Why Norway sold its gold in 2004

The decision in 2004

By the end of 2003, Norway's central bank held about 37 tonnes of gold — 33.5 tonnes of bars plus 3.5 tonnes of historic coins from the 1940 "gold transport" to England. By the end of Q1 2004 the bars were gone: Norges Bank sold 33.5 tonnes for USD 447.4 million, kept seven bars for exhibition, and the coins stayed.

It was a quiet decision. There was no political fight. No referendum. The press release ran a few lines. Twenty-two years later, the dollar gold price is many times its 2004 level — and Norway, home to one of the world's largest sovereign wealth funds, holds essentially no gold in its reserves.

This is the part of the story most Norwegians don't know. And it matters, because it changes how you should think about gold in 2026.

The reasoning Norges Bank gave

The official rationale from the central bank was simple and, on paper, defensible. Gold made up just over 1% of Norges Bank's international reserves at the time. The bank's view was that 1% of a reserve portfolio cannot meaningfully diversify risk, and the historical return on gold was low compared with the bonds and equities the rest of the portfolio held.

So gold went. The proceeds were invested as part of Norges Bank's foreign exchange reserves, and Norway moved on.

To understand why this looked reasonable in 2004, you have to remember what gold had done for the previous twenty years. Between January 1980 and July 1999, gold went from $850 per ounce to a 21-year low of $252.80. Two decades of decline, in real terms, in a period when stock markets compounded relentlessly. A central banker looking at the data in 2003 saw a sleepy, expensive-to-store asset that had been one of the worst-performing major asset classes of their entire career.

The decision wasn't crazy. It was conventional.

The cost of being conventional

Since then the dollar gold price has risen many times over: the LBMA's Alchemist notes that it passed US$1,000/oz during the 2008 financial crisis, US$2,000/oz in 2020 and set a record of US$3,500/oz on 22 April 2025.

A back-of-envelope calculation: if Norway had simply kept those 33.5 tonnes (about 1.077 million ounces), at the April 2025 record of US$3,500/oz the position would have been worth about USD 3.8 billion, against the USD 447.4 million realised in 2004. That is not a small mistake at the level of a central bank's portfolio, but it is also not a catastrophic one in the context of Norway's sovereign wealth fund. The fund's equity allocation did most of the heavy lifting Norges Bank assumed it would.

But the symbolic loss matters separately from the financial one. Many central banks have since done the opposite of what Norway did — buying gold, not selling it — with central banks in emerging and developing economies among the most active. The LBMA's Alchemist describes recent central bank demand as near unprecedented, with official-sector gold holdings approaching their 1965 all-time high.

Norway is on the other side of that trade. By choice.

What changed in the world after 2004

The arguments against gold in 2003 — low yield, high storage cost, no industrial demand of consequence — haven't changed. What changed is the rest of the financial system.

Three things in particular:

First, real interest rates collapsed. The opportunity cost of holding a non-yielding asset like gold is higher when bonds pay solid real yields. When real rates fell to around or below zero for long stretches after the financial crisis, gold's "no yield" problem mattered much less. Central bankers re-noticed.

Second, sanctions risk arrived as a portfolio consideration. The freezing of Russian central bank reserves in 2022 was a clarifying moment for non-Western central banks. Foreign-currency reserves can be turned off. Physical gold in your own country's vault cannot. This is why central banks from Asia, the Middle East, and Latin America have been the most aggressive gold buyers post-2022.

Third, the structure of fiat currency itself was tested by 2020-2022 inflation in a way it had not been since the 1970s.

None of this was visible to Norges Bank in 2003. Some of it might have been guessed at, but the institution made a decision based on twenty years of data that turned out to describe the wrong twenty years.

What this means for Norwegian gold buyers in 2026

Germany, Italy, Portugal, and France all have deep traditions of private gold ownership — partly because their 20th-century currencies failed at various points and the public learned the lesson directly. Norway's 20th century, by contrast, was a story of currency stability, oil discovery, and the gradual construction of one of the most successful sovereign investment vehicles in history. Norwegians had no reason to own gold privately because the state seemed to be handling reserves competently.

If you accept that the state's 2004 decision was, with hindsight, wrong, the question becomes whether the conditions that made that decision look reasonable are still in place. They mostly aren't. The krone has been one of the weaker G10 currencies of the past decade. The petroleum fund's returns increasingly depend on US equity beta, which is not a hedge against the things gold hedges against. And real rates have been more volatile in the 2020s than at any time since the 1980s.

This doesn't mean every Norwegian household should be allocating 20% to gold. The case for a small private gold position is what it has always been: it pays nothing, it costs a little to store, it is liquid globally, and it does well when other parts of a portfolio do badly. For most people, somewhere between 2% and 10% of investable assets is a reasonable bracket — closer to 2% if you have meaningful exposure to global equity through the fund or your pension, closer to 10% if you are concerned specifically about currency risk on the krone. Capital at risk: gold can fall in price as well as rise, and nothing here is financial advice.

The point is not that Norway sold at the bottom. The point is that the institution that was supposed to be diversifying Norway's reserves removed the one asset class designed to do exactly that, on the grounds that 1% of a portfolio could not diversify it.

If 1% is too small to matter, the obvious follow-up question is whether 5% or 10% would matter. Norges Bank never seriously answered it.

What we would do differently

This is the part where most articles tell you to buy a specific gold product. We will not, because the right answer depends on what you are trying to achieve.

If you want gold for crisis insurance, you want it in physical form, in a country that respects property rights, ideally outside the banking system. Bullion coins from a reputable mint, or 1-gram to 100-gram ingots, are the standard way to do this. For Norwegian buyers, the practical options include international bullion dealers such as Silver Gold Bull (check that it currently delivers to Norway and how shipments are insured) and Asian gold-jewellery houses like Chow Sang Sang, which sells 999.9 gold ingots and jewellery.

If you want gold for portfolio diversification but don't care about owning the physical metal, the answer is different — an ETF or an allocated gold account at a brokerage gives you the exposure without the storage problem. The cost is that you are back to relying on a financial intermediary, which is the thing gold is supposed to protect you from.

Most people who think they want gold actually want a small position they will never sell. That makes 1-gram to 10-gram ingots, or quarter-ounce coins, more useful than larger bars. You can give them as gifts, store them in three different places, and you never face the "I have to sell the whole brick or sell nothing" problem.

Norway sold the whole brick in 2004. We don't have to make the same trade.

Sources

Norges Bank press release, 28 January 2004: "Norges Bank has sold some gold reserves." Norges Bank press archive.

Norges Bank FAQ, "Gold is no longer included in Norges Bank's international reserves." Norges Bank.

Statistics Norway: International reserves and foreign currency liquidity time series, 2003-2025.

Wikipedia, "Gold holdings of Norway" and "Flight of the Norwegian National Treasury" — both well-sourced overview articles for the wartime evacuation history.

World Gold Council, central bank gold demand data 2010-2025.

Frequently Asked Questions