That disagreement is the real story. Gold is not one thing in a portfolio; it is potentially three different things, and the evidence for each is quite different. Confusing the three is how popular claims about gold — that it "protects against inflation," or that it can "replace bonds" — end up overstating what the data actually shows.
Three jobs, one metal
Treat gold the way a portfolio manager would: ask what specific problem it’s supposed to solve. According to the framing used by asset allocator Paul Kenney, there are three candidate roles — return driver, diversifier, and hedge — and gold’s case is strong in some and weak in others.
| Role | What "success" would look like | Strongest evidence | Main limitation |
|---|---|---|---|
| Return driver | Consistent capital growth over time | GLD ETF up roughly 125% over the past three years through mid-August | No yield or earnings; took nearly 28 years to reclaim its 1980 nominal peak |
| Diversifier | Low correlation with stocks, smoothing portfolio swings | 20-year correlation of GLD to the S&P 500 and Russell 2000 near 0.06 — essentially none | Low correlation doesn’t guarantee gains; gold spent almost three decades producing near-zero returns despite that low correlation |
| Hedge | Offsetting losses elsewhere under specific stress conditions | Central banks flipped from net sellers to sustained net buyers of gold since 2010, and 2026 activity remains solid, led by China and Poland | Effectiveness depends on why the stress happens — it’s not a mechanical response to inflation itself |
The table is useful precisely because it separates questions that get blended together in casual conversation. "Does gold go up?" is a different question from "does gold zig when stocks zag?" which is a different question again from "does gold protect me from a specific kind of disaster?"
The return-driver case is the weakest
If you’re holding gold expecting steady capital appreciation the way you might expect from equities over decades, the historical record is not encouraging. Gold produces no dividends, no interest, no earnings growth — its price appreciation is the entire return. And that price history includes an extraordinarily long drought: after peaking in nominal terms in January 1980, gold didn’t reclaim that high until 2008, according to London Bullion Market Association and World Gold Council data cited in the source.
The recent run has been strong, and current demand is being supported by falling real rates, still-elevated geopolitical risk, and continued central-bank buying, according to World Gold Council outlook research. But strong recent performance is a description of the past, not a forecast. Whether central banks keep buying at a similar pace, and whether institutional and retail investors deepen their allocations, are open questions — and the World Gold Council’s own commentary acknowledges that predicting the next phase of central-bank demand is "inherently difficult," partly because banks don’t all target the same metrics or manage reserves the same way. Treating a multi-year rally as evidence that gold has become a reliable growth engine risks mistaking a favorable stretch for a permanent shift.
Low correlation is not the same as a free lunch
The diversifier case is more defensible, but it comes with an important caveat that the brief’s evidence makes explicit. Over the twenty years through mid-August, GLD’s correlation to large-cap and small-cap U.S. equities was close to zero, and it delivered an annualized return of roughly 9.8% — ahead of small caps, behind the S&P 500, and still respectable even if you strip out the recent three-year surge.
But correlation measures co-movement, not profitability. An asset can be perfectly uncorrelated with stocks and still be a drag on a portfolio if it simply goes nowhere for years. That is exactly what happened for much of gold’s post-1980 history: nearly three decades of flat-to-poor performance that would have offered diversification benefits on paper while contributing little in practice. This is the crux of why "diversifier" and "return driver" have to be evaluated separately — an investor who wants both from gold at once may be asking too much of a single asset.
There’s also a structural argument for why the diversification case might be more relevant now than in the recent past. The 60/40 stock-bond framework relied on equities and bonds moving in different directions during stress. Research from FTSE Russell notes that bond-equity correlations have risen, particularly since 2022, meaning bonds have not always cushioned equity losses the way the old playbook assumed. If that pattern persists, a modest allocation to an asset with genuinely low correlation to both stocks and bonds becomes more valuable as a complement — though this is a structural observation about correlation regimes, not a guarantee that the pattern will hold going forward.
What gold actually hedges — and what it doesn’t
This is where the most common misconception lives. Gold is frequently pitched as an inflation hedge, evoking a simple story: prices rise, gold rises to match. The evidence doesn’t support that mechanical version. Gold’s performance depends heavily on how policymakers respond to inflation — aggressive rate hikes and a strengthening dollar can suppress gold returns even during high inflation, while "sticky" inflation paired with falling real yields tends to help it. In other words, gold is more sensitive to real interest rates and policy credibility than to the inflation rate itself.
The more coherent framing, laid out in the central source, treats gold as a hedge against the dollar and against fiscal/currency debasement risk rather than consumer prices directly. The reasoning: U.S. gross federal debt has grown from roughly $5.7 trillion in 2000 to about $40 trillion today, debt per capita has climbed past $112,000, and the Congressional Budget Office projects debt reaching 190% of GDP by 2056, with interest costs already consuming a growing share of the federal budget. None of this proves the dollar will falter — it’s a reason to think through what happens to a portfolio if it does, since both bonds and equities, priced in dollars, could struggle simultaneously in a currency-stress scenario. Gold’s appeal there rests on having no counterparty risk, not on any promise tied to CPI.
The takeaway is about roles, not verdicts
None of this settles whether an individual investor should own gold, and the evidence here doesn’t stretch that far. What it does clarify is that gold’s case looks different depending on the question being asked. As a return driver, it’s a yield-free asset with a genuinely rough multi-decade stretch in its history. As a diversifier, its low correlation to equities is real but not sufficient on its own — a diversifier that stagnates for years can still disappoint. As a hedge, it’s better understood as protection against currency debasement and policy stress than as an automatic inflation offset, and even that role depends on scenarios that may or may not materialize.
Strong recent institutional and central-bank demand strengthens the argument that gold deserves consideration as a strategic component of some portfolios. It does not make gold a substitute for high-quality bonds across all environments, nor does it turn a good multi-year run into a promise about what comes next. The honest answer to "what is gold good for?" is: it depends which of the three jobs you’re actually hiring it to do.
