Advertiser Disclosure

The Creation of Money

1 What is Money?

How to use money that already exists? That’s easy. How to create money in the first place? That’s more complicated. This article will explain how money comes into existence. It will also explain how the different ways money comes into existence affects its use and abuse.

First, a brief definition of money. Money is any medium of exchange. The crucial word here is “medium.” If I have eggs and you have bacon, we might agree to exchange a few eggs for several slices of bacon. But there’s no medium of exchange here. The exchange is direct, with real goods being exchanged for real goods. Thus we might say that the exchange is “immediate” — in other words, there’s nothing mediating the exchange.

By contrast, when money takes part in an exchange, it acts as a medium or intermediary to facilitate the exchange. Money, in many instances, acts as an instrumental good. That means it acts as an instrument to procure the real goods that we desire and that make a real difference to our lives. I can enjoy a $20 steak (a real good), but I can’t enjoy a $20 bill (an instrumental good) except for the benefit I can derive from it, such as by buying steaks.

But the distinction between real goods and instrumental goods is not hard and fast. Take cigarettes. In prisoner-of-war camps in WW2, cigarettes were used as money to purchase other items (bread, jam, butter). Used in this way, as a medium of exchange, the cigarettes served as instrumental goods. But a prisoner might also simply have wanted to enjoy a good smoke, In that case, the cigarette became a real good.

Gold and silver, which have an illustrious history serving as money, share this same duality of being both real and instrumental goods. Thus gold and silver coins, in their capacity as money, constitute instrumental goods. But as, say, metals that can be made into jewelry, gold and silver constitute real goods.

2 How Does Something Become Money?

Although defining money as a medium of exchange gets at the heart of what money is, this definition raises a prior question: How does something get to be a medium of exchange in the first place? This is not so much a question of how money is created, but what motivates people to use a thing as money. Cigarettes, for instance, were money in WW2 POW camps. But how did that happen? Presumably, the prisoners could have used other things as media of exchange, such as razor blades. The WW2 US Medical Research Centre provides an enlightening look at what was in WW2 American prisoner-of-war relief packages, and thus what prisoners had to trade and what might count as money in the camps.

In any case, cigarettes were the prime form of money among the WW2 prisoners, so it’s fair to ask how that happened? One’s initial response to this question might be “convenience.” Cigarettes are convenient to exchange. They come as nice discrete units. They can be readily arranged in packs and cartons. They’re light and easily transportable. All features we tend to like in money. But none of these features were strictly speaking necessary for something to serve as money in the POW camps.

Decisive in the transformation of cigarettes into money was the widespread agreement among prisoners to treat cigarettes as money. But how did the prisoners reach such an agreement? Having a medium of exchange (i.e., money) makes exchange easier and more abundant. It helps an economy, even a prison economy, to have money, and once a convenient form of it was found in cigarettes, the agreement among prisoners to use cigarettes as money happened naturally. Perhaps other items could have served equally well. Just what constitutes money in a given situation is historically contingent. It could have been different.

The Alchemist by David Teniers the Younger

The agreement that transforms a real-world item into money is more mysterious than any alchemical transformation. The subtitle of NPR money-guy Jacob Goldstein’s 2020 book on money gets at an important truth about money.[1] Titled simply Money, the book’s subtitle reads The True Story of a Made-Up Thing. Money is something “made up” in the same way that children might make up the rules to a game that they invented. There’s nothing in nature that requires any real-world item to serve as money (or a children’s game to have one set of rules rather than another). We, by an act of social agreement, turn such items into money (or set the rules for a game).

Goldstein’s view of money as a made-up thing is correct as far as it goes, but can stand elaboration. The well-known philosopher John Searle provides that elaboration in his book The Construction of Social Reality.[2] Social realities for Searle include such things as marriage, games, property, governments, law, and, yes, money. Social realities arise through what Searle calls “collective intentionality.” Collective intentionality here describes the type of agreement within the social group that brings social realities into existence: we collectively intend for something to serve a given social purpose.

Social realities, it turns out, are objective because our collective intentionality makes them so: it’s clear whether a couple is married, it’s clear what the rules of a game are, and it’s clear whether something counts as money. The problem with Goldstein’s reference to money as a made-up thing is that it gives the impression that money is a fiction. But if it is a fiction, it is a fiction in which we all agree to participate and in which we treat money as real. That’s why I find money as a “social reality” more insightful than money as a “made-up thing.” As a social reality, money is fully objective — it is not simply a matter of personal taste.

The creation of money is therefore an act of collective intentionality that brings into existence a social reality by which exchange is mediated. In the most general terms, that’s how money is created. Every instance of money that we know follows this pattern. Where things get interesting is the types of money we create and how we justify their creation. With the right agreement or collective intentionality, anything might count as money. But in fact very few things do end up counting as money. We’ll turn to these different types of money in due course. But first, we need to consider some features that we like to see in our money.

3 Features We Like Money to Have

Money is any medium of exchange. Yet for money to be useful in facilitating exchange, it helps for money to have certain features. We now turn to these features. Think of these features as desiderata, characteristics we regard as desirable for money to have. None of these features is strictly speaking necessary for something to count as money. But lacking these features, it becomes less likely that something will remain a viable form of money for long, giving way to newer forms of money that do exhibit these features.

To understand the creation of money, it helps to understand that the history of money is fraught with monetary failures — monetary experiments that turned out badly and where people abandoned one form of money only to create another.[3] Because there are so many failed monetary experiments, the creators of money typically try to build into their money certain features and safeguards to prevent the money from going bad. Typically, for money to go bad, it has to go really bad. That’s because money in an economy usually acquires a certain inertia so that its users are loath to give it up unless it really falls off a cliff (as in hyperinflation, which has killed many currencies).

Two features that one almost always sees associated with money are that it should be a unit of account and a store of value. Unit of account just means that we can do arithmetic and accounting with money. We can count it. We can divide it. And when we divide it into parts and then put the parts back together, we get back what we started with. As a unit of account, money is also fungible, which means that two items of the same amount of money are interchangeable. So, if you loan me a dollar bill and I pay back the loan with a different dollar bill, we are quits.

We also like to see money serve as a store of value, the point being that money has a certain purchasing power in the present and it should retain that power in the future. As a store of value, money should, ideally, be immune to inflation and deflation. We might think that only inflation undercuts a money’s ability of store value (as when a dollar buys tomorrow only half of what it buys today). But deflation is a problem as well for any form of money. If I’m a seller and the currency is deflating, then I need to sell my products sooner rather than later. If I can sell a widget today for $1, then that $1 will be worth more tomorrow. But if I have to sell it tomorrow, because of deflation I may have to discount the widget and so end up with less than a dollar.

It’s said that debtors like inflation and creditors like deflation. If I’m a debtor and I owe $10,000, and if inflation has cut the value of that dollar amount in half, then I should be able to earn $10,000 with work that previously would only have gained me $5,000. And the $10,000 that the creditor gets when I pay off the loan with inflated dollars will only be worth what previously was able to purchase $5,000. So the debtor will have gained half the value of the loan and the creditor lost half the value on account of inflation.

Similar reasoning applies to deflation, this time favoring the creditor and hurting the debtor. We saw debtors getting hurt in this way after the financial crisis of 2008 when housing prices fell sharply (deflation) and home owners with mortgages (debtors) suddenly had their houses under water (they were unable to sell their houses for what they owed on them).

Yet to say that debtors like inflation and creditors deflation is simplistic, missing the larger economic picture. Both inflation and deflation make for economic instability. Neither is good for the economy as a whole. Buyers and sellers as well as borrowers and lenders may make out well on one transaction because of the price instability that inflation or deflation introduces, but there will be other transactions on which they will make out badly. And the uncertainty thus introduced will undermine the economy as a whole.

As an interesting side note, gold as money would constitute a highly unstable currency, introducing bouts of serious inflation and deflation.[4] For instance, in September 1980, gold was at $674 an ounce but in March of 1982, 18 months later, it stood at $320. This more than 50 percent loss of value represents an inflation rate of over 100 percent when money is denominated in gold. An item valued at $674 would require an ounce of gold to purchase it in September 1980 but more than two ounces in March 1982, assuming no inflation in the U.S. dollar.

As it is, the dollar’s inflation rate over those 18 months was 12.5 percent, as can be verified using the Bureau of Labor Statistics’ Consumer Price Index calculator. But even factoring in that inflation rate, gold as money lost half its purchasing power during those 18 months. This loss can be seen in the following graph, which shows the price of gold from 1960 to 2020 adjusted by the Consumer Price Index:

There’s a camp of monetary theorists who hold that gold is the ideal money and that the best thing that can be done for a nation’s money is to return to the gold standard. And perhaps if everyone agreed that gold is the best form of money and that we should all go on the gold standard, then the sort of inflation described in the last paragraph would simply not arise. In that case, we wouldn’t have the U.S. dollar, which is a fiat currency decoupled from gold, to fall back on and confuse the true value of gold.

But even in a world where gold is money, gold has been subject to inflation. Thus in 16th century Spain, with the importation of gold from the New World, prices across the century rose by 300 percent. Then again, in the hundred years after the founding of the Federal Reserve in 1913, with an on-again, off-again gold standard for some of that time and a fiat currency for most of that time, inflation came to around 2,500 percent.

“Fiat” is the Latin for “let there be,” and it’s an allusion to the biblical book of Genesis, where God creates from nothing by simply saying “let there be X” and X comes into being (in the first act of creation in Genesis, X is light). Fiat money is therefore money that is made money simply by a monetary authority declaring it to be money. Its creation is in its declaration.

In any event, my point here is not to make a case for or against gold as money, but to underscore that we want in money a store of value, where the purchasing power of a unit of money stays roughly constant, varying as little as possible over time. Gold has had issues here. But so have all other currencies given enough time and stress.

Milton Friedman famously, and rightly, remarked, “Inflation is always and everywhere a monetary phenomenon in the sense that it is and can be produced only by a more rapid increase in the quantity of money than in output.”[5] The flipside is also true, namely, that deflation is a monetary phenomenon of not having enough money to track output.

So the challenge of money creation is to create enough money to avoid deflation but not too much to cause inflation. Unfortunately, achieving the Goldilocks Principle of “not too much, not too little, just right” for the amount of money in an economy is easier said than done. Insofar as money consists of some inherently rare physical (or digital) substance, its creation may be difficult, which could then result in less money than needed, and thus in deflation.

But we live in an age of fiat currencies where money can be created from nothing. The problem thus overwhelmingly becomes not having too little but having too much. As a consequence, an important feature of money that we like, and indeed need, to see, and which is crucial to rendering it a store of value, is limits on its proliferation. Often this is put in terms of money needing to be scarce.

Other features that we like to see money exhibit also exist, some specific to the type of money being considered. For instance, digital currencies can run into a double-spend problem, in which the same item of currency can appear in two simultaneous transactions, thereby doubling purchasing power via a flaw in the currency. In consequence, a feature we want to see in our money is its ability to rule out double spending . Because physical things, unlike virtual things, can reside in only one place at one time, physical currencies avoid the double-spend problem. We will turn to other features that we like to see exhibited in money as they arise. We turn next to specific types of money.

4 Physical Money

Physical money has historically taken many forms. Cowry sea shells have served as money throughout the rim of the Indian Ocean. Cocoa bags containing thousands of cocoa beans served as money for the Aztecs. The Micronesian Yap islanders made money out of circular doughnut-shaped limestones (known as fei) imported from far-away islands. These stones were often so heavy that they couldn’t be moved around very easily. One was so big that it ended up at the bottom of the sea just before reaching the island. Yet even at the bottom of the sea it was used as money, the islanders keeping track of whom it belonged to!

Thus the Yap islanders, simply by agreeing that so-and-so now owns that particular fei, were able to transfer ownership of this form of money without physically relocating it (as we typically do in ordinary cash transactions). An interesting feature of these fei is that they were all so different in size that they didn’t, and indeed couldn’t, constitute a unit of account; and yet they were still money in the sense that they mediated exchange.

Paul Einzig, in his classic Primitive Money: Its Ethnological, Historical and Economic Aspects,[6] recounts these and many other forms that physical money has taken. What follows is a partial list of the types of physical money that he considers. It shows how primal the urge in humans is to create money:

Primitive Moneys

  • Mat and bark cloth money in Samoa
  • Whales’ teeth as currency in Fiji
  • Bead currency in Palau
  • Feather money in Santa Cruz (Solomon Islands)
  • Boar tusks in New Guinea
  • Rice currency in the Philippines
  • Drum currency in Alor (Indonesia)
  • Bees’ wax as money in Borneo (Indonesia)
  • Gambling counters as money in Thailand
  • Tea brick currency in Mongolia
  • Coconut standard on the Nicobar Islands (India)
  • Grain medium of exchange in India
  • Reindeer in eastern Russia
  • Iron currency in Sudan
  • Salt money in Ethiopia
  • Livestock standard in Kenya
  • Calico currency in Zambia
  • Gold dust currency in Ghana
  • Fur currency in Alaska
  • Maize money in Guatemala
  • Sugar money in Barbados
  • Tobacco money in Bermuda

One thing that becomes immediately evident from this list is that people readily create money from the things on hand, and that they do it irrespective of the size of the group in which they find themselves. The island of Yap is small, with just over 10,000 inhabitants. (It is now a part of the Federated States of Micronesia.) As it is, you don’t need to be a country to have a currency. Local moneys, existing in small communities and independent of government control, exist even in the U.S.[7] The idea that money must be the creation of government, as promoted by monetary theorist Georg Friedrich Knapp in his widely influential 1905 book The State Theory of Money, goes too far.[8] Money can be the creation of the state, but it need not be.

4.1 Commodity Money

All the physical moneys described above are examples of commodity money, which is to say these moneys have intrinsic value by being reasonably scarce. These moneys are not like air and water, which come in unlimited supplies. As commodities, they come with built-in limitations on proliferation. They are to varying degrees scarce, and their scarcity correlates with their value in exchange.

With paper money, by contrast, the value of the money is supposed to exceed the value of the paper and ink used to print it, and so the temptation to counterfeit paper money is readily understandable (paper and ink are, after all, relatively cheap). With commodity money, on the other hand, counterfeiting is less the issue than debasing or misrepresenting it. If salt were money, as it has been historically (we get our word “salary” from it), then we might debase it by adding some white crystalline powder to it that’s cheaper than salt and yet makes it appear as salt. If sea shells are money, perhaps we can shape some pieces of wood or plastic to look like them, thus misrepresenting that we have a certain type of commodity money when in fact we don’t.

At the end of the day, however, commodity money can’t really be counterfeited in the way that paper money can. Make the counterfeit of paper money sufficiently indistinguishable from the real thing, and you’ve added bogus money to the economy, which can end up hurting the economy because the counterfeit currency artificially inflates the money supply. But make a commodity sufficiently indistinguishable from the real thing, and you’ve got the real thing. If it walks like duck, quacks like a duck, swims like a duck, and in all the ways we can tell seems like a duck, then it’s a duck. So too with commodities.

Thus with commodity money, it will always be possible to determine whether a supposed instance of it coincides with whatever standard is set for it. With gold as money, for instance, it has historically been common to use a touchstone to determine whether the quality of the gold being used in an exchange matches the standard set for it as commodity money. And even if the government requires all gold coins, say, to be minted at the government mint, it doesn’t really make sense to say that gold coins that are made with a private coining press and that are indistinguishable from the government issue are counterfeit. Economically, they are every bit as legitimate as the government issue. Indeed, one could just as well have taken the raw gold to the government mint to have it turned into gold coin. Strictly speaking, therefore, counterfeit commodity money is an oxymoron.

In western civilization over the last two millennia, probably the biggest challenge facing commodity money has been debasement. The Roman empire around the time of Augustus had silver coins that were of high purity. Over the next 200 years, the proportion of silver in those coins kept being lowered, though the Caesars that debased the coins expected people to treat these debased coins as having the same purchasing power as the earlier purer coins (as when the government wanted to buy things from the people). Of course, the people didn’t treat the debased coins as identical to the earlier purer coins. People would instead hoard the purer coins (in line with Gresham’s Law, in which bad money drives out good), and the economy suffered the resulting inflation as the currency kept losing value.

The emperor Constantine, in the fourth century, understood the problem that debasing the currency posed to the ecoomy. He therefore minted the pure gold solidus (with 4.5 grams of gold per coin) on a large scale, giving it pride of place in Rome’s money. The solidus kept itself free of debasement for 700 years, which may be a record for a metallic commodity money.

4.2 Redeemable Paper Money

When we think of paper money, we usually think of fiat money, in which the government or its designated monetary authority simply starts the printing presses and outputs pieces of paper to be used as money. But this isn’t strictly speaking true. Paper money need not be fiat money provided it is backed by something of intrinsic value, such as gold.

Look at the following two $100 bills:

1928 Federal Reserve Note redeemable in gold
2009 Federal Reserve Note not redeemable in anything

The first $100 bill states that it is “redeemable in gold on demand at the United States Treasury or in gold or lawful money at any Federal Reserve Bank.” This bill therefore substitutes for gold commodity money. The second $100 bill simply states “this note is legal tender for all debts, public and private,” which means that people in the U.S. are obliged to accept this form of money in payment even though it is unbacked by anything of intrinsic value.

In 1928, the U.S. dollar was still backed by gold. In 1933, Franklin Roosevelt made the dollar unredeemable in gold and at the same time had all U.S. citizens turn in their gold, with the government paying citizens $20.62 an ounce, only to raise it immediately to $35 an ounce (thereby devaluing the dollar and thus inflating it). In subsequent years, the dollar still kept some connection with gold, in that countries could redeem dollars in gold, though not individuals. With France, especially, redeeming a lot of U.S. dollars for gold around 1970, Richard Nixon in 1971 closed the “gold window,” as it was called, thus permanently removing the dollar from even the most tenuous connection to any gold standard. Thereafter, the dollar became a pure fiat currency.

The value of redeemable paper money should in principle coincide with what it can be redeemed for. Thus any government or monetary authority that prints redeemable paper money should be entitled to put it into circulation directly, buying stuff that it could otherwise buy directly with the commodity. For instance, transactions in the U.S. that previously would have been handled with $20 gold pieces could now legitimately be handled with the redeemable paper money.

With redeemable paper money, the question arises how much of the commodity backing the paper is on hand. Paper money that is 100 percent backed is perfectly safe in that every note can be turned in and redeemed. A government that prints paper money with 100 percent backing is entitled to treat the paper as equivalent to the corresponding amounts of the commodity, spending the paper money freely. But what if the backing is less than 100 percent, say, 25 percent? If the government can freely spend this only partially backed currency, it has essentially quadrupled the purchasing power of its actual commodity on hand. This has all the feel of a free lunch where somebody is going to be left holding the bag.

One obvious concern, when the percentage backing of a redeemable paper currency drops to less than 100 percent, is a “run on redemption” that could leave some people with unredeemed notes. Even so, partial backing seems the norm with redeemable paper currencies. Thus the Federal Reserve Act of 1913 required backing of only 40 percent in gold. Paper money offers so much convenience that people are unlikely all in one mass to redeem the entire money supply.

Granted, with only partial backing, there is a risk of redeemable paper money proving unredeemable because the store of the relevant commodity, in this case gold, has run out. But there’s also the government standing behind the money and able to offer assurances that it will secure further gold (as in this example) if too many notes go up for redemption. Of course, there’s a point at which the percentage of gold (or whatever the commodity does the backing) is so low that the partially backed paper money becomes suspect. It’s unclear that there’s a magic number here.

The deeper concern with partially backed redeemable paper currencies nonetheless remains, namely, that the monetary authority issuing such notes is indulging in a free lunch, getting value for nothing, as with fiat currencies. In fact, one might see partially backed paper currencies as a hybrid of redeemable and fiat currencies. There’s some merit to this line of thought. Nonetheless, with redeemable currencies, even if only partially backed, the good faith of the monetary authority to guarantee redemption can be seen as a value-add to the currency. Likewise, incentives can be built into the currency to add further value, such as the government’s willingness to use it for taxes. These considerations mitigate any concerns about a redeemable paper currency being only partially backed.

Even though the U.S. dollar is no longer backed by gold, it’s an interesting exercise to imagine what percentage of the dollar would be backed by gold given the dollars in circulation and the amount of gold that the U.S. government owns. According to the Federal Reserve’s website, as of January 2021, there were just over $2 trillion in paper and coin money (M-zero money, as it is called). At the same time, according to the United States Treasury, the government’s gold reserve comes to 258,641,878.085 troy ounces.

At the start of 2021, the spot price of gold has been hovering around $1,850 per troy ounce. That puts the total value of government-owned gold at around $475 billion, or just under 25 percent of total paper and coin money. Of course, there’s a lot more money in circulation in the form of checking and savings deposits in banks (known respectively as M-one and M-two money). All this bank-deposit money is supposed to be readily convertible into paper and coin money. The total of these deposits In January 2021, according to the Federal Reserve, comes to about $20 trillion dollars, putting the gold backing of the dollar, if it were redeemable in gold, at 2 to 2.5 percent.

4.3 Fiat Paper Money

Fiat paper money, though unredeemable and not backed by any commodity, is never just paper with ink printed on it. It always comes from a monetary authority, usually the government, or a government-sponsored central bank, that issues the paper money. Issuing the money is not just a matter of printing it but also getting it into circulation and doing so with incentives for people to actually use it (such as using it for taxes and requiring its acceptance for payments) as well as safeguards against counterfeiting and hyperinflation.

Thus, even though the creation of fiat paper money may seem as simple as getting a printing press and firing it up, there is in fact more to the creation of fiat paper money. If it were just a matter of printing paper tokens, its creation would be no different from the creation of Monopoly game money. But precisely because our fiat money is, by a requirement of law, made to be “legal tender for all debts, public and private,” it assumes the role of real rather than play money.

Because most government-sponsored fiat money these days is issued not in the form of printed paper money but as electronic bank deposits, the mechanisms for the creation of fiat money in general will be considered in a subsequent section where we can consider both types of issuance together (paper notes and bank deposits).

5 Money as Credit

A myth exists, still widely circulated in economics textbooks, that money arose out of the limitations of barter. To take the example at the start of this article, if Alice has some eggs and Bob has some slices of bacon, they can directly exchange eggs for bacon only if there is, as it’s called, a “double coincidence of wants.” In other words Alice must want Bob’s bacon and Bob must want Alice’s eggs, and of course these two food stuffs need to exist in the right amounts and at the right rate of exchange to satisfy both parties. Money, economics texts tell us, arose to facilitate transactions where Alice, say, wanted what Bob had, but Bob didn’t want what Alice had; or, if they wanted each other’s goods, they couldn’t work out the right proportions of the exchange.

David Graeber, in his book Debt, and Felix Martin, in his book Money, argue convincingly that this account of barter evolving into money is a fabrication with no basis in history or anthropology (as confirmed by examining exchange in currently existing “primitive” cultures).[9] They argue that barter has always been a very limited form of exchange, often ritualized, and that in fact what facilitated exchange throughout history are systems of credit in which debts and obligations, along with their payment and non-payment, were carefully tracked.

Some of the first human writing goes back 5,000 years to the Mesopotamian city of Uruk. The writing that archeologists have uncovered there is not myth or poetry but receipts and invoices. In other words, what they’ve found there are accounting ledgers and thus a system of credit. Credit always signifies a binary relation between creditor, who gives something of value, and debtor, who receives the thing of value with the promise to repay it in some agreed upon form. Credit is thus always an IOU: “I, Alice, owe you, Bob, 30 shekels of silver for the 20,000 square cubits of land next to Jacob’s tomb.”

The explanation of what thing of value exchanged hands in this credit transaction can, however, be eliminated. The IOU could simply read “I, Alice, owe you, Bob, 30 shekels of silver.” As long as this IOU is duly notarized, whether by Alice’s signature or by the elders at the city gate, it will be binding. Of course, there needs to be some limit on when Bob can collect on his debt, which often may be determined by convention.

Alice’s IOU to Bob mediates the exchange of Bob’s plot of land to Alice. Is this IOU therefore money? By our definition of money, as a medium of exchange, it is. The use of credit as a medium of exchange in this way has a long, and unbroken, history. Debt is thus turned into money. Even so, one might object that the debt itself is not the money here but rather the thing offered in repayment of the debt is in fact the real money here. But such an argument to undercut credit as money doesn’t withstand scrutiny.

To see that credit really is a form of money, consider that repayment of the IOU may be possible without actual shekels of silver ever trading hands. One way to see this is for Alice to make her IOU to Bob negotiable. So, instead of writing the IOU with Bob’s name, Alice can make it out as follows: “I, Alice, owe to the holder of this note 30 shekels of silver.” Closer to our day, with dollars rather than shekels, negotiable bearer bonds (now largely obsolete) served a similar role (also paying interest over time rather than a flat rate).

Duly notarized, Alice’s IOU now entitles anyone holding it to claim 30 shekels from Alice. This note, by being open-ended in who gets paid by Alice, is thus negotiable. It can therefore circulate among people, trading for goods worth 30 shekels of silver, or close to it, depending on Alice’s reputation for fulfilling her promises. As long as people circulating the note think that Alice is “good for it,” it can keep circulating without any shekels seeing the light of day.

True, credit money like this, issued by a debtor, always involves inherent risks to those who accept the notes because it is based on promises, and promises can always be broken. Nonetheless, societies always form legal structures to encourage/enforce the payment of debts. Money is credit.

But doesn’t credit money still presuppose some form of real money in which the underlying debt is denominated? And when a debt is finally paid, won’t that real money finally need to rear its head? It is true that credit money always depends on some more basic underlying form of money or value. And yet, any such underlying “real money” need never appear in a credit transaction.

Consider, for instance, Alice, Bob, and Carol: Alice gives Bob a negotiable IOU, signed by her, for 30 shekels of silver. Bob gives Carol a negotiable IOU, signed by him, for 30 shekels of silver. And Carol gives Alice a negotiable IOU, signed by her, for 30 shekels of silver. If Alice now comes into possession of 30 shekels of silver, she can pay Carol, who can pay Bob, who can pay Alice. That clears all the debts and Alice is back to having 30 shekels of silver.

But all three debts could equally be cleared simply by having them cancel each other, without any shekels of silver ever making their appearance. The fact is that in a system of credit and clearing, very little underlying “real money” may ever be needed. I made a similar point elsewhere not about credit money but about the velocity of money, where as money is exchanged with increasing rapidity, it may be possible to make do with less and less of it — in the limit an infinitely fast penny could do all the monetary work: see my Expensivity article “The Infinitely Fast Penny.”

6 The Multiplication of Money

Credit in the form of negotiable IOUs constitutes money. Is it therefore safe to say that credit creates money? Credit is one way of making more money out of existing money. But at the end of the day, credit money still depends on some form of more basic money, even if it is a single penny as described in my article about an infinitely fast penny. We still need to explain the existence of the penny or whatever the more basic form of money we take to be our starting point.

Nonetheless, as a way of expanding or multiplying existing money, credit money is still relevant to the creation of money. The inventor of a widget is its creator. The person who makes additional copies of the widget does not create the widget but does create ways to make more widgets, and that is relevant to the act of creating the widget since the aim of creation is not typically to make just one instance of the thing but to set the course for more things like it to be brought into existence.

To change the metaphor, imagine a flock of sheep. Given an initial male-female pair of sheep, we can explain the rest of the sheep through a process of multiplication. Creation would then explain the initial pair, multiplication the rest of the flock. Likewise, existing money can be multiplied. Credit money is a case in point. Yet the example most prevalent in our day is money multiplied through bank loans.

When money is deposited at a bank (it doesn’t matter what form the money takes — commodity, paper, or digital), banks are entitled to loan out a large piece of the deposit. This is known as fractional reserve banking, with the reserve rate being how big a fraction of the deposit the bank must keep on hand without loaning it out.

For U.S. banks with $125 million in assets, the reserve rate was, until recently, 10 percent, with smaller banks having a lower reserve rate. The reserve rate is often denoted by R, and the reciprocal 1/R is known as the money multiplier, whose product with the starting amount (i.e., the amount initially deposited) yields how much money one ends up with. For R equal to 10 percent, the multiplier is therefore 1/.10, or 10. If the reserve rate R were 5 percent, the multiplier would therefore be 1/.05, or 20.

As it is, on March 26, 2020, the Fed took emergency action to lower the reserve requirement to 0 percent. With a 0 percent reserve rate and thus an infinite multiplier, I suppose it could be said that fractional reserve banking, in which there is in fact no reserve, might constitute a form of money creation rather than money multiplication, but that is perhaps a fine point.

In any case, to see how fractional reserve banking has worked out historically, imagine a non-zero reserve rate and Alice with $10,000 in cash wanting to buy Bob’s brand new top-of-the-line BMW. Bob has put this car on the market and is asking $100,000. Alice wants it and decides to get financing from the bank. Alice, to buy Bob’s BMW, therefore goes to her bank and deposits $10,000 into her account as a down payment on Bob’s car. But Bob’s asking price is $100,000. Incredibly, through the miracle of fractional reserve banking, the bank can give Alice the loan to buy Bob’s car even if Alice’s $10,000 is the only deposit that the bank has.

With a reserve rate of 10 percent, Alice’s bank can loan 90 percent of whatever is on deposit. Since Alice has deposited $10,000, the bank can loan out $9,000 of that initial deposit, which they loan to Alice. But Alice now has an additional $9,000, which she can in turn deposit. She now has a credit of $19,000 and a loan of $9,000. But that $9,000, as a deposit, now allows the bank to loan out 90 percent of that amount, or $8,100, which goes to Alice, who adds that amount as a credit to her account but also as a debt to be repaid.

Round and round this goes. Do the calculation, and it ends with Alice having $100,000 in her account and a $90,000 loan. She can then buy the BMW from Bob by transferring $100,000 from her account. At the end of the day she has a new BMW, a zero balance in her account, and a bank loan for $90,000 that she must pay off. And one more thing: Alice needs to pay interest on that loan, interest that will always be more than the interest the bank pays on deposits.

When I first learned of fractional reserve banking, my immediate reaction was HOW DO I BECOME A BANK? Banks, by being able to engage in fractional reserve banking, are granted an exorbitant privilege. The state extends that privilege, and banks are highly regulated because of it. As a consequence, I am barred from being my own bank, which would allow me to multiply my cash 10-fold (with a 10 percent reserve rate), loaning it to myself with all the moral hazard that entails.

In sum, through the miracle of fractional reserve banking, Alice’s $10,000 becomes $100,000. Yes, she now gets to buy and use the BMW, but because she owes $90,000 on it, she really doesn’t own it. Instead, if she defaults on the loan, the bank can claim it. There’s something unreal, and even surreal, about fractional reserve banking. Bob wants to sell his BMW for $100,000, Alice wants to buy it, but only has $10,000. So she goes to a bank that multiplies her $10,000, increasing it by $90,000, making a loan to Alice for that amount, and essentially owning the BMW to boot.

The sense of unreality that surrounds fractional reserve banking comes down to bank loans making the same money to reside in two places at once on an accounting ledger. By contrast, physical objects, can reside in only one place at one time. Thus when Alice deposits $10,000 in her account and the bank then loans out $9,000 (keeping $1,000 in reserve), the money gets multiplied. Alice can still claim her $10,000 at any time because it is on deposit. But the bank can only claim the $9,000 loan at whatever point the loan is due. So if Alice claims more than $1,000 before the loan gets repaid, the bank will not be able make good on her deposit.

This entire lending scheme depends on depositors not being over-eager to withdraw their funds. So long as depositors are content to maintain their deposits (at least in line with the reserve rate) and borrowers are repaying their loans without too many defaults, the system works, or at least seems to work.

Money multiplication through loans feels a bit like a Ponzi scheme in that, banks seem to be taking from Peter (the depositor) to pay Paul (the borrower). But in a Ponzi scheme, money is continually bleeding out and there’s no way to unwind contributors’ positions if they all want to be paid — as happened with Bernie Madoff’s investors.[10]

With bank loans, everything unwinds successfully provided that all the loans get paid off. Our banking system that multiplies money through loans may seem counterintuitive. Yet much of the money that passes through our fingers results from the multiplier effect of bank loans.

7 Money Creation by Central Banks

We saw in the last section how exorbitant the privilege is for banks to engage in fractional reserve lending. But if that privilege is exorbitant, the privilege of central banks to create money from nothing is even more exorbitant. Ordinary banks that multiply money by lending are at least starting with some pre-existing money — they can only lend what has been deposited (or re-deposited). Central banks, by contrast, can lend in the absence of prior deposits

When an ordinary bank credits an amount of money to an account, it is because it has transferred or loaned that amount from another account. But when a central bank credits an amount of money to an account, no other account need be invoked or accessed. Deposits made by ordinary banks depend on other existing deposits. Deposits made by central banks can, but need not, depend on prior deposits. Deposits made by central banks without drawing on prior deposits constitute money created out of nothing.

If this sounds magical and too good to be true, don’t take my word for it. According to the Boston Federal Reserve’s 1984 booklet “Putting It Simply”:

When you or I write a check, there must be sufficient funds in our account to cover the check, but when the Federal Reserve writes a check, there is no bank deposit on which that check is drawn. When the Federal Reserve writes a check, it is creating money.

Headquarters of the Federal Reserve, Washington, DC

7.1 How the Fed Buys Securities to Create Money

Central banks obey a certain logic in creating money, and it helps to understand that this logic is independent of whether the money created consists of paper notes, accounts marked on paper-and-ink ledgers, or information digitally encoded on servers. The logic is the same. For convenience, I’ll focus on the U.S. central bank, known as the Federal Reserve or Fed for short, but the same logic applies across most central banks.

The only exception to this logic is when a central bank loses its independence as a monetary authority, creating money with no effective restriction on how much the government can spend (thereby violating the prime monetary directive of limited proliferation). This is a recipe for runaway inflation, as we are currently seeing in Venezuela. It may also be in America’s future if Modern Monetary Theory (MMT) continues here to gain traction (see section 7.2 below).

With few if any exceptions, the effective restriction that prevents governments and their central banks from creating and then spending unlimited amounts of new money is that every dollar (or peso or yuan or euro or whatever) created must take the form of a loan that must be repaid, and whose failure to repay would constitute a default by the government taking the loan. By requiring newly created money to take the form of loans, the central banks impose discipline on the creation of money, a discipline that can be further tightened or loosened by varying interest rates.

The temptation to create money and then instantly turn around to spend it can be overwhelming, especially for governments stressed by disasters. To circumvent this temptation, governments have introduced a fiscal-monetary divide. Fiscal refers to what money is spent on, monetary to its creation. The point is that when the central bank, in our case the Fed, creates money, it doesn’t get to spend the money. Spending the money becomes the job of a separate entity, in our case, the U.S. Treasury. The Fed handles the monetary side of the divide, the U.S. Treasury the fiscal side.

Here’s how this works in practice. In a typical year, Congress and the president find that taxes do not meet all government expenses. The resulting deficit needs to be covered. Congress, usually with the president’s consent, therefore agrees on the need for additional funds, tasking the U.S. Treasury with issuing bonds (“treasuries” as they’re called — more specifically T-bills, T-notes, and T-bonds depending on the time till maturity).

The Federal Reserve Act prevents the Fed from buying treasuries directly from the U.S. government. But that doesn’t prevent treasuries from ending up on the Fed’s balance sheet, as can be seen from this snapshot of the Fed’s balance sheet as of early 2021. In practice, the U.S. Treasury raises money by selling its treasuries in auctions, to large banks that are licensed to serve as primary dealers.

When the Fed buys treasuries, it then buys them on the open market. So when the Fed buys a treasury, it credits the account not of the U.S. Treasury (which has already gotten paid from the primary dealers), but of the bank or the third party which sold it the treasury. In crediting the accounts of those selling it treasuries, the Fed creates a deposit without the need to debit a prior deposit. In this way, the Fed creates money.

When the Fed buys treasuries, these bonds are shown as assets on its balance sheet (at the time of this writing, the amount of treasuries on the Fed’s books comes to almost $5 trillion). Since these bonds are liabilities of the U.S. Treasury, the U.S. Treasury is supposed to pay the interest on these bonds and then pay back the principal at the time of maturity. Just as the issuing of these treasuries resulted in the creation of money by the Fed, the retiring of these treasuries by paying them off results in the annihilation of the money so created. Accordingly, money is loaned into existence and repaid out of existence.

In sum, when the Fed buys securities, such as treasuries, and holds them, they pay for them by creating new deposits and thus creating money. Such deposits, in pre-digital times, could be made by printing paper money and simply handing it to the party selling securities to the Fed or by simply writing a duly authorized credit in the the account of that party (whose account is likely to be at the Fed — central banks are, after all, banks). In a digital age, the Fed would simply authorize an appropriate addition to the digital bank account of the party selling the securities. The logic of money creation is the same in all such cases, regardless of the precise form that the fiat money takes.

7.2 Modern Monetary Theory

It may seem like a rigmarole for central banks to create fiat money by loaning it into existence, but what is the alternative? Some limitation must be placed on the creation of money. Without it, the danger of hyperinflation looms large. I personally am unsure whether this approach to creating money by loaning it is the best approach for imposing a limit on the proliferation of fiat money. But I am convinced that the absence of such limits, as is now increasingly being mooted in the guise of Modern Monetary Theory (MMT), is economically unsound.

MMT would let the government simply create and spend the money it needs to pay for its expenses (no need, at least initially, to tax the people; no need ever to loan money into existence as in our present debt-based money creation system). Then, when this freedom to create and spend money overburdens the money supply, MMT directs the government to contract the money supply by taxation. Taxes thus become a correction to an overheated money supply

It’s remarkable to see this proposal taken seriously. History shows that governments under stress (especially in times of war) will take heroic steps to increase the money they have to spend on what’s troubling them. Runaway inflation is therefore a clear and present danger with MMT. By the time taxes are imposed to remove excess money from the economy, the inflationary damage will have already been done.

MMT is also ideologically charged, providing an ideal way to advance socialist policies not by confiscating the means of production but simply by buying them, since the government is given the power to print as much money as it needs to buy whatever it wants. In any case, MMT is not ideologically neutral: capitalism does not stand a chance with it since the public sector, with unlimited funds, can always overrun the private sector.

MMT advocates might argue that I’m misrepresenting it and that the theory can build in safeguards against unduly competing with the private sector. But without the fiscal-monetary divide to slow the creation of money and discipline its spending, it seems that the private sector would be right to worry about government policies directed by MMT. In fact, the private sector should already be worried. According to the Cato Institute’s Downsizing Government website,

Since the 1990s, federal workers have enjoyed faster compensation growth than private-sector workers. In 2018 federal workers earned 80 percent more, on average, than private-sector workers. And federal workers earned 47 percent more, on average, than state and local government workers. The federal government has become an elite island of secure and high-paid employment separated from the ocean of average Americans competing in the economy.

With MMT in the driver’s seat, not only would this trend continue, but one could imagine private sector workers insisting that the government supplement their private-sector income to match that of government workers, especially given that the government can create and spend as much money as it wants where it wants. All such payments, however, would have strings attached, with the public sector eventually capturing the private sector. We see this encroachment already with health care. MMT would universalize this encroachment.

In their defense, MMT economists will claim that the U.S. is entitled to create unlimited money because its debts are denominated in U.S. dollars, and it can therefore never default on those debts.  MMT distinguishes the U.S., whose debts are denominated in its own currency, from other countries that owe debts in foreign currencies (such as U.S. dollars). But guarantees against default are not guarantees against runaway inflation. It’s the inflation that will kill you, even if the defaults don’t.

7.3 Central Bank Digital Currency

When a bank make makes a deposit, the deposit appears simply as a single number added to an account. Thus if the U.S. Treasury gives me a tax refund for $2,000 and if I have $3,000 in my account, a snapshot of my account will look something like:

  • $3,000 [previous total in my account]
  • $2,000 [U.S. Treasury tax refund]
  • $5,000 [new total in my account]

Moreover, when the Fed created the money in the U.S. Treasury’s account that it in turn used to pay my tax refund, the Fed simply added a number to the U.S. Treasury’s account.

But what if the Fed, instead of simply adding dollar amounts to the U.S. Treasury account, also included identifying serial numbers with them, in analogy with the serial numbers that we find on our paper dollars? Such serial numbers would then uniquely identify our digital currency in the same way that serial numbers uniquely identify our paper currency.

Accordingly, when a deposit is made in an account, it would no longer just be a number added along with the party adding it, but it would also include such a serial number. These digital serial numbers would then be forever associated with the corresponding digital currency, starting at the point of origin with the Fed (or whatever central bank originally issues it), appearing at the U.S. Treasury, and from there working through the rest of the banking system.

In fact, serial numbers precisely corresponding to those on paper currency won’t be need. It will be enough to have tracking pixels that follow any unit of money deposited from its point of origin (with the Fed) to the present. These tracking pixels would be created along with the currency by the central bank, and then allow for a complete transactional history of any item of money’s use to the present. In fact, there’s no need for the users of these digital currencies to even see these serial numbers just as we don’t see the tracking pixels and cookies that are used to track our activities online. But the third parties issuing and managing this digital cash will know these serial numbers and be able to track them. Perhaps, in the context of digital currencies, such numbers should be called “tracking numbers.”

What I’ve just described is now being called central bank digital currency, or CBDC. According to Bloomberg, the Federal Reserve, working with MIT, is planning on rolling out such a digital currency in 2021. It is being referred to as “Fedcoin.” China is further along in rolling out its CBDC, a digital yuan, which it is already piloting in several cities and intends to make ready for prime time at the 2022 Winter Olympics in Beijing. Commenting on this digital yuan in Bloomberg, the Atlantic Council’s Josh Lipsky asks what international tensions might arise “if American athletes are asked to use a currency that the Chinese government can completely track.”

That’s the whole point of a central bank digital currency: to include in the creation of money also a device for tracking it. Why would anyone want to do this? Money, as we’ve defined it, is a medium of exchange. But when money is used in an exchange, it’s not necessary that there be a medium to the exchange. If I’m wiring money from my bank account to someone else’s bank account, my bank and the other person’s bank act as a medium to the exchange. In other words, they are acting as intermediaries to facilitate the exchange.

But I could also visit the person I’m paying and simply give cash in the amount that I would otherwise have wired. Now, if I give cash directly to this person, there is no medium to the exchange — I’m simply giving the cash directly. That’s the beauty of cash, namely, that it eliminates any medium or third party to an exchange. It makes the exchange private, just between me and the party I’m paying. In fact, that would be one way to define cash: any money whose exchange does not require a medium to the exchange.

Since digital money always requires the aid of a computer operated by a third party, digital cash is, according to this definition, an oxymoron. In fact, one wonders if using the term “digital cash” constitutes an exercise in misdirection, since all the privacy and control we’ve come to expect of real physical cash is absent from digital cash. A corollary thus follows: if by cash we mean money that allows us to dispense with third parties, then cash must correspond to something physical — in practice paper or coin.

Not everyone believes that cash — real physical cash! — is beautiful. Harvard economist Kenneth Rogoff has even written a recent book against cash: The Curse of Cash.[11] As Rogoff sees it, cash is a rife ground for crime, making possible the drug trade, human trafficking, and tax evasion. Also, cash is unhygienic, spreading germs and disease. Frankly, I’ll take my chances with the unsanitariness of cash. And while I’m no fan of cash’s use by criminals, I’m much less a fan of a government that can track every one of my financial transactions.

Ask yourself where this tracking would end. In a cashless society, where every item of money has been issued as a central bank digital currency and can be tracked, certainly the privacy of those transactions comes to an end. But tracking is only the first step in a surveillance economy. Once one has started down this slippery slope, what is to prevent the trackers of my (and your) financial transactions from trying to shape my behavior? What if I’m getting ready to pay for a Big Mac and I see a pop-up on my phone indicating that the purchase I’m contemplating is not the best thing for my health and that I should not make it. Such paternalistic intrusions are in line with proposals by Cass Sunstein and Richard Thayler in Nudge. [12]

But where do these paternalistic intrusions end? Suppose I decide, despite the urging on my phone, nonetheless to buy and eat the Big Mac. What if next time, my bank, which holds my digitally tracked money, or my credit card, which depends on payment in digitally tracked money, refuses to let me buy a Big Mac? Or what if, instead, the government clamps down on McDonald’s and keeps it from receiving central bank digital currency from people like me whom it deems as risking their health by consuming a Big Mac?

In such cases, it might be argued that the government is still trying to act in the best interests of the people whose money it is tracking and now increasingly controlling. But what if the government decides to shift from a paternalistic to a more punitive role? Are you late with an alimony payment? Sorry, but you can’t take that business flight — we’re going to block payment until you’re caught up with your debt. Or perhaps the government will simply transfer funds out of your account, making the alimony payment for you without your consent. Big Brother of 1984 and the Beast in Revelation (which prevents people from buying and selling except with the mark of the Beast) seem just around the corner.

Lord Acton famously said that power corrupts and absolute power corrupts absolutely. A government that is able not only to create money but also to track in minute detail all financial transactions has absolute power. It is much more worrisome, in my view, than the criminals who are using cash to advance their criminality. These criminals will in any case find ways around a central bank digital currency (as with cryptocurrencies, considered in the next section). But government tracking of all monetary transactions will sorely compromise the freedom and independence of the individual, and that’s too great a price, in my view, to pay for the supposed blessings of cashlessness.

One final benefit of cash is worth mentioning before we move to the creation of money in cryptocurrencies, and that’s the ability of cash to limit confiscatory schemes by the government. If all money is digital, it has to reside on servers, and because all banks are highly regulated, that means all money will be subject to government control. This is true even in the absence of a full central bank digital currency with full tracking of every digital dollar. If all money resides digitally in banks, the banks are in a position to do what they like with it.

One proposal that is widely mooted these days is the imposition of negative interest rates. My memory goes back far enough to times when banks offered double-digit interest rates on deposits. I remember having a CD that matured in one year and paid an 8 percent return sometime in the early 1990s. Back then banks would prominently advertise the interest rates they were offering. With interest rates close to zero for the last several years, however, banks are less inclined to celebrate the rates they are offering. But imagine how a bank might advertise a negative interest rate of 2 percent: “Come leave your money with us. Deposit $100, and in a year we’ll give you back $98.”

So long as there is cash, banks offering negative interest rates are a non-starter. If you have $100 in cash, are you seriously going to deposit it with a bank that offers a 2 percent negative interest rate? Of course not. You’ll stick that $100 in a mattress or some safe place and hang onto it there. With cash, you won’t have to concern yourself with all the supposed benefits that negative interest rates are said to be bringing to the economy. You can lay aside the compulsion to buy stuff now because a dollar spent today will be partially confiscated tomorrow. And you’ll have the satisfaction of keeping your money as much as possible out of the government’s coffers (because that’s where at least some of the money that is removed from bank accounts by negative interest rates will end up).

Of course, cash is also totally incompatible with any central bank digital currency, which requires a monopoly on all money if it is to be successful.

8 Cryptocurrencies

The creation of cryptocurrencies is our final topic. This article has grown long, and I plan a full-scale treatment of cryptocurrencies at Expensivity in the near future, so I’ll focus here on the actual creation of money within cryptocurrencies, without a lot of commentary on cryptocurrencies as such.

The screenshot above was taken today (26 February 2021) from It shows the 10 most highly capitalized cryptocurrencies in the world at this moment (with the list extending on that site if you scroll down). Bitcoin, not surprisingly, dominates the list. It has in the last days come down from its all-time high of over $50,000 per bitcoin, which would put the total market value of all existing bitcoins at close to $1 trillion.

This is remarkable. Is it a speculative bubble? To read Charles Mackay’s classic Extraordinary Popular Delusions and the Madness of Crowds, one might think that Bitcoin is the latest public mania, like the tulip mania that infected Holland in the 17th century, where people would mortgage the farm for an unusual tulip bulb. Bitcoin and other cryptocurrencies certainly have that speculative feel, but there seems to be more going on. Bitcoins have been used for exchange in countries like Argentina whose national currency has become notoriously unstable. It has found a ready home in the $700 billion remittances market, in which money needs to be seamlessly transferred abroad. And it has been used to buy illegal drugs on such darknet markets as the now defunct Silk Road.

Given a growing sentiment among economists for government to eliminate cash, as outlined in the last section, cryptocurrencies may be the next best thing to cash, and some would argue even better. Like cash, cryptocurrencies promise to stay out of the reach of meddling monetary authorities. It’s therefore no surprise that some of the biggest advocates of cryptocurrencies are also deeply skeptical of government control of money. Cryptocurrencies thus have a lasting appeal, at least to people with an anti-Big Brother streak, and this fact is likely to keep these currencies alive and holding value for some time. How much value is another matter.

So, what is a cryptocurrency? Cryptocurrencies are digital currencies. Now any digital currency is going to use cryptographic methods. Information packets sent across the internet to make transactions with digital currencies must, at the very least, be encrypted to prevent eavesdropping. But the mere use of cryptographic methods in a digital currency does not make it a cryptocurrency. Nor does public-key cryptography in a digital currency make it a cryptocurrency.

To be sure, cryptocurrencies depend on public-key cryptography, in which the owner of an item of currency is alone in knowing a private cryptographic key with which to decrypt messages sent via the public key (which everyone has access to) and with which to sign and thus authorize a transfer of funds in the cryptocurrency from one account to another (such private-key signatures are absolutely essential to all cryptocurrencies). All the same, a scheme in which people use ordinary money (say, U.S. dollars) to buy cryptographic tokens that can be transferred via a private-key signature and then redeemed in ordinary money is not, properly speaking, a cryptocurrency.

The point of a cryptocurrency is to have a self-contained currency that is entirely digital, that is protected by cryptographic methods, and that achieves its value without the aid of other currencies, such as by being redeemable in them. Bitcoin and other cryptocurrencies have achieved this status, as their market capitalization shows. It’s not that bitcoins can, by some central authority, be redeemed in dollars, but rather that people are willing to spend dollars to buy them!

Bitcoin, whose pseudonymous inventor Satoshi Nakamoto proposed it in 2008, offered a huge leap forward in the power and appeal of cryptocurrencies. Its blockchain gave a complete and unalterable record of all bitcoin transactions. Its peer-to-peer distributed protocol eliminated the need for trusted third parties (which can always prove untrustworthy). Its consensus mechanism using proof of work to generate new bitcoins incentivized people to use the currency and maintain its peer-to-peer network.

Also, Bitcoin fulfilled various desiderata that one would necessarily want to see in digital currencies (compare section 3):  privacy (protection from eavesdropping), anonymity (ability to transfer funds without revealing one’s identity), immunity to double-spending (an item of currency can’t be spent in two or more simultaneous transactions), and nonrepudiation (prevention of taking back a transaction performed — no mulligans).

With the coming of Bitcoin, prior cryptocurrencies that used trusted third parties fell by the wayside. Cryptocurrencies developed since then have all taken Bitcoin as their point of departure, modifying some of its features but not altering its basic blockchain/ peer-to-peer network/ consensus-mechanism infrastructure. In this infrastructure, the consensus mechanism is key to the creation of new money within a cryptocurrency.

Each of these cryptocurrencies exists on a peer-to-peer network, which facilitates the transaction of its currency. But the peer-to-peer network also oversees the creation of new currency over time. The details of how this creation happens depend on how the cryptocurrency was structured from its inception.

It could happen that all the currency within a given cryptocurrency is specifed from the start, the maximum thus being available all at once, with it being distributed according to some scheme. Or it can dribble out over time, as with Bitcoin. The way the cryptocurrency is structured from the start may specify an upper bound to the number of currency items (as with Bitcoin, where the maximum number of bitcoins is set at 21,000,000, with it becoming progressively more computationally difficulty to generate more bitcoins over time). Or it can be more open-ended, as with Polkadot, whose DOTs started with an upper limit of 10,000,000 but then, according to Coin Bureau, was modified “to allow for a somewhat alarming degree of inflation.”

Cryptocurrencies use two forms of consensus mechanisms to create new currency: proof of work (PoW) and proof of stake (PoS). Bitcoin, with a market capitalization of close to $900 billion, as of today, uses proof of work. Among cryptocurrencies that use proof of stake, the one with the highest capitalization is Polkadot, which as of today sits at about $30 billion.

A consensus mechanism creates money within a cryptocurrency in a clear and well defined way. Because a cryptocurrency exists in a peer-to-peer network, users on that network are all eligible to receive new currency (unless the currency has reached its agreed-on maximum). Users that fulfill certain requirements then receive currency, whether with certainty or through some randomization, after a block of trading (these are blockchain-based cryptocurrencies, in which trading happens in discrete blocks).

For cryptocurrencies that use proof of work, the user on the peer-to-peer network who gives proof of having performed a difficult computational task (with Bitcoin there is a competition so that only one user gets rewarded with new bitcoins after each transaction block) then gets rewarded with new currency. To say that the user gets rewarded is simply to say that the peer-to-peer network recognizes that additional currency in a given amount has, because of the proof of work, been added to the user’s account of cryptocurrency on the network. The creation of new currency therefore consists in the agreement by the peer-to-peer network to assign the new currency to one or more of its users.

Proof of stake is likewise a consensus mechanism, though in this case the user’s ownership of the cryptocurrency and interactions with its peer-to-peer network over time determine the receipt of newly created currency. The bottom line is that in cryptocurrencies new currency is created through a carefully articulated consensus mechanism by which the peer-to-peer network agrees to assign newly created currency to users. This all happens automatically, the consensus mechanism itself being programmed into the cryptocurrency from the start.

In conclusion, this article has explained all the main ways that money has been created from ancient times to the present. Up until the present section’s treatment of cryptocurrencies, the treatment of the creation of money in this article has been reasonably thorough. Much more can be said about the creation of money within cryptocurrencies (especially the nuts and bolts of proof of work and proof of stake), but this will need to await an upcoming article on cryptocurrencies that I plan to write.

Reference Notes

[1] Jacob Goldstein, Money: The True Story of a Made-Up Thing (New York: Hachette, 2020).

[2] John R. Searle, The Construction of Social Reality (New York: Free Press, 1995). Searle followed up this book with Making the Social World: The Structure of Human Civilization (Oxford: Oxford University Press, 2010).

[3] Hyperinflation is the worst type of monetary failure. The link here is to an infographic showing 152 such failures.

[4] Click on “all data” at to see how gold has varied sharply in price over the last 50 years.

[5] Milton Friedman, The Counter-Revolution in Monetary Theory (London: Institute of Economic Affairs, 1970), p. 24.

[6] Paul Einzig, Primitive Money: Its Ethnological, Historical and Economic Aspects, 2nd edition (Oxford: Pergamon, 1966).

[7] Peter North, Local Money: How to Make It Happen in Your Community (Totnes, Devon: Transition Books, 2010).

[8] Georg Friedrich Knapp, The State Theory of Money, translated from the 1905 German edition (London: Macmillan, 1924).

[9] David Graeber, Debt: The First 5,000 Years, updated expanded edition (London: Melville House, 2014) and Felix Martin, Money: The Unauthorized Biography — From Coinage to Cryptocurrencies (New York: Knopf, 2013).

[10] Harry Markopolos, No One Would Listen: A True Financial Thriller (Hoboken, N.J.: Wiley, 2010).

[11] Kenneth Rogoff, The Curse of Cash: How Large-Denomination Bills Aid Crime and Tax Evasion and Constrain Monetary Policy (Princeton: Princeton University Press, 2016).

[12] Richard H. Thaler and Cass R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness (New York: Penguin, 2008).

Image Notes

[top] Cover image is self-explanatory and comes from two images, one signifying Bitcoin, the other Michelangelo’s creation from the Sistine Chapel, Wikimedia Creative Commons.

[sec 2] Image of alchemist is from a painting by David Tenier (1610-1690), Wikimedia Creative Commons.

[sec 3] Gold prices from 1960 to 2020, both nominal and CPI adjusted, Wikipedia.

[sec 4] Image of gold coins from 1st Century under Vima Kadphises, Wikimedia Creative Commons.

[sec 4] Image of fei taken from Wikipedia.

[sec 7] Photo of Federal Reserve headquarters, taken from Wikipedia.

[sec 8] Screenshot from the homepage of