GIC Says It Chose a Lower Return. Its Twenty-Year Number Fell to 3.4 Per Cent

SINGAPORE, 8 AUG 2026 — GIC's headline number fell this year, from 3.8 per cent to 3.4 per cent. Asked why in Parliament on 5 August, the Finance Ministry said the fund had chosen to take less risk.

That is a more interesting answer than it looks, because it describes a decision rather than a market.

The number and the question

GIC reports its performance as an annualised rolling twenty-year real rate of return — the compound annual return over the past two decades, after inflation. For the year ended 31 March 2026 that figure was 3.4 per cent, against 3.8 per cent the year before.

The Leader of the Opposition, Mr Pritam Singh, asked for the reasons behind the drop.

Senior Minister of State for Finance Mr Jeffrey Siow replied that the decline reflects, in part, GIC's more cautious investment approach in recent years amid heightened market volatility and uncertainty. GIC has prioritised portfolio resilience by diversifying its investments and adopting a lower-risk posture. That moderated returns, he said, while strengthening resilience and providing greater downside protection, consistent with GIC's mandate to preserve and enhance the international purchasing power of its assets over the long term.

Why a twenty-year figure moves at all

The measure is unusual and the mechanics explain more than the commentary does.

A rolling twenty-year return recalculates each year over a window that shifts forward by one year. A new year enters at one end and an old year drops off at the other. The figure can fall for two reasons. Either the new year added to the average was weak, or the old year that dropped off was strong.

The window ending 31 March 2026 covers the twenty years from 2006. The window ending a year earlier covered the twenty years from 2005. The calculation changed by dropping a year from the mid-2000s and adding a recent one.

The reply attributes the decline "in part" to the cautious posture. That phrasing leaves room for the arithmetic of the window itself, and the reply does not separate the two effects. Neither can we: doing so would require the annual series, which GIC does not publish.

What a lower-risk posture buys

The substance of the answer is a trade that most institutional investors describe and few state so plainly.

Diversifying and reducing risk lowers expected return. It also narrows the range of outcomes, which is what "downside protection" means. For a sovereign fund managing a small country's reserves, the trade-off is asymmetric: avoiding a bad decade matters more than capturing a good one.

GIC's mandate, as the reply restates it, is to preserve and enhance international purchasing power over the long term. Preserve comes first in that sentence, and the posture described is consistent with reading it that way.

The honest counterpoint is that a lower-risk posture is easy to justify after a period of volatility and harder to unwind. A fund that de-risks in uncertain markets and re-risks in calm ones is buying high and selling low on a very long cycle. Nothing in the reply addresses when or whether the posture reverses.

What the figure does and does not tell a reader

This number is quoted more often than it is understood, and two points of caution are in order.

It is a real return, meaning after inflation, so it is not comparable with the nominal figures most funds and index benchmarks publish. A nominal return over the same period would be several points higher, and comparisons that ignore this flatter the wrong side.

It is also the only headline figure GIC gives. The public sees no annual return, no benchmark comparison, and no breakdown by asset class. The twenty-year window is a deliberate choice, and its effect is to smooth precisely the variation that would let an outsider judge a single year's decisions.

That design is defensible for a fund with an explicit long-term mandate. It also means that when the number moves, the explanation has to come from a minister rather than from the data.

The mandate is doing the work in that sentence

The reply ends on GIC's mandate, and the wording repays attention.

It is to preserve and enhance the international purchasing power of assets under management over the long term. Every part of that mandate constrains the fund. "International purchasing power" sets the benchmark as what the reserves can buy abroad, not just their value in Singapore dollars. "Preserve and enhance" puts capital protection before growth. "Over the long term" is the reason for the twenty-year window.

A fund with that mandate is not trying to beat an index. A lower-risk posture that moderates return while narrowing outcomes is consistent with it. Judging the 3.4 per cent against equity market returns over the same period would be measuring the wrong thing.

What the mandate does not settle is how much return should be given up for how much resilience. That is a judgement, it was made, and the reply describes it without quantifying either side.

Why this number carries weight in Singapore

The figure attracts scrutiny for a reason that is structural rather than financial.

Investment returns on the reserves contribute to government revenue under Singapore's fiscal framework, which makes long-run performance a live matter for the budget rather than an abstraction. A question from the Leader of the Opposition about a move of that size is, in that light, a question about the public finances.

It is also one of very few observable numbers. When an institution publishes only one headline number, that number gets all the scrutiny. Some of the weight this figure carries is a consequence of how little else is published.

What "in part" is carrying

The two most consequential words in the reply are the hedge.

Attributing the decline "in part" to the cautious posture concedes that other factors contributed without naming them. The rolling-window arithmetic is the obvious candidate. Others include currency movements in an international portfolio and the inflation deflator itself — a real return falls when inflation rises, even if nominal performance holds.

The inflation effect is the least discussed of these. A real return is a nominal return minus inflation, and the period covered includes years of unusually high global inflation. Some part of a declining real return over a twenty-year window can come from the subtraction rather than from the investing.

The reply does not disaggregate any of this, and a reader who takes "cautious posture" as the whole explanation is taking more from the sentence than it offers.

What to watch

Two things.

The first is whether the figure falls again next year. One year is a window effect plus a posture; two consecutive falls with the same explanation would make the posture the dominant story.

The second is whether the lower-risk stance is ever described as ending. The reply frames caution as a response to heightened volatility and uncertainty. If those conditions ease and the posture does not, then what was presented as a cyclical response was a structural change, and it would be worth saying so.