Gold Price History 1971-2026: Every Major Rally and Crash, and What Forecasts Say Now

Gold is one of the most talked-about investments in America, and also one of the most misunderstood. Advertisements tend to show only the rising parts of the chart. Critics tend to show only the falling parts. This guide walks through the whole thing, from the last days of the official $35 price to the record of early 2026 and the sharp drop that followed, so you can see the full pattern for yourself.

Here is what you will find below:

• Every major rally and crash since 1971, with the dates, the prices, and what was happening in the world at the time.
• Year-by-year tables, including year-end prices for 2015-2025, annual averages, and snapshots from earlier decades.
• How the gold price is actually set, and why two respected sources can quote different "record" prices for the same day.
• Gold after inflation, gold against stocks, and silver's history in brief.
• The six forces that have moved gold, each with real historical examples.
• Every major drawdown and how long recovery took - the part of gold's history that matters most to someone living off their savings.
• What the big banks forecast now, and why their record suggests you should not lean on any single forecast.
• What this history means for a retirement saver, plus answers to the questions we are asked most.

Gold's modern history at a glance

Each of these rows has its own section below, with sources. Where sources disagree - and for older dates they often do by a few dollars - we say so.

How we wrote this

• Every figure is dated and sourced. Current prices are as of October 19, 2026. Historical prices come from the World Gold Council, the LBMA, the Federal Reserve, Kitco, Reuters reports and other named sources. Older figures in particular vary slightly between data sets, so treat them as approximate.
• We show our arithmetic. Where we calculate a percentage change, we show the inputs so you can check them. Calculated figures are rounded and ignore taxes, dealer markups and fees, which a real investor would pay.
• We make no predictions of our own. In the forecasts section we report what banks have published, with dates, and show how often those forecasts have been revised. We do not tell you where gold is going, because nobody knows.
• Past performance is not a guide to the future. History shows the range of things that have happened. It does not tell you what will happen next.
• This is education, not advice. Whether gold belongs in your retirement savings depends on your whole financial picture. A fee-only financial adviser - one you pay directly, not one paid by commission on what you buy - can help you decide.

If you are considering gold for retirement specifically, our companion guide Is a Gold IRA a Good Investment? uses the same data to weigh the pros and cons, including what fees do to your returns. This article focuses on the price itself. New to the topic? Our gold IRA guide explains how gold is held in a retirement account.

When a news report says "gold hit a record," it sounds as if there is one official price. There is not. There are several closely linked prices, each measured a slightly different way. Understanding them will save you a lot of confusion when you compare charts, read a dealer's website, or check what your gold is worth.

The four prices you will see

The LBMA (London Bullion Market Association) is the trade body for the London market, which is the largest hub for physical gold trading in the world. Its twice-daily auction gives a single published number that everyone can refer to, which is why official statistics usually use it. The World Gold Council, for example, counts "record highs" using the LBMA PM price.

Why sources quote different records: January 29, 2026

The best example is gold's all-time high. On January 29, 2026:

• Spot gold touched about $5,595 an ounce intraday (Reuters reported a record of $5,594.82).
• The LBMA AM price that morning was $5,501.70.
• The LBMA PM price that afternoon was $5,405.00.
• Some data services that track daily closing prices list the record close as about $5,318.

All four numbers are "correct" - they just measure different moments. The spot high was a brief spike during trading; the LBMA prices were set at fixed times; the close was the price at the end of the trading day. The gap between the intraday high and the LBMA PM price was about $190, or 3.4% ($5,595 / $5,405 - 1 = 0.035). On a fast-moving day, that kind of gap is normal.

The same thing happens in history. The 1980 peak is usually given as $850, which was the London price on January 21, 1980, but New York futures traded as high as about $875 that day before closing near $825, according to a Numismatic News retrospective. The 2011 peak is variously given as $1,895 (LBMA PM), $1,920.70 (spot intraday) or $1,923.70 (COMEX futures). When you see two different "record" figures, check which price each source is using before assuming one is wrong.

Why the dealer price is never the spot price

If you buy a one-ounce gold coin, you will not pay spot. You pay spot plus a premium that covers minting, distribution and the dealer's margin. When you sell, the dealer pays you spot minus a small discount, or occasionally a little over spot for popular coins. Typical figures from our research in 2026:

Phone-based gold IRA dealers often charge considerably more - sometimes 10% or higher on bullion and far more on "premium" or proof coins - which is why the price history in this article is always a best case for an actual buyer. Our gold IRA fees guide and fee calculator show what those costs do to your results.

Which price we use in this article

• For records and lows, we give the figure most commonly reported and say whether it is intraday spot, an LBMA price or a close.
• For year-by-year tables, we use year-end prices and annual averages from the sources named under each table. The World Gold Council's annual averages are based on the LBMA PM price.
• For today's price, we use Kitco's spot quote on October 19, 2026: about $4,170.

Differences of a few dollars - or, for very old dates, a few percent - between sources do not change any of the big lessons in this article.

To understand why 1971 is the starting line for almost every gold chart, you need to know that for most of American history the price of gold was not a market price at all. It was set by law.

1900-1933: $20.67 an ounce

Under the classical gold standard, a US dollar was defined as a fixed weight of gold, which worked out to $20.67 per troy ounce. Paper money could be exchanged for gold coin at that rate. Because the price was fixed, there was no "gold price chart" in the modern sense - gold did not go up or down against the dollar; the dollar was gold.

1933-1934: the government takes the gold

During the Great Depression, banks failed and people hoarded gold. In 1933, President Franklin Roosevelt issued an executive order requiring Americans to turn in most of their gold coins, bullion and gold certificates to the Federal Reserve. Private ownership of gold bullion was effectively banned, with exceptions for jewelry, some collectors' coins and industrial uses.

The Gold Reserve Act of 1934 then raised the official price from $20.67 to $35 an ounce - an overnight devaluation of the dollar against gold of about 41% (1 - $20.67 / $35 = 0.41). Seen the other way, the gold the government had collected became worth about 69% more in dollars ($35 / $20.67 - 1 = 0.69).

1944-1971: Bretton Woods and the $35 promise

At the end of the Second World War, the Allied nations met at Bretton Woods, New Hampshire, and set up a new international money system. Under it:

• Other countries pegged their currencies to the US dollar.
• The US promised that foreign governments and central banks could exchange dollars for gold at $35 an ounce.
• Ordinary Americans still could not own gold bullion.

The system worked for about two decades, but the strain grew through the 1960s. The US was spending heavily on the Vietnam War and domestic programs, inflation was rising, and the number of dollars held abroad grew far larger than the US gold stock could cover. In 1968 a group of central banks known as the London Gold Pool gave up trying to hold the market price at $35, and a "two-tier" system appeared: central banks still dealt with each other at $35, while a private market in London traded at whatever price buyers and sellers agreed.

August 15, 1971: Nixon closes the gold window

By the summer of 1971, foreign governments were lining up to swap their dollars for American gold. Over the weekend of August 13-15, President Nixon met with about fifteen advisers at Camp David, including Federal Reserve Chairman Arthur Burns, Treasury Secretary John Connally and Treasury Undersecretary Paul Volcker. On Sunday evening, August 15, he announced that the US would stop exchanging dollars for gold, along with a 90-day freeze on wages and prices.

According to the Federal Reserve's own history of the decision, the first order was to close the gold window so that foreign governments could no longer exchange their dollars for gold. The move was presented as temporary. A few months later, the Smithsonian Agreement tried to patch the system together with new exchange rates (and a new official gold price of $38, raised again to $42.22 in 1973), but by 1973 the major currencies were floating freely and the Bretton Woods system was over for good.

That is why most gold price histories start in 1971. From that point, gold's dollar price was set by buyers and sellers in the market rather than by the US Treasury - and it began to move.

What this means for reading long-term charts

• A "100-year gold chart" mixes two regimes. Before 1971 the line is flat because the price was fixed by law. Comparing gold's "return" over 1920-2026 with stocks over the same period is misleading, because gold could not rise for half of that time.
• The $35 starting point matters. Many observers believe $35 was well below what a free market would have paid by 1971, so part of gold's early-1970s jump was a catch-up rather than a normal market rise.
• Americans could not legally own bullion until the end of 1974. The early part of the 1970s boom was driven by foreign buyers and by Americans buying gold indirectly (for example through mining shares or legal coins).

The 1970s were gold's first decade as a free-floating asset, and they produced the largest rise in its modern history in percentage terms. The decade had two separate surges with a painful drop in between.

The first surge: 1971-1974

Once the dollar was no longer tied to gold, the price rose quickly. Inflation was climbing, the 1973 Arab oil embargo quadrupled oil prices, and confidence in paper currencies fell. Federal Reserve monthly data put gold at about $43.48 in December 1971 and about $63.91 in December 1972. By the end of 1974 it was close to $200: Numismatic News and contemporary newspaper reports say gold topped $195 an ounce in London on December 30, 1974, before slipping to about $186.50 the next day.

December 31, 1974: Americans can own gold again

Public Law 93-373, signed by President Gerald Ford in August 1974, took effect on December 31, 1974. For the first time in about 41 years, US citizens could legally "purchase, hold, sell, or otherwise deal with gold" (the World Gold Council marks this as one of the key moments in its 50-year history of the US gold market).

Many people expected a buying frenzy. It did not happen. Newspaper reports from the first day describe only a trickle of purchases instead of the widely predicted gold rush, and the price had already peaked in anticipation. This is one of the oldest lessons in gold's history: by the time a widely expected event arrives, it is often already in the price.

The mid-decade crash: 1975-1976

After the legalization, gold fell for about 20 months. The US Treasury and the International Monetary Fund both sold gold from their reserves, inflation eased for a while, and the investors who had bought in anticipation sold out. The US Bureau of Mines' Minerals Yearbook lists a 1976 low of about $103.05 (in September; other daily data sets show a low near $100). From the December 1974 high of about $195, that was a fall of roughly 47% ($103 / $195 - 1 = -0.47).

Anyone who bought gold the day it became legal for Americans would have lost almost half their money within two years. This is worth remembering whenever gold is in the headlines.

The second surge: 1977-1979

From 1977 the price climbed again, and then rose very fast in 1979. The forces behind it were some of the most powerful in gold's history:

• Double-digit inflation. US consumer prices rose more than 11% in 1979 and more than 13% in 1980.
• Negative real interest rates. For much of the decade, interest on savings and Treasury bills was below inflation, so holding cash lost purchasing power every year - which made a non-paying asset like gold relatively more attractive.
• A second oil shock after the 1979 Iranian revolution.
• Geopolitical fear: the Iran hostage crisis from November 1979 and the Soviet invasion of Afghanistan in late December 1979.
• A weakening dollar and doubts about US economic policy.

The 1970s year by year

Sources: Federal Reserve monthly gold series (1970-1972); Macrotrends annual closes (1977-1979); Minerals Yearbook and MetalCharts for 1976; contemporary press for 1974. The 1973 and 1975 figures are commonly cited London prices that we could not confirm in a primary data set in October 2026, so treat them as approximate. Different data series can differ by several percent for the 1970s because the market was thinner and was quoted in several places.

The 1970s in numbers

• From $35 to about $512 (year-end 1979): roughly 14.6 times ($512 / $35 = 14.6), or about +1,360%.
• From $35 to the January 1980 peak of $850: about 24 times ($850 / $35 = 24.3) in less than nine years.
• The worst drop along the way: about 47% (1974-1976).
• From the 1976 low of ~$103 to $850: about 8.3 times ($850 / $103 = 8.25) in about three and a half years.

Two important caveats. First, the 1970s started from a price that had been held artificially low by law for decades, so some of the rise was a one-off adjustment that is unlikely to be repeated. Second, inflation was very high, so the gain after inflation was much smaller than the headline - though still large. We return to inflation-adjusted prices in a later section.

What the 1970s taught investors

The decade created gold's reputation as an inflation hedge, and much of what gold sellers still say today traces back to it. But the decade also contained a near-halving of the price in 1975-1976, and - as the next section shows - it ended with a peak that took 28 years to regain in plain dollars. Both halves of the story are true.

Gold's 1970s boom ended in a spectacular spike. In the first three weeks of January 1980, with the Soviet army in Afghanistan, American hostages held in Tehran and inflation in double digits, the price shot up by hundreds of dollars. On January 21, 1980, the London price reached about $850 an ounce. In New York, futures briefly traded around $875 before closing near $825, according to a Numismatic News retrospective.

It was a classic mania. Silver was rising even faster, helped by the Hunt brothers' attempt to corner that market. (APMEX notes that the Hunts' buying helps explain silver's spike but not gold's.) Ordinary people queued outside dealers to sell jewelry and family silverware for cash.

Why the peak ended

The turn came from interest rates. Paul Volcker, who had become Federal Reserve Chairman in August 1979, set out to break inflation by squeezing the money supply. Short-term interest rates rose toward 20%. Suddenly, cash in the bank or in Treasury bills paid far more than inflation. That reversed the single most important force behind the 1970s boom: when you can earn a high real (after-inflation) return on safe savings, the case for holding gold, which pays nothing, weakens sharply.

Gold fell back below $500 within weeks of the peak and, after a rebound later in 1980, spent the next two decades mostly drifting down.

Twenty years of decline: 1980-1999

The bear market was long and grinding, with occasional rallies that came to nothing. The main reasons cited by historians and market analysts:

• Inflation fell. From over 13% in 1980 to around 3-4% for much of the late 1980s and 1990s, removing gold's most powerful selling point.
• Real interest rates stayed positive. For most of the period, safe savings paid more than inflation.
• Stocks boomed. The great bull market of 1982-2000 gave investors a far more rewarding home for their money.
• Central banks were sellers. European central banks sold large amounts of their reserves in the 1990s, and mining companies sold their future production forward ("hedging"), adding supply.
• A strong dollar for much of the late 1990s.

The bottom: the Bank of England sales, 1999

In May 1999 the UK Treasury announced it would sell more than half of Britain's gold reserves by auction. The first auction took place in July 1999, and gold fell to about $252.80 - a level the LBMA's Alchemist magazine describes as a historic low. Sources date the exact low to July 20, 1999 (some summaries round it to "about $252 in August 1999" - the difference is a few weeks). The UK eventually sold about 395 tonnes in 17 auctions between 1999 and 2002 at an average of about $275 an ounce, a decision that Kitco and others have since described as badly timed.

The selling stopped at the bottom. On September 26, 1999, 15 European central banks signed the Washington Agreement on Gold, limiting their combined sales to 400 tonnes a year for five years. Gold jumped from around $260 to around $330 within about two weeks. It would still be another two years before the long-term trend clearly turned.

The bear market in numbers

Decade snapshots: 1985-2000

The figures marked "commonly cited" appear consistently across price-history sites but we could not check them against an LBMA or Federal Reserve series in October 2026, so treat them as approximate. The overall picture - gold stuck in a roughly $250-$500 range for about two decades - is not in doubt.

What it was like to own gold through this

Think about someone who bought $10,000 of gold at the January 1980 peak at age 45. At age 64, in 1999, it would have been worth about $3,000 ($10,000 x $252.80 / $850 = $2,974), before any costs. Meanwhile, prices in the shops had more than doubled. They would have been in their early 70s before the gold was back to $10,000 in plain dollars. A buyer at a more ordinary price - say $400 in the mid-1980s - did better, but still waited roughly 20 years for a meaningful gain.

This is the single most important episode in gold's history for anyone close to retirement. It shows that gold can lose most of its value and stay down for longer than many people's entire retirement. We look at what that means for planning in the retirement section.

After two decades in the doldrums, gold began a rise that lasted about ten years and multiplied its price more than seven times. Unlike the 1980 spike, this rise was steady for most of its length, with only one serious interruption, in 2008.

The starting point

Gold made a second, slightly higher low in early 2001, at roughly $255-$260 an ounce (commonly cited; the exact low varies by source). From there, it rose in every calendar year from 2001 to 2012 - one of the longest winning streaks of any major asset.

What drove the rise

The key milestones

• 2005: year-end close about $513 (Macrotrends) - roughly double the 2001 low.
• March 2008: gold crossed $1,000 for the first time, on March 13-14, as the investment bank Bear Stearns needed emergency funding. Reports from The Northern Miner and BullionVault differ on the exact day.
• Autumn 2008: during the worst of the financial crisis, gold fell - to somewhere around $680-$710 in October and November 2008 (sources differ). Investors sold whatever they could to raise cash and meet margin calls, and the dollar rose. From roughly $1,000 to roughly $700 is a drop of about 30%.
• 2009-2010: gold recovered and set new records. It ended 2010 at about $1,410 (Macrotrends: $1,410.25).
• August-September 2011: after the US lost its AAA credit rating from S&P and fear about European banks peaked, gold made its high. Mining Journal reports an intraday record of $1,920.70 on September 6, 2011; COMEX futures touched $1,923.70. It is usually rounded to $1,921 or $1,920.

The 2001-2011 bull market in numbers

• From about $255 (2001) to $1,921 (2011): about 7.5 times ($1,921 / $255 = 7.53), or about +650%, over roughly ten and a half years.
• Annual rate: about 21% a year compounded (7.53 to the power of 1/10.4 = about 1.214). That is an exceptional run for any asset and should not be treated as normal.
• The 2008 interruption: a fall of about 30% in a few months, in the middle of the crisis that gold buyers often expect it to protect against.

The 2008 lesson

Gold's behavior in 2008 is often misremembered. Over the whole of 2008, gold ended slightly higher, while the S&P 500 fell by more than a third. That is the good news for gold. But in the most frightening weeks of the crisis - September to November 2008 - gold fell sharply alongside stocks, because investors were forced to sell anything liquid. It then recovered much faster than stocks did. The same pattern appeared in March 2020. Gold has often helped over a crisis; it has not always helped during the worst days of one.

The top in September 2011 was followed by a four-year slide that cut gold's price nearly in half. For many people who had bought near the top - often encouraged by heavy advertising - it was a painful lesson that gold can fall a long way even when the headlines still feel frightening.

Why gold fell

• The crisis faded. The worst fears about the US economy and the euro did not come true. Stocks began a long bull market.
• Real interest rates rose. In 2013, the Federal Reserve signaled that it would slow ("taper") its bond-buying program. Ten-year Treasury yields jumped, and real yields moved from below zero to positive - historically bad news for gold.
• The dollar strengthened, especially in 2014-2015.
• ETF investors sold. Investors who had piled into gold ETFs in 2009-2012 sold large amounts in 2013, adding to the pressure.
• Inflation never took off. Many had expected the money-printing after 2008 to cause high inflation. It did not, at least not in consumer prices, which undermined one of the main reasons people had bought.

April 2013: the crash

The sharpest part came in April 2013. On Friday, April 12, and Monday, April 15, gold fell about 14-15% in two trading days. Reuters reported that Monday's drop of about $125 an ounce was the biggest one-day dollar fall on record at the time, with spot gold falling as low as about $1,321 - its lowest since January 2011. In percentage terms, the roughly 9% one-day fall was the biggest since 1983, and ABC News put the two-day decline at about 14.9%. Bloomberg called it the end of a 12-year bull run.

Explanations at the time included worries that Cyprus might sell part of its gold reserves, large sell orders in the futures market, and a stampede for the exits once key price levels broke. Whatever the trigger, the episode showed how quickly gold can move when many investors try to sell at once.

Over the whole of 2013, gold fell by roughly 28% - its worst calendar year in more than three decades. (Reported figures range from about 27% to 30% depending on the price series and dates used; we could not confirm a single exact figure.)

The bottom: December 2015

Gold kept drifting lower through 2014 and 2015. In early December 2015, as the Federal Reserve prepared to raise interest rates for the first time since 2006, spot gold fell as low as about $1,045.85, the lowest since early 2010. After the Fed's rate rise on December 16, gold futures briefly tested the same area again. The year ended at about $1,061, down about 10.7% for 2015.

The decline in numbers

What this episode taught

The 2011-2015 decline matters because it happened recently, to a generation of buyers who had seen gold rise every year for a decade and assumed it would continue. Three things stand out:

1. Gold fell even though many of the "reasons to own gold" were still being advertised. Government debt was still high and money-printing continued for years. The price fell anyway, mostly because real interest rates rose and fear faded.
2. A 45% fall on a well-known, widely held asset happened in about four years. That is comparable to a severe stock bear market.
3. Buyers at the top waited most of a decade to break even - longer if you include dealer markups and annual fees, which can easily add 10-20% to the break-even point for a physical gold IRA (see our gold IRA fees guide).

After the 2015 low, gold spent four years moving gradually higher, with setbacks along the way. It was an uneventful period compared with what came before and after, but it is useful because it shows what gold looks like in "normal" times: modest gains, modest losses, and no obvious trend for long stretches.

Sources: year-end prices via JM Bullion; reported yearly changes from published annual returns; silver averages from our research notes (October 2026). Reported yearly changes can differ slightly from the change you would calculate from the year-end prices, because sources use different closing times or benchmarks. For example, $1,152 / $1,061 - 1 = +8.6% for 2016 versus a reported +6.8%.

What stands out

• 2018 shows the interest-rate link clearly. As the Fed raised rates and the dollar rose, gold slipped even though stocks had a rough end to the year.
• 2019 shows the reverse. When the Fed switched to cutting rates, gold rose more than 18%. The annual average price in 2019 was about $1,393.
• Central bank buying was growing quietly. Countries such as Russia, China, Turkey and Poland were adding gold to their reserves - a trend that would become far more important after 2022.

From the end of 2015 ($1,061) to the end of 2019 ($1,517), gold rose about 43% ($1,517 / $1,061 - 1 = 0.43), or roughly 9% a year. That is a respectable result - but at the end of 2019, gold was still about 21% below its 2011 high ($1,517 / $1,921 - 1 = -0.21).

The Covid-19 pandemic produced gold's first new all-time high in nearly nine years - and, briefly, a sharp fall first.

March 2020: the dash for cash

When markets panicked in March 2020, gold initially fell along with almost everything else, dropping more than 10% in about a week from its early-March level as investors sold assets to raise cash. It was the same pattern seen in late 2008. Then central banks and governments responded with enormous support - interest rates cut to near zero, trillions of dollars of bond buying and large government spending programs.

The rise to a record

With real interest rates deeply negative (ten-year Treasury yields fell well below the rate of inflation), money pouring into gold ETFs and widespread fear about the economy, gold rose rapidly. In early August 2020:

• The LBMA gold price exceeded $2,000 for the first time in its more than 100-year history, at $2,034.45 on August 5 (Mining Weekly).
• Spot gold peaked at about $2,067-$2,075 on August 6-7, 2020. Price-history sites commonly give $2,067.15; the US Geological Survey's monthly report lists $2,060; a COMEX futures contract settled at a record $2,069.40.

As with every record in this article, the exact figure depends on which price you use. We use "about $2,067".

2020 in numbers

• Year-end: about $1,899, up about 25% for the year.
• Annual average: about $1,770.
• Silver averaged about $20.55, up from about $16.21 in 2019.
• Gold ETF inflows: 2020 was the strongest year on record for gold ETF demand; the World Gold Council's full-year 2025 report describes 2025 as the second strongest, behind 2020.

The 2011 high finally broken

The August 2020 record took gold past its September 2011 peak of about $1,921 for the first time - roughly nine years later. But it did not stay there: by the end of the year gold had slipped back below $1,921, and it would not close a year above that level until 2023. A buyer at the 2011 top had, in effect, a lost decade.

What 2020 showed

2020 is a textbook example of the conditions that have historically been good for gold: a crisis, central banks cutting rates to near zero, real yields falling well below zero and a weaker dollar in the second half of the year. It also showed, in March, that gold can fall in the first, most chaotic phase of a crisis. Both lessons appear again and again in gold's history.

The three years after the pandemic record were a puzzle for anyone who thought they understood gold. Inflation surged to a 40-year high - which "should" have been good for gold - yet gold went almost nowhere for two years. Then, in 2023, it rose to a new record even though interest rates were at their highest in more than 15 years.

2021: inflation returns, gold slips

US inflation began rising sharply in 2021. But gold fell about 4-6% over the year (reported change -5.8%; year-end about $1,829). The reason was that markets expected the Federal Reserve to respond with higher interest rates, and investors moved money back into booming stock markets. The annual average was about $1,799.

2022: war, record inflation and the fastest rate hikes in decades

Russia's invasion of Ukraine in February 2022 sent gold briefly back above $2,000 in early March. Then the Federal Reserve began raising rates at the fastest pace since the early 1980s, and the US dollar climbed to its strongest level in about 20 years. Gold fell more than 20% from its March high to a low of about $1,620 on September 26, 2022 - its lowest since April 2020, according to Reuters reports carried by Mining.com and others.

Then something unusual happened. Gold recovered in the final months of the year even though the Fed kept raising rates. It finished 2022 at about $1,823, roughly flat for the year (reported +0.4%; our calculation from year-end prices is -0.3%). The annual average was about $1,800.

The main reason, in hindsight, was central banks. After Western countries froze a large part of Russia's foreign-currency reserves in 2022, many central banks - particularly in emerging markets - stepped up their gold buying. The World Gold Council reports that central bank net purchases reached a record of about 1,136 tonnes in 2022, and stayed above 1,000 tonnes in each of the following two years. Gold held in a country's own vaults cannot be frozen by another government, which made it newly attractive as a reserve asset.

2023: a record despite high rates

In March 2023 several US regional banks, including Silicon Valley Bank, failed, and gold rose sharply. In October, the war between Israel and Hamas added to geopolitical worry. By December 2023, gold had set a new all-time high above its 2020 record (commonly reported around $2,135 intraday on December 4, 2023), and it ended the year at about $2,063, up about 12.8%. The annual average was about $1,941.

That year-end close of $2,063 was also the first year-end close above the September 2011 peak of $1,921 - a recovery that took about 12 years.

Why 2021-2023 broke the old rules

The lesson is not that the old relationships stopped working, but that several forces act at once, and the balance between them shifts. Anyone using one simple rule - "inflation is coming, so buy gold" - would have been disappointed in 2021-2022 and surprised in 2023.

2024 was the year gold's rise moved from the financial pages to the front pages. It set record after record and finished with its best calendar year since 2010.

The year in numbers

Reuters reported on December 31, 2024 that gold's gain of more than 26% was its biggest annual rise since 2010. Note the small differences between sources: one Reuters story earlier in the autumn said gold was up 27% for the year at that point, and the final figure depends on the closing price used. We use "about 27%" for consistency with our other guides.

What drove 2024

• Central banks kept buying. The World Gold Council's figures show about 1,092 tonnes of net central bank purchases in 2024, the third year in a row above 1,000 tonnes.
• The Fed began cutting rates. The Federal Reserve made its first rate cut of the cycle in September 2024, lowering the "cost" of holding a non-paying asset.
• Strong buying in Asia. Investors in China in particular bought gold heavily as the country's property and stock markets struggled.
• Geopolitics. The wars in Ukraine and the Middle East, plus a contentious US election year, kept demand for "safe haven" assets high.

What was unusual

Gold rose strongly in 2024 even though US real interest rates were still positive and the US stock market also had a strong year. For decades, analysts had expected gold to struggle in that kind of environment. The fact that it did not was widely attributed to buying by central banks and Asian investors, who are generally less sensitive to US interest rates than Western ETF investors.

From the 2015 low to the end of 2024

Gold went from about $1,061 at the end of 2015 to about $2,617 (mid-point) at the end of 2024 - about 2.5 times ($2,617 / $1,061 = 2.47) in nine years, or roughly 10.5% a year. Over the same nine years, the price fell in two calendar years (2018 and 2021) and was roughly flat in a third (2022). The next year would dwarf all of this.

2025 was the strongest year for gold in more than four decades. The price rose about 65% from start to finish, set a record on 53 separate days, and pushed total gold demand above 5,000 tonnes for the first time.

The year in numbers

The figures in this table come from the World Gold Council's Gold Demand Trends: Full Year 2025 and its accompanying press release, published in early 2026, unless otherwise stated. The 5,002-tonne total includes over-the-counter demand, so it does not exactly equal the sum of the categories shown.

What drove 2025

• Investors joined central banks. In 2022-2024 central banks did much of the buying. In 2025, Western investors came back in force through ETFs, and bar and coin buying hit a 12-year high. Investment demand nearly doubled.
• Central banks kept buying, at a slower pace. Net purchases of about 863 tonnes were down from more than 1,000 tonnes in each of the previous three years, but still far above the levels of the 2010s. The WGC notes that much of this buying was not publicly reported at the time: a separate WGC tally of officially reported purchases came to only about 328 tonnes for 2025 (Kitco, Feb 4, 2026). Poland was the largest reported buyer, adding about 102 tonnes.
• Uncertainty. The WGC's own headline was "Gold investment rockets in 2025, setting a new high as uncertainty bites" - citing trade tensions, geopolitical risk and concerns about government debt and the dollar.
• Momentum. Rising prices attracted more buyers, which pushed prices higher still. That works in both directions, as early 2026 would show.

A milestone: the 1980 peak broken after inflation

In September 2025, Bloomberg reported that gold's rise to a then-record of about $3,674 had taken it above its January 1980 peak adjusted for inflation for the first time - about 45 years later. We look at inflation-adjusted prices in more detail in the real price section.

Why the average and the year-end differ so much

Notice that 2025's average price ($3,431) was far below its year-end price (about $4,316). That is because most of the rise came in the second half of the year. Someone who bought steadily through 2025 paid around $3,431 on average, not $2,617 - so their gain by year-end was about 26% ($4,316 / $3,431 - 1), not 65%. This is one reason headline yearly returns can be misleading: few real investors buy everything on January 1.

2026 began with gold's most dramatic move since 1980: a near-vertical rise to an all-time high in January, a violent one-day collapse, and a slide of nearly 30% over the following five months. At the time of writing in October 2026, gold sits roughly a quarter below its peak.

January: the record

Gold entered 2026 at about $4,310-$4,323 and kept climbing quickly. On January 29, 2026, spot gold reached an intraday record of about $5,595 (Reuters: $5,594.82). The LBMA AM price that day was $5,501.70 and the PM price $5,405.00. Silver hit a record of about $121.62 the same day (MetalCharts; we could not independently verify this figure).

From the end of 2025, that was a further gain of about 30% in four weeks ($5,595 / $4,316 - 1 = 0.296). From the end of 2023 ($2,063), the price had risen about 171% in just over two years.

January 30 - February 2: the break

The next day, gold fell sharply. Reports differ on exactly how far: Reuters-based reports described an intraday fall of close to 10% on January 30 - the sharpest one-day drop since 1983 - while other price records show spot around $5,183 at one point that day, about 3.9% lower on a closing-price basis. Silver fell far more, with reported one-day losses of roughly 27-36%.

Analysts quoted by BNN Bloomberg and others pointed to several triggers:

• A new Fed chair nominee. The White House's nomination of Kevin Warsh to lead the Federal Reserve was read by markets as a sign of tighter monetary policy. The dollar strengthened.
• Higher margin requirements. CME Group raised the margin (deposit) required to trade precious metals futures - reportedly from 6% to 8% for gold and from 11% to 15% for silver - which forced leveraged traders to sell, which pushed prices lower, which triggered more margin calls.
• Crowded positioning. After such a fast run, many buyers were sitting on large gains and rushed to lock them in. One analyst called it a classic "air pocket" after an extraordinary rally.

February to June: the slide

Gold bounced back in February - one less-established source reported it back above $5,100 by late February - but the recovery did not last. Through the spring, the backdrop changed in a way that has historically been bad for gold:

• Rate cuts turned into rate-hike expectations. By June, markets were pricing in rate increases from the Federal Reserve rather than cuts. CME FedWatch data cited by Reuters on June 25 showed a 66% chance of a September hike.
• The dollar strengthened to its highest level in more than 13 months.
• ETF investors sold. The World Gold Council recorded net outflows of about 16 tonnes from gold ETFs in May, continuing into June.
• Money rotated into stocks, especially AI-related shares.

On June 24, gold fell below $4,000 for the first time since November 2025. Reuters, via Kitco, put spot gold at about $3,982 on June 25, more than 28% below the January record. The LBMA price low in late June was about $3,994.50 - about 29% below the peak ($3,994.50 / $5,595 - 1 = -0.286). The 52-week low in spot trading was around $3,900.

July to October: stabilizing at a lower level

Since the June low, gold has partly recovered. The Federal Reserve did raise rates in September 2026, according to reports of a Goldman Sachs note that followed the decision. Gold futures settled at about $4,157 on October 5, 2026, and spot gold was about $4,170 on October 19, 2026 (Kitco).

2026 so far in numbers

For buyers at different points:

These figures ignore dealer markups, storage and account fees, all of which make the results worse for a real buyer - especially in a gold IRA, where an all-in markup of 5-15% is common. The gold IRA fee calculator shows how much the price must rise just to cover costs.

Demand stayed strong even as the price fell

One notable feature of 2026 is that demand did not collapse. Our research notes record total demand at a record of about 1,231 tonnes in Q1 2026, with central banks still buying. The price fall was driven mostly by Western investors and leveraged traders reacting to the change in interest-rate expectations - another reminder, explored in how much gold to hold, that gold's price in the short run depends on who is buying and selling at the margin, not on the long-term story.

This section collects the price data in one place. Use it to check claims you see in advertisements, or to see what gold did in any year that matters to you - the year you retired, for example.

Year-end prices and yearly changes, 2015-2025

Sources: year-end prices via JM Bullion; reported yearly changes from published annual returns; annual averages for 2019-2023 from our research notes, 2024 from the LBMA and 2025 from the World Gold Council. "n/a" means the figure is not in our verified data set. Calculated changes are our arithmetic - for example, 2017: $1,303 / $1,152 - 1 = +13.1%. Reported and calculated changes differ slightly because sources use different closing prices or benchmarks. These are the same figures used in our gold IRA investment guide.

Decade snapshots, 1971-2010

Five-year changes from those snapshots

Our arithmetic; for 1985 we used the mid-point of $327-$331. Because several of the older prices are approximate, treat the older percentage changes as rough. Of the ten five-year periods shown, gold rose in six, was roughly flat in one and fell in three. In the periods it fell, it fell by 25-44%.

How to use these tables

• Check the start and end dates of any return figure you are shown. The difference between 2010-2015 (-25%) and 2015-2020 (+79%) is entirely about when you looked.
• Remember these are prices, not your returns. A buyer paid more than spot and received less than spot when selling, plus any account fees.
• For the full daily history, the LBMA publishes its historical gold prices, and the World Gold Council's Goldhub has downloadable price data in many currencies.

A dollar in 1980 bought far more than a dollar today, so comparing $850 in 1980 with $5,595 in 2026 is not a fair fight. Adjusting for inflation changes the story considerably.

Estimates for the 1980 peak in today's money vary a lot. Analyses using annual-average CPI put it at about $3,325; those using the January 1980 monthly CPI reach about $3,590; and one series quoted by Eco3min puts it near $2,900 in 2026 dollars. Our own estimates in the table are rough and rounded; treat them as an order of magnitude.

Three things the real price shows

• The 1980 peak was not beaten in real terms until 2025. Bloomberg reported in September 2025 that gold at about $3,674 had passed its inflation-adjusted 1980 high for the first time - roughly 45 years later.
• The long-run real gain is much smaller than the headline. In dollars, gold is up about 119 times since 1971 ($4,170 / $35). After inflation, it is up roughly 15 times ($4,170 / ~$280).
• Gold has not reliably kept pace with inflation over every period. Over 1980-2001 it lost most of its real value. Over 1995-2025, Kiplinger puts gold's real return at about 5.4% a year - positive, but well behind stocks at about 8.4%.

Silver tends to move in the same direction as gold but with bigger swings, because it is a smaller market and about half of demand comes from industry. Its history has the same peaks as gold's, only sharper.

The gold/silver ratio - how many ounces of silver one ounce of gold buys - is about 72:1 today ($4,170 / $57.70 = 72.3). It has ranged from roughly 17:1 in January 1980 to just over 125:1 in March 2020, and MetalCharts puts the average of annual readings since 1971 at about 60:1. Silver's bigger swings cut both ways: it fell more than 50% from its 2026 peak, against about 29% for gold. Our silver IRA guide covers silver in retirement accounts.

No single factor explains gold's price. These six have mattered most since 1971, and they often pull in different directions at once.

Two practical points follow. First, whenever someone says "gold will rise because of X," history usually offers a period when X happened and gold fell. Second, the force that dominates changes over time: in the 1970s it was inflation and negative real rates; in 2022-2025 it was central banks; in the first half of 2026 it was rate expectations and leveraged selling. For a longer discussion of these drivers in an investment context, see our guide how to invest in gold for retirement.

Whether gold has "beaten" the stock market depends almost entirely on the dates you pick. These are the long-run figures we use across our guides:

All are before taxes, fees and dealer costs. Our own rough check on the longest period: $35 in August 1971 to about $4,170 in October 2026 is about 119 times over about 55 years, or roughly 9% a year - close to the Money Metals figure.

Why the start date matters so much

• Start in 2000 and you begin at the top of the dot-com stock bubble and near the bottom of gold's 20-year slump. Gold wins.
• Start in January 1980 and gold took about 28 years just to get back to its starting price, while stocks multiplied many times over. Stocks win by a mile.
• Start in 1971 and you capture every boom and bust for both. Stocks are ahead by about 2.4 points a year, which compounds into a large gap over decades.

The lesson: whenever a salesperson or commentator quotes a gold return, ask "from when to when?" Our gold IRA investment guide shows what $10,000 would have become in each over these periods, and our gold IRA vs gold ETF comparison shows how the way you hold gold changes your actual result.

A drawdown is the fall from a peak to the lowest point before a new peak. For someone living off savings, drawdowns and how long they last matter at least as much as average returns.

Our arithmetic. The gain needed is peak / low - 1: for example $1,921 / $1,050 - 1 = +83% and $5,595 / $3,994.50 - 1 = +40%. The 1980-1999 figure uses the exact low ($850 / $252.80 - 1 = +236%); our companion guide rounds the fall to 70% and shows +233%. Dates marked "approximate" are commonly cited but not confirmed in a primary data series.

What the table tells you

• Falls of 20-30% have come every few years, and falls of 45% or more have happened three times since 1974.
• Recovery has ranged from about 18 months to about 28 years in plain dollars - and longer after inflation.
• Losses hurt more than gains help. A 50% fall needs a 100% gain to recover. That is true of every asset.
• Selling during a drawdown locks in the loss. If you may need to raise cash from gold, read how to sell a gold IRA before you buy, so you know the steps and costs.

We do not forecast gold prices. Here is what major banks have published, with dates where we could find them. Most figures come from news reports of bank research rather than the banks' own documents, and several have already been revised.

How reliable are these?

Look at the revisions. Goldman's end-2026 forecast moved from $4,900 to $5,400 in January, back to $4,900 in June and to $4,650 in September. J.P. Morgan went from $6,000 to $4,500 in about four weeks. Wells Fargo started the year at $6,100-$6,300; gold is about $4,170. Earlier figures in our notes - Goldman $4,900 and BofA $5,000 - were themselves superseded. Few, if any, published forecasts at the start of 2025 anticipated a 65% rise, and few in January 2026 anticipated a 29% fall. Treat any forecast as one informed opinion, not a plan.

This is general education, not advice. But five lessons from the record above apply to almost anyone using gold in retirement savings:

1. Sequence risk is real. If you must sell gold for income or required minimum distributions just after a crash, you sell more ounces to raise the same cash, and those ounces cannot recover. At $1,921 (2011), $10,000 took about 5.2 oz; at $1,050 (2015), about 9.5 oz. Many planners keep several years of withdrawals in cash or short-term bonds so they are not forced to sell.
2. Size matters more than timing. A small slice can fall 45-70% without wrecking a plan; a large one cannot. Commentators commonly cite 5-10% for gold; see how much gold to hold in a retirement portfolio.
3. Rebalancing turns volatility to your advantage. Setting a target share and trimming after big rises (2020, 2025) or topping up after big falls is a disciplined alternative to guessing.
4. Costs come off every result above. A gold IRA buyer pays a dealer markup plus yearly custodian and storage fees. Check them with our fee calculator and read gold IRA fees explained.
5. Beware sales pitches timed to headlines. Records bring advertising and pressure. Read our gold IRA scams and red flags guide first.

A simple rebalancing illustration

Suppose a $200,000 portfolio held a 5% gold target ($10,000) at the end of 2024, with gold at about $2,617. By the end of 2025 (about $4,316), that gold would be worth about $16,500 ($10,000 x $4,316 / $2,617), before costs. If the rest of the portfolio were unchanged, gold would now be about 8% of a $206,500 total. Trimming back to 5% would have meant selling roughly $6,200 of gold near the year-end price - before the 2026 drop. The same rule would have meant buying after the 2013 crash. This is an illustration of how rebalancing works, not a recommendation or a prediction; real results depend on costs, taxes and what the rest of your portfolio does.

If you decide to proceed, our complete gold IRA guide, rollover guide and best gold coins for an IRA explain the steps, and how to sell a gold IRA covers the exit. A fee-only adviser can help you decide whether gold fits at all.

For an IRA, also check the value your custodian reports on statements; it will usually use a recognized benchmark, while your actual sale price will be the dealer's bid. Checking daily is rarely helpful for a long-term holding.

About $5,595 an ounce intraday spot on January 29, 2026 (Reuters: $5,594.82). The LBMA PM price that day was $5,405.00, and some services list a record daily close of about $5,318. Adjusted for inflation, it is also the highest real price on record. On October 19, 2026, gold was about $4,170, roughly 25% below that peak.

The official US price was $35 an ounce until President Nixon closed the gold window on August 15, 1971. The free-market London price was a little higher, and Federal Reserve data put it at about $43.48 by December 1971. $35 in 1971 is roughly $275-$285 in today's money by our CPI estimate.

From about $1,061 at the end of 2015 to about $4,316 at the end of 2025, gold roughly quadrupled - about 4.07 times, or about 15% a year. From the end of 2016 ($1,152) to October 19, 2026 (~$4,170) it rose about 3.6 times, or about 14% a year. That decade was unusually strong: it began near the bottom of a 45% slide and ended near a record. Gold's long-run average since 1971 is about 8.8% a year. Past performance does not predict future results.

Mainly because expectations for interest rates flipped. Markets moved from expecting Fed rate cuts to expecting hikes, the dollar rose to a 13-month high, ETF investors sold (about 16 tonnes of outflows in May) and money rotated into stocks. The initial January 30 plunge was worsened by the Kevin Warsh Fed nomination, higher CME margin requirements and crowded positioning after a ~30% rise in four weeks. Gold fell about 29% to roughly $3,994 by late June.

No. Gold rose over the 1973-1975 and 2007-2009 recessions and in 2020, but it fell sharply during the worst weeks of late 2008 and March 2020 as investors sold assets to raise cash, and it fell during the 1981-1982 recession as real interest rates soared. Its record in crises is mixed, and it depends heavily on what interest rates and the dollar are doing at the time.

Nobody knows. Published bank forecasts for 2027 range widely: Goldman Sachs $5,400 by end-2027 (September 2026), UBS $5,400 by September 2027 (unverified), and one report of a Bank of America 2027 average of $4,813 (unverified). J.P. Morgan had suggested $6,300 was possible in 2027 before cutting its 2026 targets sharply. These banks revised their 2026 forecasts by $500 to $1,500 within months, so treat any 2027 figure as a rough opinion. We make no forecast of our own.

Since the dollar's link to gold ended in 1971, gold has risen from $35 to about $4,170 - roughly 119 times in dollars and about 15 times after inflation. But the path was anything but smooth: a 47% fall in the mid-1970s, a 70% collapse that took 28 years to recover, a 45% slide after 2011 and a 29% drop within months of the January 2026 record.

• Gold has had long stretches of beating stocks and longer stretches of trailing them; the start date decides the winner.
• Real interest rates, the dollar, central banks and investor flows have mattered more than any single headline.
• Forecasts, including those from major banks, are revised often and should not be the basis for a retirement decision.
• For a retirement saver, position size, costs and the ability to avoid forced sales matter more than predicting the next move.

If you are weighing gold for retirement, start with is a gold IRA a good investment?, see how we assess providers on how we rank, or return to our homepage for our full set of guides. Past performance does not guarantee future results.

This article is general information, not financial, tax or legal advice. Fees, minimums and promotions change often and vary by custodian and account size - confirm current terms in writing with the company and your custodian before you invest.