The history of fiat currencies

Fiat means “let it be”. Money is valid because someone has decided so.

Imagine a village of a hundred people. I am good at picking apples and hand them out to everyone. You are good at building houses. We need no money, because everyone knows who contributes what. But as soon as the economy grows beyond everyone knowing everyone, we need something to store our work in, something that holds its value until you want to trade it for something else. That is what money is. A way to save time, energy and focus for later.

Fiat is Latin and means “let it be”, as in “let there be light”. It shall be so, because someone has said it. A fiat currency is money that cannot be redeemed for anything. The note in your wallet and the figures in your salary account are not receipts for gold. They are worth something because the law says they must be accepted, because taxes must be paid in them and because everyone else uses them.

Money has gone through three stages.

In a fiat system new money is created in two ways. The central bank creates notes and the banks' reserves. But most of it arises when ordinary banks lend. The loan is deposited as a new balance in someone's account. Nobody needs to dig anything up first.

That makes the system flexible in a crisis. But it also means the new money always reaches someone first. In my view this is the most important and least discussed part. The money flows from the central bank to the state, on to the largest companies and the most creditworthy borrowers, and last to you. By the time it reaches you, asset prices have already risen. If I had been a dictator I would have loved this system.

And the money is supposed to lose value. The Riksbank, the Fed and the ECB all aim for two percent inflation per year. Two percent sounds small. But it means the money loses half its purchasing power in 35 years and 86 percent in a hundred years. If everything goes according to plan.

Sweden gave Europe the banknote. Three years later it collapsed.

The history of fiat money actually begins in Stockholm. And the first lesson came immediately.

35 years when the krona held its value.

Between 1879 and 1913 both the krona and the dollar could be redeemed for gold at a fixed price. What happened to prices during those years? In Sweden they rose by 0.5 percent a year. In the US by 0.2 percent. Anyone who put a ten-krona note in a drawer in 1879 could still buy more than 80 percent of the same goods with it 35 years later.

That was not because prices stood still. They fell at first, as the world economy grew faster than the gold supply, and rose again when new gold fields in South Africa and the Klondike began producing. But they came back. That is the big difference from today. Under the gold standard there was an anchor line. Prices could move, but they were pulled back.

And saving worked. Treasury bills returned 4.3 percent a year above inflation. The Stockholm Stock Exchange with dividends returned 7.1 percent. Swedish companies built railways, sawmills and factories with money that did not lose value in the meantime.

The krona has lost 98 percent. Almost all of it after 1914.

So what has happened since? A krona from 1879 buys today what 1.6 öre bought then. A dollar buys 2.9 cents. Prices in Sweden have risen 64-fold. The one who takes the hit is the one who receives their salary and keeps their savings in the currency.

Look at the shape. Until 1914 the curve moves sideways. Then the First World War arrives, gold is switched off and prices almost triple in six years. After the war Sweden returns to gold, and the krona recovers part of the loss. When gold is abandoned in 1931 a long decline begins that never turns back.

What surprised me most was how much disappeared before 1971. When Nixon closed the gold window the krona had already lost 85 percent since 1879. Bretton Woods tied the dollar to gold for central banks, not for you and me. Since 1971 the krona has lost a further 90 percent of what was left.

So let us ask the question plainly. If you received your salary in kronor and saved it, has it ever held its value? I looked at every conceivable saving period since the krona became fiat in 1931.

I recalculated River's chart. The figures hold up.

I recalculated purchasing power since 1971 using the World Bank's inflation data for 21 currencies. The figures match within one percentage point. The Swiss franc, the world's most stable fiat currency, buys a third of what it did in 1971. The dollar buys an eighth. The krona a tenth.

I followed one krona from 1879. Here is what it became.

But almost nobody keeps their money in a drawer. So the real question is what would have happened if you had done something with the krona. The chart follows one krona from the end of 1878 in six different forms of saving, with the Riksbank's historical series up to 2012 and official statistics thereafter. Switch to “real” to see everything measured in what the money actually buys.

Three things stand out.

The first is the savings account, here treasury bills. In Sweden they have returned 1.9 percent a year above inflation over 147 years. That is more than I thought. But almost all of it comes from the period before 1932 and from the high interest rates of the 1980s and 90s. Between 1932 and 1970 they returned zero above inflation, before tax. With tax, the savings account has been a guaranteed loss for long periods. That is the saver's real fiat tax, and it never shows on the account statement.

The second is gold. In kronor it has returned 1.5 percent a year above inflation. It has preserved purchasing power, which no fiat currency has managed. But it is also the lowest return of anything except cash.

The third is the stock market. With dividends reinvested, the Stockholm Stock Exchange has returned 6.4 percent a year above inflation. One krona became 575,000 kronor. Purchasing power grew almost 9,000-fold. Without the dividends it is 2.4 percent a year. More than 60 percent of the stock market's return above inflation over 147 years is dividends bought back in.

The same pattern in dollars. The S&P 500 with dividends buys 100 times more gold than in 1925.

The American stock market has longer and denser data from 1926. There the picture is the same, only clearer. The savings account has returned 0.4 percent a year above inflation, gold 2.4 percent and the S&P 500 with dividends 7.2 percent. Without the dividends, 3.5 percent.

Gold was pinned down for 37 years. Then it made up for everything in nine.

In my view gold is the true yardstick for money. It happens to be the only commodity that can hardly be destroyed, that costs enormous amounts of energy to produce more of and that has never found a more important role than preserving value. But a yardstick that someone holds still measures nothing.

147 years contain four monetary systems. The gold standard. The war years and the brief return to gold. A fixed exchange rate without gold, when the gold price was locked at 35 dollars while prices in Sweden quadrupled. And fiat after 1971. The same assets have behaved completely differently in each of them.

Under the gold standard everything except cash and gold gave a real return. Treasury bills 4.3 percent a year, the stock market 7.1. During the war years and the 1920s it was the opposite. The stock market returned minus 3.5 percent a year for 18 years, through the world war, the crisis of the 1920s and the crash of 1931.

Between 1932 and 1970 gold lost 2.3 percent a year in purchasing power measured in kronor. The price was fixed while everything else rose. When the price was released in 1971, gold had decades of inflation to catch up on, and it took nine years: from 37 dollars per ounce in 1970 to 455 dollars in December 1979.

That is why I do not read gold's 10 percent a year in kronor since 1971 as a return on gold. A large part is a one-off adjustment that should have happened gradually since 1934. The gold price shows how much the money has lost. It is not a return you can count on getting again.

And here comes what surprised me most. Swedish equities have returned the most above inflation under fiat, 9.8 percent a year since 1971. More than under the gold standard. When money loses value, it is the assets that absorb it. Fiat forces savers to become investors, whether they want to or not.

Over ten years it is a lottery. Over thirty, gold has almost never won.

Nobody saves for 147 years. So the question becomes what you would have got depending on when you started and how long you held on. Every point in the chart is a starting year.

In Sweden gold has beaten the stock market with dividends in 37 of 138 ten-year periods and in 19 of 128 twenty-year periods. Over thirty years in 8 of 118, and all eight began between 1903 and 1913. Anyone who bought Swedish shares just before the First World War had to wait through war inflation, the crisis of the 1920s and the Kreuger crash before the stock market caught up.

In the US the picture is even clearer. Over thirty years gold has never beaten the S&P 500 with dividends. Not once in 71 possible starting years since 1926.

Still with me? Over ten years, timing matters almost more than what you buy. Over thirty years, history has given almost only one answer. And the exception is worth remembering: it came when the war broke the gold standard.

Without dividends the Stockholm Stock Exchange has quadrupled in gold over 147 years.

In The big trend since 1971 I measured the stock market in gold instead of in money. Here I do the same thing as far back as it goes. With dividends reinvested, the Stockholm Stock Exchange today buys a thousand times as much gold as in 1878. The price alone, without dividends, has quadrupled. That is 1.0 percent a year.

The curve follows the monetary system almost perfectly. It rose under the gold standard, fell during the war years and rose again under Bretton Woods. Then came the 1970s. In 1980 the Stockholm Stock Exchange's price index bought less gold than in 1878. A hundred years of price gains, erased when measured in gold. The peak came in 1999, when gold was most forgotten. Since then the price index has lost 80 percent measured in gold.

The S&P 500 shows the same shape. Without dividends it buys three times as much gold as in 1925, and 64 percent less than in 1999.

Gold has been under water for 92 years. The stock market for 32.

Return per year says nothing about how it felt. Here is the longest time each asset has spent below its previous peak, measured in purchasing power. That is how long you would have waited, if you bought at the top, just to get back to zero.

Gold reached its highest purchasing power in kronor in 1887, in the middle of the gold standard's price decline, and was not back there until 1979. That is 92 years. If you bought gold in kronor at the 1982 peak it took 27 years to get back. In the US it took 30 years from the 1980 peak, and in the year 2000 gold was down 75 percent in purchasing power.

Government bonds from 1938 took 58 years to recover. Treasury bills 51 years. Cash has never recovered from the 1887 peak, and it never will.

The stock market has had deeper falls but shorter waits. With dividends the longest is 32 years, from 1912 to 1944. Without dividends, 74 years. In the US the S&P 500 with dividends has at most been under water for 14 years, after the dot-com bubble in 1999.

What River's charts show, and what they do not.

Gold was the best money we have had. We believe the next is digital.

Gold has worked as money for a simple reason. It happens to be the only material that cannot be destroyed, that costs enormous amounts of energy to produce more of and that nobody can print. A kilo of gold from 1879 buys eight times more today. No currency in history can say the same.

But gold also has a problem, and it is the same problem Johan Palmstruch solved in 1661. It is heavy, hard to send and must be stored by someone. That is why paper took over, and that is why paper could be printed.

That is where bitcoin comes in, in my view. There will never be more than 21 million. Nobody can print more. And it can be sent anywhere in the world in minutes without anyone needing to store it for you. Gold's properties, without gold's problems.

But I want to be clear about what the data in this analysis says and does not say. Bitcoin has existed since 2009. It has not passed a single thirty-year test, because it has not existed for thirty years. It has fallen more than 70 percent four times. This section is our assessment of the future, not a result in the figures above.

That is why we have built Futureproof. A simple way to save in assets that nobody can print more of, gold and bitcoin, gathered in one and the same portfolio. A portfolio of scarce assets. Personally I see it as the saving I myself have been missing: money built to preserve what I have earned, instead of losing it.

But the same honesty applies here. Scarce assets fluctuate. The data above shows that even gold has been under water for decades, and bitcoin has a considerably shorter history. What you save can become worth less.

What I take with me

That connects back to the true-flation bar from my previous analysis. Since 1971 the S&P 500 with dividends has returned 11.2 percent a year in dollars, against the bar of 10.3. That is the entire margin after half a century.

I could bury a gold coin today and ask my descendants a thousand years from now to dig it up. It will still be gold. No fiat currency has so far held its value for even a hundred years. Money should preserve. Ownership should grow. The problem with fiat is that it forces you to take the owner's risk just to keep what you have already earned. That is the problem I believe we can solve.