticisms, and my immediate reaction is – these are stupid problems to have. I’m not saying they aren’t real or don’t occur. They absolutely do. But an organization facing these problems doesn’t have an investment approach issue. It has a governance issue. And likely some team structure and incentive issues too.
An SAA framework doesn’t preclude investment teams from coordinating across the portfolio. Equity risk can still be tracked and managed across public and private holdings. Reporting might be organized by asset class, but risk exposures (sector, geography, factors, etc.) can still be monitored and evaluated at the portfolio level. And investment decisions can and should still be optimized for total portfolio outlook.
What SAA does provide over TPA is a clearer approach to benchmarking and relative performance measurement at both the investment and total portfolio level. I suppose reasonable minds can differ over how much that matters.
I suspect it’s not a coincidence that the early TPA adopters have been those with extraordinarily large pools of capital. The challenge of managing over 700 billion CAD2 (which translates to over $500 billion USD) for the Canadian Pension Plan is different from that of managing a $5 billion or even $50 billion endowment or foundation. It’s easier to coordinate across a smaller team, and similarly, it’s easier to invest nimbly across a smaller portfolio.
At a certain portfolio size, TPA probably does make the most sense. It focuses the team on a unified picture of portfolio exposures, risks, and opportunities. It’s a different kind of discipline in portfolio management. Again, I’ll caveat that this is my understanding of TPA as an external observer. I’m sure I’m missing some nuances.
So which portfolio strategy holds up over the next 10 years? Probably both? I don’t see E&Fs moving away from using the SAA approach, provided they’re not facing the very avoidable problems we discussed above. And the TPA approach seems to serve large sovereign and pension funds well. I don’t think we should be surprised that these fundamentally different investors might ultimately prefer different approaches.
And as for additional innovations, your guess is as good as mine. Are there better ways to measure risk? Are there better ways to improve decision making across the portfolio? Are there ways to do these things more quickly and cost effectively across the portfolio? I think so, and I think investment offices are experimenting around the edges. I don’t know how portfolio management will evolve, but I’m certain it will continue to advance, especially as (circling back to where we started) teams figure out how to incorporate AI into their processes.
Thanks for the question, and as always, feel free to reach out: askacio@ivyinvest.co!
See you in two weeks,
Wendy
If you enjoyed this post, you might also enjoy the Ivy Invest app!
Click below to create an account and learn more about our endowment-style fund:
As of 6/30/25