Start here: TPA in a minute

Under TPA, every investment faces one question: does it help the whole fund meet its goals? Ideas from across the fund compete for the same money and the same risk, which is what competition for capital means.

Many funds make this concrete with a reference portfolio: a low-cost blend of listed equities and bonds that the board chooses to express its appetite for risk, plus an active risk limit on how far management may stray from it. Each new investment is funded, on paper, by selling a slice of that blend with the same market risk, and it must beat that slice after costs. The device is common but optional: Australia's Future Fund has no reference portfolio.

Sources: CFA Institute Research & Policy Center 2026, pp. 3–4, 20–21; Thinking Ahead Institute 2019, p. 2; A. Ang, M. Brandt, D. Denison for the Norwegian Ministry of Finance 2014, p. 9; CalPERS 2025, p. 2; GIC 2022, Policy and active portfolios section; CalPERS Board 2025, slides 7–8; Investment Magazine 2026, P4–P5; Future Fund Board of Guardians 2025, whole document

TPA is often misread. It is not:

• A new theory. A 2026 review for the CFA Institute Research Foundation finds no new portfolio theory in it; what changes is who decides, who owns the risk and who answers for the results.
• One fixed recipe. It is a spectrum: many funds sit part-way along it as hybrids, and going all the way is not always worth the cost.
• Constant trading. Funds act when conditions change, and sometimes the right move is to do nothing.
• A new label. Renaming old tools TPA, without changing who decides, how people are paid and how results are judged, changes little.

Sources: CFA Institute Research Foundation 2026, pp. 1, 3, 17, 40; CFA Institute Research & Policy Center 2026, pp. 3, 7, 19; CAIA Association & Thinking Ahead Institute 2025, pp. 31–32; Thinking Ahead Institute 2025, p. 8

SAA spread in the 1980s, largely through US consultants, because it gave boards a stable mix that they could sign off, track and explain.

From the mid-2000s, a handful of large funds, mainly in Canada, Australia, New Zealand and Singapore, began running their money as one portfolio. A 2014 review for Norway's government pension fund wrote the method up as the Opportunity Cost Model.

Its spread is not proof that it works: much of the case for adopting it now comes from the Thinking Ahead Institute, part of the consultancy WTW, which sells advice on making the switch.

Sources: Roger Urwin, Thinking Ahead Institute 2020, web page; CFA Institute Research Foundation 2026, p. 19; fn 9, p. 13; CAIA Association 2024, p. 13; CFA Institute Research & Policy Center 2026, pp. 9, 39; A. Ang, M. Brandt, D. Denison for the Norwegian Ministry of Finance 2014, p. 9; Thinking Ahead Institute & Future Fund 2025, front matter; Thinking Ahead Institute 2025, pp. 11–12

Under SAA, the board's central decision is the mix: a target weight and a benchmark for each asset class. Under TPA, it sets the destination and the guardrails, not the route: the fund's goals, how much risk it may take and the limits within which management invests.

At CalPERS, which switched in July 2026, the board now sets a reference portfolio split 75/25 (as of 17 Nov 2025 · source) between global equities and US Treasuries and an active risk limit of 400bp (4 percentage points) (as of 17 Nov 2025 · source), replacing its asset-class targets and ranges. Staff told the board this leaves management no more room for risk than the old ranges allowed.

Changing the mission or the reference portfolio stays with the board, and oversight has to grow as management's freedom grows.

Sources: CFA Institute Research Foundation 2026, Exh. 2, pp. 10–11; pp. 20–21, 24–25; CFA Institute Research & Policy Center 2026, p. 14; CalPERS newsroom 2025, paras 1–2; CalPERS 2025, p. 1; CalPERS 2025, pp. 3–4

SAA still suits many funds, and the CFA Institute's 2026 guide expects it to stay dominant for the foreseeable future. Among the world's largest asset owners, a preliminary count by the Thinking Ahead Institute found only 8 of 100 (as of 2025 · reported · source)¹ using TPA in full.

Full TPA demands strong governance, data and people, so it suits large or sophisticated funds best; smaller ones can take parts of it. SAA may be the better choice when:

• there is no clear problem for TPA to solve;
• the investment team's skill is in doubt;
• the horizon is short or the appetite for risk is low;
• the fund cannot act, will not act or is not allowed to act on whole-fund views;
• opportunities are stable and constraints are simple.

Sources: CFA Institute Research & Policy Center 2026, pp. 3, 9–10, 34; CFA Institute Research Foundation 2026, pp. 17, 27, 41

Thin, so far. No controlled study shows that TPA beats a traditional SAA, and the CFA Institute Research Foundation's 2026 review finds that the literature does not establish systematic outperformance.

The figure most often quoted, a gap of 1.3% a year (as of 2025 (study 2024) · source)¹ in favour of TPA funds, comes from one peer study in which funds sorted themselves into groups: a prompt for questions, not proof.

The review's view is conditional. TPA is a way of running a fund, not a source of extra return in itself; with strong governance, skills and data it may improve the odds, but it guarantees nothing.

Sources: CFA Institute Research Foundation 2026, pp. 3, 36–41; web page; Thinking Ahead Institute & Future Fund 2025, closing report, pp. 19–20; CFA Institute Research & Policy Center 2026, pp. 10, 28