The total portfolio approach (TPA) is a way of building portfolios in which every investment competes for capital based on its contribution to the whole fund's objective, measured against a single reference portfolio, rather than being slotted into fixed asset-class buckets with their own benchmarks. Funds like CPP Investments, GIC, NZ Super and Australia's Future Fund use it.
For half a century, institutional investing has been organized around a simple idea: decide how much to put in each asset class, set a benchmark for each, and rebalance back to those weights. That framework — strategic asset allocation — built the modern pension and endowment world. But a growing group of the most sophisticated funds has concluded it has structural flaws, and has moved to something different: the total portfolio approach. This guide explains what TPA is, how it differs from the model it is replacing, and why funds managing trillions are adopting it.
The problem TPA is trying to solve
Strategic asset allocation (SAA) divides a fund into asset-class buckets — equities, bonds, real estate, private equity, infrastructure — each with a target weight and its own benchmark. Teams are then judged on whether they beat their bucket's benchmark.
The approach has two well-known weaknesses. First, it encourages silo behaviour: each asset-class team optimizes its own slice and defends its allocation, even when capital would do more good elsewhere in the fund. Second, asset-class labels are a crude way to measure risk — a "real estate" allocation and an "equity" allocation can share much of the same underlying economic exposure, so a portfolio that looks diversified by asset class may be concentrated by risk factor. SAA can therefore misstate how much risk the fund is really taking and where its returns actually come from.
What the total portfolio approach does differently
Under TPA, the fund is managed as one integrated portfolio rather than a collection of buckets. Three principles define it.
Every investment competes for capital. There are no fixed asset-class allocations to defend. A prospective private-equity deal, a listed-equity position and an infrastructure asset all compete for the same capital based on what each contributes to the total fund's objective — its return, its risk, and how it interacts with everything else the fund already owns.
Diversification is measured by risk factor, not asset class. Instead of slotting investments into labelled buckets, TPA looks through to the underlying drivers of return — economic growth, interest rates, inflation, credit, liquidity — and diversifies across those factors. This gives a truer picture of the portfolio's real exposures.
Success is measured against a reference portfolio. The fund defines a simple, low-cost, investable reference portfolio — often a basic blend of global equities and bonds — that captures the level of risk it is willing to take. The actual portfolio must beat that reference portfolio over time. Every active decision and every illiquid, expensive private-market position has to justify itself against that simple alternative, which imposes real discipline on where the fund spends its risk and fee budget.
TPA versus strategic asset allocation, side by side
The contrast is sharp. SAA is set-and-forget and segmented: fixed weights, asset-class benchmarks, rebalancing back to target. TPA is dynamic and holistic: no fixed weights, a single fund-level objective, capital flowing continuously to its best use. SAA measures success as relative performance within each bucket; TPA measures it as total fund return versus the reference portfolio. SAA diversifies by asset class; TPA diversifies by factor.
The practical effect is to move power and accountability to the centre. Under TPA, the CIO and a unified investment team own the whole-of-fund outcome, free to build an unconventional mix if it better achieves the goal — without worrying about "tracking error" to a dozen sub-benchmarks.
Which funds use it
TPA emerged among large, long-horizon funds with the governance and internal capability to run it. New Zealand Super Fund is widely credited as an early pioneer. CPP Investments in Canada, Australia's Future Fund, and Singapore's GIC are among the best-known adopters; GIC, for instance, frames its portfolio as a policy portfolio capturing systematic (beta) risk plus an active overlay seeking alpha. These are precisely the kinds of institutions — large, patient, internally resourced — for which the approach was designed.
Industry research has been encouraging: analyses cited by bodies such as the Thinking Ahead Institute and CAIA suggest TPA-adopting funds have delivered a meaningful performance edge over SAA peers over multi-year periods, though such comparisons should be read with care given differences in fund size, mandate and skill.
How a TPA decision gets made in practice
The mechanism that makes TPA work is opportunity cost. When a new opportunity is assessed, the question is not "does this beat its asset-class benchmark?" but "does adding this improve the total fund relative to the reference portfolio, after accounting for the risk it adds and the capital it consumes?" Every candidate investment is, in effect, funded by selling down a slice of the reference portfolio, so each one must earn its place against that simple, cheap alternative.
In practice this means a private-equity commitment competes directly with listed equities, infrastructure and credit for the same risk budget. If the private deal does not offer enough extra return to compensate for its illiquidity, fees and added risk, it is not funded — regardless of whether the "private equity bucket" is below target, because there is no bucket. This discipline is the source of TPA's appeal: it pushes capital relentlessly toward its most productive use and strips out the institutional inertia that lets underperforming allocations survive simply because they have a target weight to defend.
How does TPA differ from a policy portfolio?
The clearest contrast is with the traditional policy portfolio. A policy portfolio sets fixed long-term target weights for each asset class — say 30% private equity, 20% real estate, 15% infrastructure — and the fund's job becomes maintaining those weights and rebalancing back to them. Decision-making is organised around filling and defending buckets. The total portfolio approach removes the buckets entirely: capital is fungible across the whole fund and flows to the best risk-adjusted opportunity, judged against the total-fund objective rather than an asset-class benchmark. Many leading funds, including GIC, blend the two — keeping a policy or reference portfolio to express overall risk appetite while running the actual portfolio on total-portfolio principles. We compare the two models in detail in [policy portfolio vs total portfolio approach](https://www.universalassetowners.com/policy-portfolio-vs-