How much gold does the stock market cost?

Since 1925, the S&P 500 has risen 605-fold in dollars. Measured in gold, it has tripled. Today the index costs 58 grams of gold, just over a third of what it cost in 1999.

By , CEO, Goobit Group AB (publ) · Published 6 November 2026 · Data through 30 October 2026 · 12 min read

Also available as a video

×605 / ×3

S&P 500 since 1925, in dollars and measured in gold. Price only.

161 → 58 g

grams of gold for one unit of the S&P 500, at year-end 1999/2000 and today

+13.3 / −1.5

% per year since 2020, in dollars and measured in gold

17 of 17

crashes have been recovered in dollars. Measured in gold, the 1999 peak has still not been reached.

Two weeks ago, I went through 155 years of the stock market, in dollars and in gold, directly in the charts. I received a lot of positive feedback, but also heard from many people who found it too detailed and struggled to follow. So here is the same comparison again, zoomed out and brought together in simpler pictures.

This is about the S&P 500, the 500 largest companies in the US. I think it is the best index for measuring the stock market over long periods, and researchers have reconstructed it all the way back to the nineteenth century. This time I start in 1925, between the world wars, when countries were trying to return to gold after the First World War.

Before we begin: I am not saying that shares or the stock market are bad. I am building a company myself, and the stock market is an excellent way to build and own companies. What I want to offer is a perspective I think many people miss. What you see depends on what you measure with, and gold can be an alternative for preserving value over time.

The uncomfortable part first

  • Price only. All figures show the price without dividends. With reinvested dividends, the S&P 500 has risen roughly 100-fold measured in gold since 1925, not 3-fold.
  • Stocks have won too. Between 1981 and 2020, when interest rates fell for almost 40 years, stocks beat gold by 5.0 % per year.
  • My conflict of interest. I am CEO of Goobit Group, which sells gold and bitcoin, and I own shares in the company. This is my analysis, not advice.

We are measuring the stock market. Gold is the ruler.

We are used to measuring the stock market in a currency. The S&P 500 is measured in dollars, and here in Sweden we measure in kronor. But you can also ask how much gold you get for one unit of the index. Or, put another way: what would it have looked like if I had held gold instead of being invested in the 500 largest companies in the US?

The same stock market, two price tags

Levels on 14 October 2026. Price in gold = S&P 500 divided by the gold price in dollars per ounce.

7,722 dollars

1.85 ounces = 58 grams of gold

How to read this: −50 % in gold means one unit of the S&P 500 buys half as much gold as at the peak.

Figure 1. The same stock market, two price tags, on 14 October 2026.

Past performance is no guarantee of future returns

The price in gold is the S&P 500 divided by the gold price in dollars per ounce. An ounce is 31.1 grams, and since grams are easier to grasp, I show both. When I say the stock market fell 50 % measured in gold, it means one unit of the S&P 500 buys half as much gold at the bottom as at the peak.

The ruler itself has had three settings. Until 1933, an ounce of gold was 20.67 dollars. That was not a price, it was the definition of a dollar. In 1934, the government decided the same ounce was 35 dollars, and it stayed fixed there until 1971. Since then, the price has floated.

One ounce of gold (31.1 grams) in dollars, 1925–2026

In 1934, the dollar’s gold value was reduced by 41 %. Since 1971, the dollar has lost 99 % measured in gold.

Logarithmic scale, annual values. Gold: official price until 1959, World Bank (December average) from 1960, 14 Oct 2026.

1925–1933: fixed price 20.67 dollars

1934–1971: fixed price 35 dollars

Oct 2026: floating price 4,169 dollars

Figure 2. The gold price in dollars per ounce, 1925–2026.

Past performance is no guarantee of future returns

Between 1934 and 1971, measuring in gold therefore effectively measures only the dollar. A yardstick someone holds fixed measures nothing.

It is always the stock market moving in these figures. All I change is the yardstick, and until 1971 that yardstick was fixed by the government.

Since 1971, everything has grown when measured in dollars

On 15 August 1971, President Nixon ended the ability of other countries’ central banks to exchange dollars for gold. From that day, the dollar was tied to nothing. Gold continued to be valued for what it is, a limited asset that many trust to hold its value. Meanwhile, the US kept increasing the quantity of dollars, and eventually a limited asset could no longer be held together with a currency that only grows.

Since 1971, everything has grown when measured in dollars

Multiples in dollars, logarithmic scale. Consumer prices, S&P 500 and gold: year-end 1970/71 to 14 Oct 2026. M2: Jan 1971 to Sep 2026. Sources: BLS (CPI), Damodaran (S&P 500, price only), World Bank (gold), Fed H.6 (M2).

S&P 500 measured in gold −26 %. One unit of the S&P 500 buys a quarter less gold today than in 1971.

Figure 3. How many times everything has grown since 1971, measured in dollars.

Past performance is no guarantee of future returns

Looking at the gold price after 1971, it is hard to reach any other conclusion than that something has happened to the dollar. The money supply has grown 37-fold, consumer prices 8.5-fold, the S&P 500 84-fold and gold 113-fold. When there are so many more dollars, everything scarce rises when measured in dollars. Shares too.

Measured in gold, one unit of the S&P 500 has instead lost 26 % since 1971, counting price only. With dividends, it has more than tripled even in gold, and that needs to be said.

A large part of the stock market’s rise in dollars since 1971 reflects the increase in the number of dollars, not just companies becoming more valuable.

Four phases, in dollars and in gold

When I put 155 years of the stock market side by side in the previous analysis, I saw the same four phases returning again and again. First, the market moves sideways, often for many years. Then it breaks out and trends. At the end of the trend comes a bubble, where the rise is faster than during the entire preceding period. Then comes a major correction, and afterwards the phases begin again.

Four phases that keep returning

Schematic illustration, not data. The phases are my judgements in the charts in analysis 4.

Consolidation

Sideways, often for many years.

Trend

Breaks out and rises for a long time.

Bubble

Faster than anything before.

Correction

Plunges, often taking ten years to heal.

155 years of the stock market: The same four phases appear in dollars and in gold. The difference is depth: declines become much larger in gold.

Figure 4. The four phases. A sketch of the pattern, not data.

Past performance is no guarantee of future returns

Plotting the phases on the S&P 500 in dollars since 1971, it starts with a major correction shortly after 1971, the oil crisis. Then the market moves sideways until 1982, trends, and has a fairly small correction in the 1987 crash. Then the same thing again: consolidation, trend and the dot-com bubble.

Next comes a strange phase with a double crash, 2000–2002 and 2007–2009, without the index rising much higher in between. Since 2009 there has been a long trend, with covid becoming only a brief dip. Every major peak has been higher than the previous one, and every decline since 1971 has been recovered, taking at most seven and a half years.

S&P 500 in dollars: short declines, new peaks

Month-end Jan 1971 – Sep 2026, logarithmic scale. Phases are my judgement. Data: TradingView, own calculations.

1. Consolidation2. Trend3. Bubble4. Correction

Major peaks in dollars: 118 → 330 → 1,518 → 4,766 → 7,686. All declines recovered within 7.5 years. Since 1971: ×80 in dollars.

Figure 5. S&P 500 in dollars, month by month since 1971, coloured by phase.

Past performance is no guarantee of future returns

Many argue that there have been no really big crashes since the 2008 financial crisis, just one long rise. In dollars, that is true. But measured in gold, the same period begins with an enormous crash. From 1971 to 1980, stocks fell 93 % measured in gold, from 2.60 to 0.17 ounces, lasting almost ten years. Had you held gold instead, you could have bought more and more of the stock market throughout that period.

Then consolidation, a trend, a smaller crash in 1987, a long trend and the dot-com bubble. But what looked like two crashes in dollars, 2000 and 2008, was one long eleven-year crash measured in gold: from 5.47 to 0.67 ounces, −88 %. After a new trend in 2012–2021 came another correction, from 2.61 to 1.30 ounces by February this year, and stocks remain below their 2021 peak.

S&P 500 in gold: the same phases, deeper declines

S&P 500 divided by the gold price, month-end Jan 1971 – Sep 2026, logarithmic scale. Phases are my judgement. Data: TradingView, own calculations.

1. Consolidation2. Trend3. Bubble4. Correction

Major bottoms in gold: 0.17 → 0.67 → 1.30 ounces. 5 → 21 → 41 grams. But the 2021 peak was half the 2000 peak.

Figure 6. The same stock market and the same period, measured in gold.

Past performance is no guarantee of future returns

Two things are worth noticing. Every major bottom has so far been higher than the last: 5 grams of gold in 1980, 21 grams in 2011 and 41 grams in February this year. And the 2021 peak was only half the peak in 2000.

This is how the stock market worked before 1971, when the dollar was tied to gold. And this is how I think we should view the stock market today too.

In dollars, stocks look like a staircase where every fall is recovered. In gold, they move in waves, with declines of 50–93 %.

The same stock market, two completely different pictures

Here is the whole point in one picture. At the top, the S&P 500 in dollars: a staircase rising 605-fold since 1925. Below, exactly the same index measured in gold: no trend at all, just large waves.

The same stock market, two completely different pictures

Logarithmic scale, year-end 1925 – 14 Oct 2026, price only. S&P 500: Damodaran. Gold: official price and World Bank.

×605 in dollars since 1925 · ×3 measured in gold since 1925 · ×100 in gold with dividends

Figure 7. S&P 500 in dollars and measured in gold, year-end 1925 – 14 Oct 2026.

Past performance is no guarantee of future returns

Measured in gold, stocks have not even returned to their 1999 level. One unit of the S&P 500 today buys 64 % less gold than at year-end 1999/2000. It buys roughly as much as in the early 1960s, before gold was released from the dollar.

You can make of that what you will, but it gives a completely different picture from the one we usually see of how stocks perform.

Both pictures are true. The upper one shows how many dollars the stock market is worth. The lower one shows how much of a limited asset it can buy.

From 7 to 161 grams of gold

How much gold does the stock market cost?

Grams of gold per S&P 500 (1 ounce = 31.1 grams), year-end. Today = 14 Oct 2026. S&P 500: Damodaran. Gold: official price and World Bank.

1979: 7.4 grams. Less gold than at the bottom of the 1930s.

1999: 161 grams. 22 times more, twenty years later.

Today: 58 grams.

Figure 8. Grams of gold for one unit of the S&P 500 at peaks and bottoms.

Past performance is no guarantee of future returns

In 1928, the stock market cost 37 grams of gold. At the bottom in 1932 it cost 10 grams, and in 1967 almost 90 grams. Then, in 1979, only 7.4 grams. That is less than at any year-end since 1925, even during the Depression. Twenty years later, in 1999, the same index cost 161 grams, 22 times as much.

The peaks in 1928, 1967 and 1999 grew higher and higher. But since 1999, stocks have not done better than gold. The 2021 peak, 83 grams, reached neither 1967 nor 1999. Today one unit costs 58 grams.

The stock market’s price in gold has swung more than twenty-fold in twenty years, in both directions. Anyone who bought at year-end 1999/2000 gets just over a third as much gold for their unit today.

No one wins forever

If you look only at who wins, stocks or gold, you see a pendulum. Between 1932 and 1967, stocks were 8.2 times better than gold. But gold was fixed against the dollar then, so we are effectively measuring only the dollar. Since 1967, the winner has changed every ten to twenty years.

No one wins forever

S&P 500 divided by the gold price, year-end between turning points. ×22 = stocks bought 22 times more gold in 1999 than in 1979.

Stocks beat gold / Gold beat stocks. Except when the gold price was fixed, the winner has changed every 10–20 years.

Figure 9. Who won between the turning points, stocks or gold.

Past performance is no guarantee of future returns

Gold won 11.6-fold in 1967–1979. Stocks won 22-fold in 1979–1999, and I say that openly. Then gold again until 2011, stocks again until 2021, and since 2021 we have been in a gold phase.

The swings are getting smaller and smaller in both directions: 11.6 and 22 times, then 6.8 and 3.5. To me, that suggests it is becoming harder and harder to achieve outperformance in either asset. I think this is because the world and monetary policy have become so complicated, and we have printed a great deal of money to keep the economy going. This has been happening since 1971, but in extreme quantities since the 2000s. The current gold phase is not over, so it says nothing about its size yet.

Except when the gold price was fixed, the winner has changed every 10–20 years. Anyone drawing conclusions from a single period can get any answer they want.

When the government took the gold

During the Depression in the 1930s, many countries ran into problems with their currencies and economies. Britain and Sweden left the gold standard in 1931 and let their currencies float freely. That was precisely what the US later did in 1971.

The US chose a different path. In April 1933, all private individuals were forced to hand their gold to the government at the fixed price of 20.67 dollars per ounce. Anyone who did not comply risked heavy fines. Nine months after the final surrender date, on 31 January 1934, the government decided the same ounce was worth 35 dollars.

When the government took the gold

Source: Federal Reserve History, Executive Order 6102 (1933) and Gold Reserve Act (1934).

20.67 dollars per ounce

Private gold ownership is banned. All gold must be handed to the government.

35 dollars per ounce

The same ounce, 31 grams of gold. The government sets a new price.

41 % less gold per dollar.

1971 was not the first time the dollar let go of gold. For ordinary Americans, it happened in 1933.

Figure 10. The gold ban in 1933 and the new price in 1934.

Past performance is no guarantee of future returns

The revaluation made the government’s gold worth much more in dollars without increasing its debts. The government’s balance sheet looked stable again, and from its perspective the crisis was solved.

Let me be blunt: the American people paid for this, receiving 41 % less for their gold than the government shortly afterwards said it was worth. And it has become a pattern. Sweden and Britain in 1931, the US in 1933 and the US again in 1971. When a government has serious economic problems, it is the value of money that changes.

1971 was not the first time the dollar let go of gold. For ordinary Americans, that had already happened in 1933.

The same crisis, two depths

As long as the dollar was tied to gold, the two measures gave the same answer: 1937, 1962 and 1966 have the same figure in dollars and in gold. After 15 August 1971, they diverge. The 1973–1974 oil crisis was −48 % in dollars but −78 % in gold. And it goes both ways: in 1980–1982 stocks fell in dollars but rose in gold, because gold fell even more. Gold is a moving yardstick too.

The same crash, two answers

Declines in dollars and in gold over exactly the same period, daily close. Data: TradingView (SPX, GOLD), own calculations.

The same crash, two answers
CrashS&P 500 in dollarsS&P 500 in gold
1937–1938−54.5 %−54.5 %
1961–1962−28.0 %−27.9 %
1966−22.2 %−22.2 %
15 August 1971: The dollar lets go of gold. The yardsticks diverge.
1973–1974−48.2 %−78.1 %
1980–1982−27.1 %+35.9 %
2007–2009−56.8 %−66.2 %
2020−33.9 %−29.4 %
2022−25.4 %−18.1 %

Before 1971, the gold price was fixed. Both measures gave the same answer.

After 1971, they diverge in both directions.

Figure 11. The same crash, measured in dollars and in gold over exactly the same period.

Past performance is no guarantee of future returns

Starting with the worst crashes measured in dollars, the 1930s top the list, the one still called the worst crash of all time: −86 % in three years. Second is the 2007–2009 financial crisis at −57 %. All 17 crashes of at least 20 % since 1928 have returned to new records in dollars.

The ten largest declines, measured in dollars

Daily close 1928 – 30 Oct 2026. Crash = decline of at least 20 %, ending at a new peak or +50 % from the bottom. Data: TradingView, own calculations.

17 of 17 crashes since 1928 have recovered to the peak in dollars. The longest took 25 years, after 1929.

Figure 12. The ten largest declines in the S&P 500, measured in dollars.

Past performance is no guarantee of future returns

Measured in gold, the list looks completely different. The largest is not the Depression but 1967–1974, when stocks fell 87.6 % measured in gold but only 31 % in dollars. Second is the dot-com bubble and financial crisis, 1999–2009: −87 % measured in gold against −52 % in dollars.

Third is 1976–1980. In dollars, stocks rose 7 %, but measured in gold they fell 87 %, because the gold price rose so much. That crash was not even visible, because it was measured as a rise in dollars. Can we call it a crash? I think this is one of the most interesting questions in the entire analysis. And in 2022–2026, the years after Russia’s currency reserves were frozen, stocks rose 42 % in dollars but fell 52 % in gold.

The reason the 1971–1980 decline is one fall in figure 6 but two here is the rule I use: a crash is considered over when stocks have risen 50 % from the bottom, and that happened measured in gold between 1974 and 1976.

The ten largest declines, measured in gold

S&P 500 divided by the gold price, daily close. Gold 20.67 dollars until 1933, 35 dollars in 1934–54, then market price; 1933–35 estimated. Data: TradingView, own calculations.

3 of 10: During three of the ten largest declines in gold, stocks rose in dollars. Four declines were greater than 85 %.

Figure 13. The ten largest declines in the S&P 500, measured in gold.

Past performance is no guarantee of future returns

Then there is the time it takes to get back to the peak. In dollars, it has become faster and faster: 25 years after 1929, 5.5 years after the financial crisis and half a year after covid. Measured in gold, it took 29 years after 1929 and 30 years after 1967. The July 1999 peak has still not been reached. 27 years later, stocks are 66 % below it.

How long does it take to reach the peak again?

From the peak to the day it was reached again, daily close to 30 Oct 2026. Data: TradingView, own calculations.

Not back yet: The July 1999 peak in gold has still not been reached. 27 years later, stocks are 66 % below it.

Figure 14. From the peak to the day the peak was reached again.

Past performance is no guarantee of future returns

In dollars, stocks have always recovered, and increasingly quickly. Measured in gold, it has taken a generation, and from the 1999 peak they have not recovered at all.

Only when rates fell did stocks clearly beat gold

Dividing history according to how interest rates moved reveals something interesting. In dollars, stocks have risen faster in every period. Measured in gold, they have delivered negative returns in three of four periods. During those periods, it would have been better to own gold than to have money in stocks.

S&P 500 per year, in dollars and in gold

Per cent per year, price only, year-end. 2020–2026 = to 14 Oct 2026. S&P 500: Damodaran. Gold: official price and World Bank.

Only when rates fell for 40 years did stocks clearly beat gold.

Since 2020: +13.3 % per year in dollars, −1.5 % per year in gold.

Figure 15. S&P 500 per year in different interest-rate periods, in dollars and measured in gold.

Past performance is no guarantee of future returns

The exception is 1981–2020, when rates fell for almost 40 years. Then stocks beat gold by 5 % per year. I say that openly, because it is part of the truth. But notice what happened when rates turned upwards in 2020. In dollars, stocks have never risen faster. Measured in gold, they have fallen.

The figures are calculated using annual data and differ slightly from analysis 4, where I measured directly in the charts using different start and end points.

Higher and higher returns in dollars, but in gold the same negative returns as in the periods before 1981. That suggests much of the difference is the dollar, not the companies.

The same cycles, only faster

Measuring bitcoin in gold reveals the same cycles as in the S&P 500: trend, bubble, correction and consolidation, again and again. The difference is that the cycles move much faster. I think this is because bitcoin is a much younger asset. It has existed for almost 18 years.

Four cycles in fourteen years

Bitcoin divided by the gold price, month-end Jan 2012 – Sep 2026, logarithmic scale. Phases are my judgement. Data: TradingView, own calculations.

1. Consolidation2. Trend3. Bubble4. Correction

Every bottom higher than the last: 0.17 → 2.6 → 9.1 → 12.7 ounces

5 → 80 → 282 → 395 grams of gold

But peaks are flattening: 34 → 37 ounces.

Figure 16. Bitcoin measured in gold, month by month since 2012, coloured by phase.

Past performance is no guarantee of future returns

Four cycles in fourteen years, with peaks in 2013, 2017, 2021 and around year-end 2024/25. Every bottom has been higher than the last, from 5 grams of gold per bitcoin in 2015 to almost 400 grams in February this year. But the peaks are flattening: 34 ounces in 2021 and 37 in 2025. Declines have become smaller with each cycle, and rises too.

Now put stocks and bitcoin side by side, both measured in gold. From 2012 to 2021, when stocks rose against gold, bitcoin rose even more. Since December 2021, when stocks have fallen against gold, bitcoin has fallen by roughly the same amount.

Bitcoin has followed stocks, not gold

Both measured in gold, month-end Jan 2012 – Sep 2026, logarithmic scale. Stock market peak against gold: Dec 2021. Data: TradingView, own calculations.

2012–2021: Stocks ×3.4 Bitcoin ×8,000

Dec 2021 – Sep 2026: Stocks −32 % Bitcoin −30 %

Figure 17. S&P 500 and bitcoin, both measured in gold.

Past performance is no guarantee of future returns

Bitcoin has followed stocks, not gold. I think that is the biggest reason bitcoin has not done particularly well in recent years: its price has moved more with stocks than with gold.

The uncomfortable part: today one bitcoin costs 20 ounces of gold, just over 600 grams. That is less than half the peak.

The same cycles, only faster

Bitcoin measured in gold, two-week bars, my annotations in analysis 4, as of 23 Oct 2026. Intramonth peaks and bottoms are included, hence larger declines than in the monthly chart.

+8 991 % · 2013

+3 448 % · 2017

+41 % · since the bottom

−84 % · 2018

−74 % · 2021–22

−69 % · after the 2024/25 peak

Today one bitcoin costs 20 ounces of gold, just over 600 grams, less than half the peak.

My assessment, not data: So far bitcoin has followed stocks more than gold. I still think it has much in its favour over the coming 5–20 years.

Figure 18. Bitcoin’s largest rises and declines, measured in gold.

Past performance is no guarantee of future returns

Bitcoin solves two problems, gold solves one. Gold’s biggest problem has always been that it is hard to pay with. Its strength, really its only function, is preserving value. Bitcoin can do both: it works for payments in a digital world and has a limited supply, just like gold. That is why I think bitcoin has a great deal in its favour over the coming 5–20 years.

But bitcoin has fallen more than 70 % four times since 2009 measured in dollars, and it does not yet have a single thirty-year period to measure it against. I see bitcoin as money for generations, not for years.

Anyone who bought bitcoin as protection against stocks falling against gold has not received that protection yet.

Two truths about the same stock market

Two truths about the same stock market

Year-end 1925 – 14 Oct 2026, price only. With dividends: ×100 in gold since 1925.

×605

since 1925. All 17 crashes have been recovered. Since 2020: +13.3 % per year.

×3

since 1925, without dividends. The 1999 peak has not been reached again. Since 2020: −1.5 % per year.

Shares are ownership with risk, and that risk should be rewarded. But if money held its value, you would not have to become an investor just to stand still.

Figure 19. Two truths about the same stock market.

Past performance is no guarantee of future returns

1

In dollars the S&P 500 has risen 605-fold since 1925. All 17 crashes have been recovered, and since 2020 it has risen faster than ever, +13.3 % per year.

2

Measured in gold the same index has tripled in a hundred years, without dividends. The 1999 peak has not been reached again, and since 2020 stocks have fallen 1.5 % per year. I think the difference between the pictures is largely the money supply.

3

No one wins forever. The winner between stocks and gold has changed every 10–20 years, and bitcoin has so far followed stocks more than gold.

Shares are ownership with risk, and you should be paid for that risk. You have been, even measured in gold, if you include dividends. The question is whether the reward has been worth the risk, and everyone must judge that for themselves.

But I do not think you should have to take risk to save your money. Money should be a way to save your time, energy and focus so you can use them in the future. That is the purpose of money. It should also be possible to pay with it, and both things should be true. For the past fifty to a hundred years, we have mostly lived with the second truth: money is what we pay with, and then everyone has to do something with their money for it to retain its value.

I would argue that it is the other way round. Something becomes money because people trust it to hold its value over time. That is why it then becomes a means of payment, not the reverse.

We are not there today; we are closer to the opposite. Most people have never been so active with their investments, because it has become accepted that money which is not invested loses value. I think that is a sad development. If money itself held its value, you would not have to become an investor just to stand still. My assessment is that we may be moving towards a world where money can be money again.

Please write what you see in the numbers, and challenge me.

Declaration of interests and risk. Futureproof is a service from Goobit Group, which sells gold and bitcoin. I am CEO of Goobit Group and own shares in the company. This is my analysis and not personal investment advice. Past performance is no guarantee of future results, and your savings can lose value.

Past performance is no guarantee of future returns

Sources and method

  • The figures are based on slides from the video accompanying this analysis.
  • Annual data (figures 1–3, 7–9, 15 and 19, except the number of crashes): S&P 500 price only, year-end 1925–2025, from Aswath Damodaran, NYU Stern, Historical Returns on Stocks, Bonds and Bills; the level is reconstructed from annual price returns with 31 Dec 1999 = 1,469.25. Gold: official price until 1959 and the World Bank (Commodity Price Data, December average) from 1960. Last point 14 Oct 2026: S&P 500 7,721.54 and gold 4,169 dollars per ounce. With dividends: the same source’s total return. Consumer prices: BLS (CPI-U). Money supply M2: Federal Reserve H.6.
  • Monthly and daily data (figures 5, 6, 11–14, 16 and 17, and the number of crashes in figure 19): S&P 500, gold price and bitcoin from TradingView, own calculations. Month-end Jan 1971 and Jan 2012 respectively to Sep 2026; daily close 1928 to 30 Oct 2026. Crash = decline of at least 20 % from a peak, ending at a new peak or +50 % from the bottom. Gold is counted as 20.67 dollars until April 1933 and 35 dollars in 1934–1954, then market price; the 1933–1935 decline measured in gold is estimated.
  • Figure 18: my annotations in TradingView charts using two-week bars, as of 23 Oct 2026 (the same as in analysis 4, The stock market in dollars and the stock market in gold).
  • History: Federal Reserve History (Executive Order 6102, 1933; Gold Reserve Act, 1934; Nixon’s decision on 15 Aug 1971) and Sveriges Riksbank (leaving the gold standard in 1931).

Uncertainties

  • Annual, monthly and daily data give different figures for the same decline. Within months and years, both peaks and declines have been greater.
  • The gold price was fixed before 1971. Measuring in gold then effectively measures only the dollar.
  • Everything shows price without dividends. With reinvested dividends, shares look significantly better against gold.
  • Where a phase or interest-rate period begins and ends is my judgement, and the rule for what counts as a crash determines how many crashes there are.
  • The latest bottom measured in gold, in February 2026, is not yet confirmed. If it breaks, the pattern of rising bottoms disappears.

Be there from the start

The service has not launched yet. Join the waitlist and we will be in touch when we open.

Investing involves risk. Your savings can go both up and down in value.