Meryl Pick, Portfolio Manager at Old Mutual Investment Group
“It is not what we don’t know that gets us into trouble. It’s what we know for sure that just ain’t so.” — Mark Twain
Imagine it’s the 1970’s and you are managing a South African equity fund.
Your benchmark looks nothing like today’s market. Resources make up roughly 68% of the All Share Index. South Africa produces close to 70% of the world’s gold. A decade later, gold shares still account for 38% of the JSE.
So how do you build a diversified portfolio? Do you hold less gold because the index looks too concentrated, or accept what the market is telling you?
The monetary system also changed. Bretton Woods had broken down, the US closed the gold window in 1971, and investors worried about inflation and monetary instability bought gold aggressively. Its price eventually rose from US$35 an ounce to around US$800 by 1980.
Fast forward to February 2026. Gold shares reached 18.9% of the FTSE/JSE All Share Index, prompting a familiar question: is that too much gold for a diversified portfolio?
History makes that weight look different. For many investors, the past 20 or 30 years provide the main frame of reference. Gold was relatively small, inflation fell, bonds diversified equities, and the dollar sat comfortably at the centre of the global financial system. Through that lens, 18.9% looks extraordinary. Through the longer JSE history, it does not.
Familiar does not mean permanent
Investment frameworks are backward looking. We observe how assets behave, identify persistent relationships, and use them to construct portfolios for the future. Yet investors may transfer wealth to the next generation in a world different from the one in which they accumulated it.
For much of the past four decades, the US dollar sat at the centre of the global financial system, inflation generally declined, developed market interest rates trended lower and government bonds became a reliable counterweight to equity risk.
Those relationships shaped modern portfolio construction, but they shouldn’t be extrapolated indefinitely. Monetary systems evolve, and so do relationships between assets. Bretton Woods is a reminder. The dollar was convertible into gold at US$35 an ounce until President Nixon closed the gold window in 1971.
The lesson isn’t that another rupture is imminent. Monetary systems reflect economic and geopolitical circumstances. Familiarity does not mean permanence.
Central banks are changing their behaviour
Deep US capital markets and network effects support the dollar’s dominant reserve role. Yet countries are reassessing how much their reserves should depend on another sovereign state’s financial infrastructure.
The freezing of Russia’s foreign exchange reserves after its invasion of Ukraine in 2022 sharpened that debate by showing how geopolitical decisions can affect reserve assets.
Central banks offer another sign of change. They sold significant amounts of gold throughout the 1990s and early 2000s. That trend reversed around the Global Financial Crisis and accelerated after 2022. Recently, central banks have absorbed roughly a quarter to a third of annual mined gold production. Reserve managers consider risks like sanctions, geopolitical fragmentation, currency concentration and confidence in reserve assets.
Gold has distinctive characteristics. It’s not another entity’s liability and, when held domestically, does not depend on another country’s financial infrastructure.
This doesn’t require forecasting the dollar’s decline or a return to the gold standard. The measured conclusion is that plausible monetary outcomes have widened.
From monetary architecture to portfolio architecture
Diversification depends on expected returns and how assets respond to economic shocks. For much of the past four decades, developed market government bonds provided an effective counterweight to equity risk. When growth weakened, subdued inflation often allowed central banks to cut rates, pushing bond prices higher.
The simultaneous decline in global equities and bonds in 2022 challenged that assumption. South African balanced portfolios have never fitted neatly into the 60/40 model, but the question is universal: what provides resilience when historical relationships between traditional asset classes become less reliable?
The answer is not to abandon equities or developed market government bonds. Equities remain a key source of long-term growth, while government bonds can still provide liquidity, income and protection in particular environments. Portfolios built for more varied economic and monetary regimes may need a broader mix of return drivers.
Gold can play a different role because its drivers extend beyond the conventional economic cycle to include real interest rates, currency confidence, geopolitical risk and demand for monetary reserves. Central bank accumulation has also added a structural source of demand that mattered far less during the previous investment regime.
For multi-asset portfolios, this supports considering a wider role for gold, through physical gold and gold equities, together with diversifiers such as private equity and emerging market equities. Their rationales differ, but together they can reduce reliance on the equity bond relationship that worked well in the previous regime.
Updating frameworks, not simply forecasts
We don’t need to forecast the demise of the dollar or predict a new gold standard. What matters is recognising when the range of possible outcomes has widened enough to warrant a change in portfolio construction.
Today, gold’s strategic case doesn’t rest on recent price appreciation. Central banks have become structural buyers, geopolitical considerations increasingly influence reserve management, and debate around the international monetary system has returned.
The broader lesson goes beyond gold. Investment frameworks become most vulnerable when their assumptions fade from view. Investors formed their assumptions about inflation and interest rates, equities and bonds, and reserve assets and geopolitical risk during a particular period.
As that environment changes, portfolios may need to change with it. Rather than position for one alternative future, investors can build portfolios resilient across several plausible futures. When the evidence changes, investors should update more than their forecasts. They should update their frameworks and the portfolios built from them.
The portfolios we build today shouldn’t simply reflect the world we inherited. They need to navigate the world the next generation will inherit.
ENDS
