I would describe what is happening as a sharp correction within a still-intact long-term bull market in gold, rather than the beginning of a structural bear market.

Gold ended 2025 at approximately $4,368 per ounce, then rose to nearly $5,595 in January 2026, briefly fell to around $4,000 in June, recovered to $4,563 by the end of August, and is trading at roughly $4,288 on September 15.

Gold is therefore only about 2% below its end-2025 level, but approximately 23% below its January peak.

Why gold has corrected so sharply

The main factor right now is interest rates and, even more importantly, real yields.

Gold does not generate interest income. When investors can earn around 5% a year on long-term US government bonds, holding gold becomes relatively less attractive.

The yield on the 10-year US Treasury recently moved above 5%, reaching its highest level since 2023. At the same time, the market is now pricing in a greater chance of Federal Reserve rate hikes rather than cuts because inflation has started to accelerate again.

This is important because people often say:

In practice, the relationship is more complicated.

If inflation rises while interest rates remain low, gold usually performs very well.

But if inflation forces the Federal Reserve to raise rates aggressively, real yields and bond yields rise, creating serious competition for gold.

That is essentially what is happening now.

The World Gold Council has also identified high real yields as one of the main obstacles facing gold in 2026.

In addition, gold's January move had a significant speculative component. Gold rose by approximately 67% during 2025 and then gained another 24% at its January peak.

The move from roughly $2,600 to almost $5,600 in little more than a year was so rapid that the price had clearly run ahead of the fundamentals. Profit-taking and a reduction in speculative positions were therefore almost inevitable.

Investment flows created additional pressure.

In the second quarter, gold ETFs recorded outflows of approximately 45 tonnes of gold, with about 74 tonnes leaving in June alone.

Central-bank buying also slowed temporarily. Net central-bank demand in the first half of the year was approximately 345 tonnes—the lowest first-half figure since 2022.

This was partly related to gold sales by Turkey, Russia, and Azerbaijan in the first quarter.

As a result, several negative forces were operating at the same time:

• an excessive January valuation → profit-taking;
• rising US real yields → a higher opportunity cost of holding gold;
• a shift from expectations of Federal Reserve rate cuts to possible rate hikes;
• periods of US dollar strength;
• selling by gold ETFs;
• a temporary slowdown in central-bank purchases;
• a very high gold price that reduced physical jewelry demand.

The long-term fundamentals remain strong

This is where my view becomes considerably more positive.

1. Central banks continue to accumulate gold

In my view, this is one of the most important changes in the gold market compared with what we saw 10–15 years ago.

In the second quarter, central banks purchased approximately 289 tonnes of gold, a record for the quarter and about five times the first-quarter total. China and Poland remained major buyers.

The World Gold Council's survey of central banks is particularly interesting:

This is no longer short-term speculation. It is a strategic diversification of international reserves.

Countries increasingly want assets with the following characteristics:

• no counterparty risk;
• no default risk;
• relatively low political and sanctions risk;
• independence from any single currency.

Gold performs this function extremely well.

I do not believe that de-dollarization will cause the US dollar to disappear as the world's reserve currency—we are still a very long way from that.

However, even a relatively small reallocation of global reserves into gold could have a significant effect on its price because the investable gold market is much smaller than the global bond market.

2. ETF investors are already returning

This is one of the most positive recent signals.

After heavy selling during the first half of the year, August brought approximately $18 billion of inflows into global gold ETFs, the second-largest monthly inflow on record.

Gold holdings in ETFs increased by approximately 121 tonnes to a record 4,189 tonnes. By the end of August, year-to-date net ETF inflows had reached about $29 billion.

Gold itself gained approximately 13% in August, supported by both ETF inflows and a weaker dollar.

To me, this indicates that strategic demand for gold has not disappeared.

3. Government debt and budget deficits

In my view, this factor is underestimated.

The US government must issue increasing amounts of debt to finance large budget deficits. This puts upward pressure on long-term government-bond yields and intensifies concerns about the sustainability of public finances.

The paradox is that this can be negative for gold in the short term because bond yields rise.

But if the market begins to see high yields not as a sign of a strong economy but as evidence of fiscal risk, the situation could change.

In that environment, gold becomes especially attractive.

The World Gold Council has already noted that concerns about government finances were one reason European and US investors returned to gold ETFs in August.

This is potentially a very powerful long-term theme.

4. High interest rates eventually create problems of their own

This is another reason I am much more optimistic about gold over the next 2–5 years than over the next 2–5 months.

With Treasury yields around 5%, servicing government debt becomes substantially more expensive.

High interest rates also put pressure on:

At some point, something usually begins to break.

If economic growth slows sufficiently, the Federal Reserve will eventually have to stop raising rates and later begin cutting them.

Once the market believes that real yields have peaked, one of the main current headwinds for gold will disappear.

Some major banks already expect rate cuts later in 2027, even though near-term expectations have become more hawkish. That transition could begin the next major upward wave in gold.

The next few months could remain difficult

At approximately $4,288, gold still faces significant short-term macroeconomic pressure.

The market is considering the possibility of another Federal Reserve rate hike, oil is above $100, inflation is accelerating again, and the 10-year US Treasury yield has moved above 5%. This is almost a textbook negative environment for an asset that generates no interest income.

I would therefore not be surprised to see gold make another move towards:

if US bond yields continue to rise.

However, I would view gold in the $3,800–4,100 range very differently from gold above $5,500 in January.

At $5,500, an enormous amount of optimism and fear was already reflected in the price. Near $4,000, you are buying the same asset at a substantially lower price while many of the long-term fundamentals remain intact.

On the other hand, if gold moves decisively back above approximately $4,550–4,700 and bond yields stop rising, I would see that as a sign that the correction is probably approaching its end.

In that case, $5,000 becomes the obvious next major target, followed by a possible return to the all-time-high area of $5,400–5,600.

The factor I would watch most closely

Not CPI. Not even the Federal Reserve's policy rate itself.

I would watch real yields on US government bonds.

Then add the behavior of the dollar. The most favorable environment for gold looks roughly like this:

If three or four of these factors begin to work at the same time, I believe gold could easily retest $5,500 and subsequently set new all-time highs.

The current situation is rather unusual

This is why gold is under pressure now even though, in other respects, the fundamental environment looks almost ideal for it.

My overall view

Over the next 3–6 months, I am neutral to moderately cautious on gold. I would not rule out another test of $4,000.

Over the 2027–2028 horizon, my view is considerably more positive.

I believe the probability that gold eventually rises above its January record of approximately $5,595 is significantly greater than the probability that $5,595 marks the final top of the current long-term gold cycle.

The reason is not simply “inflation.”

Importantly, the current correction has already removed a significant portion of the speculative excess that existed when gold traded above $5,500.

If we consider gold specifically as an investment rather than only as a macroeconomic theme, the next step is to identify the levels at which small / medium / aggressive position accumulation may be appropriate—for example, $4,200, $4,000, $3,800, and below—while assessing the potential return and risk at each level.