Understanding Money, the World's Most Powerful Tool

Written by Olga Cooperman

John Maynard Keynes called the gold standard “a barbarous relic”; Alan Greenspan saw gold as “inseparable from economic freedom.” Cathie Wood has praised Bitcoin as “a new paradigm for monetary systems”; Warren Buffett dismissed it as “rat poison squared.” The ECB presents the digital euro as a statement of European sovereignty; the U.S. administration sees a digital dollar as a threat to sovereignty. These sharp disagreements point to a deeper problem: debates about the past, present, and future of money often rest on different assumptions about what money is and how it works. This paper examines these assumptions. Its goal is to provide a clearer basis for judging which forms of money should be trusted, treated with caution, or rejected.

Part 1: What is Money?

The butcher, the brewer, and the baker

Before money existed, people relied on barter, the original and most natural form of exchange. This idea is etched into our system of beliefs. Partly because it seems plausible, and partly because it has been repeated in textbooks for 250 years. In The Wealth of Nations (1776), Adam Smith argues that natural differences in talents lead people to specialize. He imagines a village in which people trade surplus goods directly. He recognizes that the likelihood of two people wanting exactly what the other has at the same time is low. He therefore concludes that the inefficiencies of barter would drive people to adopt a universally desired commodity as a medium of exchange. He put it in these words:

The butcher has more meat in his shop than he himself can consume, and the brewer and the baker would each of them be willing to purchase a part of it. But they have nothing to offer in exchange, except the different productions of their respective trades, and the butcher is already provided with all the bread and beer which he has immediate occasion for… In order to avoid the inconveniency of such situations, every prudent man in every period of society, after the first establishment of the division of labour, must naturally have endeavored to manage his affairs in such a manner as to have at all times by him, besides the peculiar produce of his own industry, a certain quantity of some commodity or other, such as he imagined few people would be likely to refuse in exchange for the produce of their industry.

This classic origin story provided the foundation for what later became known as the commodity theory of money. It is the idea that money emerged naturally from barter and that it derived its value from the material of which it was made. In this view, precious metals were widely accepted in exchange because they had value independent of their use as money. Subsequently, states formalized and regulated monetary metals through official coinage. Fiat currency and credit appeared much later.

“Money is a creature of law”

The first refutation of the canonical origin story appeared in Georg Friedrich Knapp’s The State Theory of Money (1905). Knapp rejected the idea that money has intrinsic value. His central claim is that money becomes money because the state declares it acceptable for settling debts and taxes. If the state requires taxes to be paid in a particular token, people must obtain that token. This requirement alone can make an otherwise worthless object function as money. “Money is a creature of law” is Knapp’s first and most famous line in his book.

Knapp saw legal authority as the essence of money: the state names the unit of account, designates acceptable means of payment, and enforces obligations denominated in that unit. Later scholars proposed that money arose and spread as a result of imperial conquests. David Graeber, in Debt: The First 5,000 Years (2011), calls coinage a military technology for provisioning standing armies. Empires paid soldiers in newly minted currency and then imposed taxes payable in that currency, compelling populations to accept it. This practice likely turned locals into willing military suppliers. Graeber cites archeological evidence that proliferation of coins in the Mediterranean coincides with the rise of professional soldiers. Large markets appear to form around military camps. He concludes that money did not emerge from the need to trade. Rather, money itself gave rise to markets. By providing a standardized, convenient, state-backed means of payment, currency enabled trade among strangers and made specialization viable at scale.

Knapp’s insight that money is fundamentally legal and political did not depose the established commodity theory and its foundational myth. People understood the butcher, the brewer, and the baker’s dilemma more readily than tax liabilities. It also did not help that Knapp published his work during the classical gold standard era. From roughly 1870 until the outbreak of World War I in 1914, all major powers defined their currencies in terms of gold, believing it to be the natural foundation of money.

“Credit and credit alone is money”

Although Knapp rejected the barter origin of money, he offered little explanation of how people traded before state-backed currencies were introduced. This question was taken up by Alfred Mitchell-Innes, who proposed that before money people relied on credit. He presented his argument in two essays, What Is Money? (1913) and The Credit Theory of Money (1914).

Adam Smith believed that if a baker or brewer wanted meat but had nothing the butcher wanted in return, then no exchange could happen. This situation appears to support the need for money as a medium of exchange. But if the baker and brewer are honest, the butcher could simply record what they owe him. When he received bread or beer in the future, that debt would be settled. Because such a system could work on its own, there is no need to assume the existence of a medium of exchange.

What sometimes appeared as barter or commodity money, after closer examination turned out to be a credit relationship. Smith cited an example of nails functioning as money in a Scottish village, but later historians showed he had misunderstood the situation. Nail makers received supplies from dealers on credit and simply repaid their debts with finished nails at their market value. Physical currency was not in use because the system ran on accounting and credit, not on nails. Similarly, colonial laws sometimes allowed things like corn or tobacco to be used to pay debts or taxes. However, their value was measured in official money, and they were accepted only at their market price. There’s no evidence that these commodities were used as a general medium of exchange. The laws simply gave debtors an alternative way to settle what they owed.

Mitchell-Innes emphasized that coinage was not a commodity money but a token of debt. In ancient Greece, coins were made of gold, silver, bronze, or electrum (a gold–silver alloy), and none bore any indication of value. Coins varied so much in weight and metal content that they could not have had a consistent intrinsic value. Roman coins had marks indicating their value, yet their weights were extremely irregular. This suggests that coins were treated mainly as tokens. The important feature was the issuer, which was always indicated. The source of monetary value was the credibility of the issuer, not the metal. Mitchell-Innes noted that coinage was not a state monopoly. In the medieval period, private tokens circulated alongside official coin, and rulers repeatedly attempted to suppress them.

Mitchell-Innes criticized Adam Smith, writing that “no scientific theory has ever been put forward which was more completely lacking in foundation.” Yet he also failed to shake the dominant belief. He was a diplomat rather than a professional economist. Without the academic standing, his ideas were largely ignored. And the world around him did not appear to confirm his claims. He insisted that money was fundamentally credit relations while piles of gold still sparkled in vaults.

We lack direct evidence that could decisively confirm or rule out the commodity, legal, or credit theories of money’s origin. Still, we can test them through a range of anthropological and narrative examples. And even if money’s beginnings remain obscure, we can still ask what makes something function as money. Is there one element that is both necessary and sufficient: material substance, legal authority, or a credible promise?

Argentina’s barter clubs

During Argentina’s 2001–2002 economic crisis, clubes de trueque (barter clubs) emerged as a response to mass unemployment. Initially they were intended for direct exchanges of goods between members. However, barter networks quickly evolved internal credit systems that used vouchers, créditos. Upon joining a club, a new member would receive a small initial allotment of créditos. Then participants were expected to earn most of their créditos by selling something. A voucher had value because members expected other members to take it in payment later. But as the network grew, forgery and over-issuance became a problem. Fake vouchers flooded exchanges, prices exploded, and the system eventually collapsed.

In the context of this grassroots movement, genuine barter proved impractical, and no commodity with intrinsic value emerged to serve as a medium of exchange. What the episode most clearly reveals is the central role of credit in sustaining exchange.

Witch’s Money

Witch’s Money (1939) is a satirical short story by John Collier. In the story, a traveling American painter stumbles upon a charming French village and decides to work there for a while. However, there is no rental market in the secluded village. The artist roams around until he finds a vacant property and offers to buy it for 30,000 francs. The owner accepts a check from the painter, although he’s never seen one before. At the end of the week, he comes back with the check and demands that the painter gives him 30,000 francs. Instead, the painter refers him to the bank, which is a daylong trip from the village.

At the bank, a teller explains to the villager that he needs to open a checking account, wait for the check to clear, and come back in a week to withdraw money. The villager is frustrated but has no choice other than to wait. He comes back a week later and asks again for his 30,000 francs. Since he is taking all the money out, the teller closes the checking account and withholds a small fee for the service. The villager is now furious. After all his troubles, he does not get the nice round sum he anticipated. He returns home and vents to his friends about this gross injustice. They all agree that this is unacceptable. Moreover, they note that the painter has a full book of checks, each evidently worth 30,000 francs. How can an honest person have so much money? The friends kill the painter and split the checks among themselves. Immediately, they start trading with each other. Soon, they start trading with their neighbors. While in circulation, blank checks inspire all sorts of new business projects, arranged marriages, and even illicit activity.

The broke, sleepy village is soon transformed into a place bustling with commerce. The painter’s checks change many hands before making a full circle back to the perpetrators. The story ends with the friends, dressed up in their best clothes, laughing as they enter the bank to cash their checks.

Witch’s Money is a fictional monetary experiment. Before the stranger’s arrival, the money-poor village operated on deferred settlement and trust. The painter’s check is immediately understood as a promise to pay later, and the landlord accepts it without hesitation. The new “money” that enters circulation has no intrinsic value. No legal authority enforces its acceptance. But crucially, villagers believe that the bank stands behind blank checks and will readily convert them into external money. After the painter’s checks began changing hands, the village economy expanded dramatically. No new resources entered the system, yet transactions multiplied. The only change was the introduction of money itself.

The story suggests that money is fundamentally a credit network sustained by trust in future settlement. Confidence that a recognized authority (the bank) stood behind the blank checks made them more readily accepted than personal promises and increased the pace of trade.

Limestone disks as money

In The Island of Stone Money (1991), Milton Friedman recounts the peculiar practices of the people of Yap. The small island in Micronesia had a population of a few thousand at the time. In their close-knit society, wealth was represented by massive stone disks known as rai. These stones were often too large to move and remained permanently in place. Transactions occurred not through physical transfer but through collective acknowledgment within the community. The stones were quarried on distant islands and brought to Yap on bamboo rafts pulled behind canoes. In one case, a stone was lost at sea during transport. Yet it was recognized on the island as a good delivery. Everyone believed it existed and was beautiful. And they knew who owned it.

Friedman’s essay presents a fascinating case study: what guaranteed the value of rai and why these stones retained value even when they were lost at the bottom of the ocean? The Yap example challenges the claim that money must primarily function as a medium of exchange. Instead, it shows that recognized ownership matters more than circulation. It also suggests that value did not come from the material of the stones as they had little practical use. Nor did it come from state or authority because no one enforced them. Arguably, the value of rai rested on social relationships within the island. The entire community agreed on the stones’ existence, their ownership, their relative importance. Even lost stones retained value at the bottom of the ocean because their existence and ownership remained socially recorded.

Limestone disks probably functioned as markers of social status and reputation. Ultimately, they signaled the owner’s creditworthiness. The Yap economy can therefore be understood as a system of social credit backed by personal reputation expressed through ownership of rai.

Gold as money

Friedman used the Yap story to show that even gold-based monetary systems ultimately depend on shared belief and accounting records rather than material value and physical possession. By the mid-twentieth century, most of the world’s monetary gold was held in the United States and rarely moved from the underground vault of the Federal Reserve Bank of New York. Instead, nations relied on mutual recognition of claims to gold resting in bedrock beneath Manhattan. This arrangement was not so different from a stone lying at the bottom of the ocean but still recognized as a valid claim to wealth on Yap. What sustained both systems was not the underlying asset itself, but shared confidence in its value and the record of ownership.

Modern advocates of the gold standard argue that tying money to gold anchors it to something real and keeps governments from excessive issuance. They often regard owning gold as a safeguard against unstable fiat currency. The opposing view is that money is not a thing with value in itself, that tying it to gold can be economically constraining, and that holding gold as an asset carries its own risks. Which view is better supported by the evidence?

The modern gold standard emerged during the Industrial Revolution to support international trade. Britain was already on gold, and other nations followed the lead of the world’s dominant financial power. National currencies were defined as fixed quantities of gold, and central banks promised to convert paper money into gold at that rate.

During World War I, governments were forced to suspend convertibility and print money to finance expenditures. Currencies became overvalued and simply returning to pre-war gold parity would surely cause runs on gold. To defend gold reserves, governments raised interest rates. The logic was that sufficiently high interest rates would compensate currency holders for the risk of devaluation. For example, if investors expected a 10% devaluation but could earn a 12% interest rate on holding the currency, they would have little incentive to convert it into gold. But high interest rates discouraged investments, and, consequently, caused mass unemployment, weak demand, deflation, and depressions.

When Britain left gold in 1931, the pound fell and its exports became more competitive. This boosted domestic producers and helped economic recovery. Other nations followed, triggering competitive devaluations and a period of global monetary disorder. The Bretton Woods agreement (1944) attempted a partial restoration of the gold standard to stop price wars. The U.S. dollar became the anchor currency, with other nations pegging their exchange rates to the dollar rather than directly to gold. The United States, in turn, promised to convert dollars into gold at $35 per ounce, but only for foreign governments and central banks. But as foreign holders, notably France, increasingly demanded gold, the system became unsustainable once again. In 1971, convertibility was terminated, marking the beginning of the modern fiat currency era. Since then, money has no longer been tied to gold or any other commodity.

The scale of modern economies far exceeds what global gold reserves could support, even symbolically. Yet gold continues to be regarded by many as a superior monetary asset and a safeguard against inflation. But would it truly provide security in periods of financial or political instability?

Monetary gold ownership takes several forms. Iconic gold coins such as American Gold Eagle, Canadian Gold Maple Leaf, or South African Krugerrand are official money issued by their respective states. Yet they function poorly as a medium of payment. Imagine offering a Gold Eagle to a doctor or a plumber for their service. They would likely refuse this form of payment, unable to value the coin on the spot. They could also be discouraged by price volatility and the need to pay tax because legally gold is an asset. In a genuine crisis it is even more uncertain who would accept them. Gold coins are typically easy to acquire at a premium but can usually be sold back only to brokers, and often at a significant discount. Gold bullion bars introduce different frictions. Their ownership involves significant costs for secure storage, insurance, and verification of authenticity. In practice, bullion is not accepted as payment anywhere. In any case, its lack of divisibility would make it unusable for ordinary purchases.

A more abstract form of gold ownership is a claim on bullion held in an official repository. Bullion banks and dealers typically sell unallocated gold, meaning ownership is linked to a stated quantity rather than to specific bars identified by serial number. These institutions often operate on a fractional-reserve basis, issuing paper claims that exceed the amount of physical gold held in vaults. Because these trades are conducted over the counter, precise figures are difficult to know. However, various estimates suggest that paper claims may exceed physical gold by ten to one hundred times. Physical delivery is generally possible under normal conditions, but it could become unlikely if many investors simultaneously demanded their gold. What is really owned is not the metal itself but a financial claim that would likely be settled in fiat currency. Gold exposure through futures and exchange-traded funds extends this abstraction even further. These instruments primarily track the price of gold rather than provide access to the metal and may not be fully backed by physical holdings.

The realities of direct ownership suggest that gold is illiquid in practice and hardly usable as money. Indirect ownership, meanwhile, amounts to a speculative investment in financial instruments that reference gold yet ultimately settle in fiat currency.

Conclusion

We looked at barter club credit vouchers, the traveling painter’s blank checks, the limestone disks of Yap, and the modern gold standard to probe the true nature of money. Together, these examples suggest that money is not a physical object with intrinsic value, nor does it have to serve as a literal medium of exchange. They show that money can arise socially, without a central authority, even if authority often gives it scale, liquidity, and stability. Above all, they reveal money’s credit nature.

The fact that “credit and credit alone is money” is often obscured by our everyday experience. Modern monetary systems are built on layers of liabilities and settlement claims. But because this architecture works so seamlessly, money can appear to be a thing in itself. It is not. A bank deposit is not cash sitting in storage but a legal claim on the bank. Cash itself is a direct claim on the central bank. We now turn to how this layered network of claims, which we call money, works from first principles.

Part 2: How Money Works

During the Global Financial Crisis, Ben Bernanke, then Chair of the Federal Reserve, was asked whether the money used to bail out banks was “tax money.” He replied: “It’s not tax money. The banks have accounts with the Fed, much the same way that you have an account in a commercial bank. So, to lend to a bank, we simply use the computer to mark up the size of the account that they have with the Fed. It's much more akin to printing money than it is to borrowing.” Yet in another interview, he said: “One myth that’s out there is that what we’re doing is printing money. We’re not printing money. The amount of currency in circulation is not changing.”

Bernanke’s remarks are puzzling. He rejects the idea that the bailout funds came from tax revenue, and he does not frame them as borrowing. He suggests that the operation is like printing money, while also insisting that no additional currency is entering circulation. It appears that Bernanke is working with a different model of money than the one most people implicitly assume.

To make sense of his remarks and to understand the monetary architecture, we will examine the respective roles of private banks, the central bank, and the government in the economy. We use the United States as our example, but the same basic structure operates across virtually all advanced economies. As with the nature of money itself, there are multiple theories about how these institutions function and interact. We will review those perspectives, trace the traditions from which they came, and test them against a central question: which framework best explains money from the point of creation rather than assuming an already existing stock of money in circulation?

The Role of Private Banks

Financial Intermediation Theory

The conventional view is that banks are financial intermediaries that channel funds from savers to borrowers. In this model, savings come first. Households save, banks gather and pool those savings and then lend them to those who need financing. This understanding of banking, though very familiar, is relatively recent. It emerged in the 1960s as a logical consequence of general equilibrium theory.

Historically, the discipline of economics had been largely a blend of moral philosophy and policy analysis, concerned with wealth, society, and governance. But in the nineteenth century, when the exact sciences gained enormous prestige, economists aspired for comparable status. They especially looked up to classical mechanics for its precision and rigor. But economics lacked a formal theory and a mathematical framework with systems of equations.

Léon Walras first proposed general equilibrium theory in 1874. He tried to represent the economy as an interdependent system in which markets and participants are linked through prices. He defined equilibrium as the set of prices and quantities that balance supply and demand across the whole economy. Walras wrote down a system of equations, but he lacked mathematical tools needed to prove rigorously that such a system had a solution. It took eighty years of development across several fields before the existence proof finally arrived in 1954. Over the following years, the theory spread widely and profoundly reshaped the discipline.

At its core, general equilibrium is a theory of how prices allocate scarce resources across the economy. As this framework took hold, it also became natural to describe banks as intermediaries between savers and borrowers. If the economy is understood primarily as a mechanism for coordinating the supply and demand of scarce resources, then lending appears as the transfer of money from one person to another. “Fidelity Fiduciary Bank,” a song from Walt Disney’s 1964 film Mary Poppins, shows how deeply this view entered popular culture. The song explains how two pence deposited in a bank will fund “railways through Africa” and “dams across the Nile.”

A key flaw of the financial intermediation view is that it never truly explains how the first money entered the economy. Lending requires prior savings, and savings require prior income, which comes from employers who must first borrow in order to invest in their businesses. This circular dynamic explains how existing funds are reallocated, but it pushes the question of origination backward rather than answer it.

Fractional Reserve Theory

For centuries, banks had operated on a fractional reserve basis. They issued claims in excess of the gold stored in their vaults, expecting that not all customers would seek redemption at once. However, a more specific formulation of fractional reserve theory was given by Chester Phillips in his 1920 book Bank Credit. He proposed a “money multiplier” model of banking. Each individual bank is only a financial intermediary: it takes deposits and lends them out but does not itself create money. However, money creation happens at the level of the banking system as a whole. As deposits are re-lent and spread through the system, the combined activity of many banks leads to an overall expansion of the money supply. Consider an initial deposit of $1,000 at Bank A under a 10% reserve requirement. Bank A keeps $100 in reserves and lends out $900. That $900 is deposited in Bank B, which keeps $90 in reserves and lends out $810. The $810 is then deposited in Bank C, and the process continues in the same way. In total, the banking system can generate about $10,000 in deposits by re-lending the original $1,000. Around the same time, the Federal Reserve was seeking to understand how central bank actions affected the broader economy. The money multiplier offered a simple model linking the requirement to keep a certain fraction of deposits as reserves to the total money supply in the economy.

For decades, this theory influenced both economic teaching and monetary policy. Then, during the global financial crisis, it became clear that reserves do not mechanically govern bank lending or deposit expansion. The Bank of England has explicitly stated that the money multiplier does not provide an accurate account of money creation, while the St. Louis Fed has recommended retiring the concept from classroom teaching. Today, in most advanced economies, banks are no longer required to hold a fixed fraction of deposits as reserves. Since March 2020, the Federal Reserve has set a reserve requirement of 0%. The United Kingdom has no reserve requirement at all. The European Central Bank retains a 1% minimum reserve requirement, though it is generally not regarded as a meaningful constraint on lending.

Credit Creation Theory

There is a third and the oldest theory about the functioning of the banking sector. This was the default understanding from the late 1800s to the 1920s. It states that each individual bank creates new money when it makes a loan.

When banks still issued physical banknotes, the credit creation theory was easier to observe in practice. A bank made a loan by handing the borrower its own notes. Depending on the period and country, private banknotes were redeemable in gold, silver, or state-issued currency. In everyday use, however, they often circulated as a means of payment in their own right and were only rarely converted into other forms of money.

Once banknotes were replaced by electronic accounting records, the process of money issuing by banks became less visible. Then central banks gained monopoly over the issuance of physical currency. It became easier to think of money creation as something done exclusively by the state. Commercial banks were seen as secondary players, merely moving around existing funds.

Since the global financial crisis, the credit creation theory of banking has undergone a remarkable revival. The crisis highlighted that macro-economic models treating banks as simple intermediaries can miss the very source of financial instability. Central bankers increasingly concluded that commercial banks create money by making loans, rather than lending out pre-existing deposits.

In the academic literature, Richard Werner’s 2014 paper presented evidence in favor of the credit creation theory. Werner arranged to take out an actual loan from a bank and then closely observed what happened inside the bank’s accounting system. He also interviewed staff to understand the operational process. According to his findings, the bank did not transfer existing funds, reduce another customer’s balance, or wait to obtain reserves before crediting his account. No prior pool of savings was used. Instead, the bank simultaneously recorded two new entries: a loan as an asset and a deposit in his account as a liability. Werner concluded that deposits are not a prerequisite for loans. Rather, deposits are created in the process of lending.

The three theories of banking, however, are not entirely mutually exclusive. The intermediation view still captures the role banks play in channeling purchasing power across time. Likewise, while the money multiplier is not an accurate description of how lending works, it illustrates monetary expansion. Still the credit creation view provides the most important insight, because it explains how money enters the economy through lending.

When a bank makes a loan, it creates a new deposit in the borrower’s account. This operation is a zero-sum game for the bank: the client owes borrowed money to the bank, and the bank owes deposit money to the client. Yet a new deposit instantly increases the total money supply in the economy. When the loan is eventually repaid, the deposit is extinguished and that money is destroyed. Bank deposits are commonly, but mistakenly, understood as prior savings. In reality, for nearly every deposit there is a corresponding debt. As Hartley Withers noted in The Meaning of Money (1909), much of money deposited in banks is not preexisting cash saved by the public, but bookkeeping credit.

The Role of Central Banks

Trust Anchor in a Two-Tier Monetary System

In a 2022 BIS speech, Hyun Song Shin said, “If central bank money did not exist, it would need to be invented.” The sentiment may sound self-serving coming from within the central banking world, but it captures an important idea. The “traveler paradox” illustrates this idea.

The story takes place in a small, impoverished town where everyone is in debt. A traveler arrives and decides to stay the night at a local hotel. He places cash on the counter and goes upstairs to inspect his room. The hotel owner immediately uses the cash to repay a debt to his chef. The chef rushes to settle his bill with the grocer. The grocer pays the doctor. The doctor pays the nurse. The nurse, in turn, uses the same money to settle a past debt to the hotel, where she once stayed. The cash has now traveled full circle. At this point, the traveler comes back downstairs. He decides the hotel is too shabby and asks for his money back. The hotel owner returns the cash, and the traveler leaves. The money left the town, but all debts have been extinguished.

If all the debts in the town ultimately cancel one another out, why couldn’t the residents settle them without passing around the traveler’s cash? The answer is trust. Each resident owed one neighbor and was owed by another. On a net basis, they were flat: neither burdened by net debt nor in possession of net money. But the claims they held on others could not be passed along in payment. These were personal claims, tied to the trust in specific individuals. A nurse with a claim on the doctor might not accept, in repayment, a claim on the grocer. The traveler’s cash solved that problem. Unlike private claims, it was accepted by everyone as a common and trusted means of settlement.

Commercial banks with the central bank at the middle operate on much the same principle. To bank customers, it appears that money held on deposit can be used to pay bills, buy goods, or send funds across the world. But deposits are private claims issued by commercial banks. They are like gift cards issued by different department stores, may carry the same face value, but each ultimately depends on the credibility of its issuer. Deposits are not themselves acceptable for settling obligations between banks. Interbank payments require central bank money, which serves as the system’s common high-trust settlement medium. Without it, deposits would not reliably function as universal money. The banking system would start to resemble the impoverished town: a web of offsetting debts, but no dependable way to complete payments, leaving the economy frozen.

The central bank is the institution that issues the country’s official currency. In legal terms, the country’s currency is a liability of the central bank. What does the central bank owe people in return for money? In a fiat system, it does not promise gold. The key promise is legal acceptability: the currency must be accepted for taxes and for settling debts.

The electronic form of central bank money is called reserves. These reserves are held in accounts that commercial banks keep at the central bank. Banks may convert part of their reserves into physical cash for use in branches and ATMs. How do banks acquire reserves in the first place? One way is by exchanging high-quality assets they hold for newly issued reserves. Another is by receiving reserves from other banks through interbank settlement, or from the government through the payment of salaries to government employees into commercial bank accounts.

In this two-tier monetary system, commercial banks create money dynamically in response to customer demand for loans. The central bank, in turn, ensures that payments move smoothly throughout the system. By doing so, it makes money appear to be a single, unified thing rather than a patchwork of different claims.

The Role of Governments

By this point, it should come as no surprise that there are different views on the government’s role in money issuance and control.

The Classical Understanding of Government Finance

Each fiscal year, the government decides how much it will spend on public services and how much it will collect in taxes. When government spending exceeds tax revenue, a budget deficit results. To pay for the deficit, the federal government borrows money by selling bonds to the public. The national debt is the accumulation of government borrowing. The U.S. has carried debt since its inception. As the government experiences recurring deficits, the national debt grows. Recent events triggering large spikes in the debt include the Afghanistan and Iraq Wars, the 2008 Great Recession, and the COVID-19 pandemic. Over the past 100 years, the U.S. federal debt has increased from $381 billion in 1925 to over $37 trillion in 2025.

Many politicians and economists support reducing government debt and keeping budgets balanced. This position is rooted in liberal values such as prudence, frugality, and personal responsibility. Benjamin Franklin, one of America’s Founding Fathers, warned against living beyond one’s means, saying, “Rather go to bed without dinner than to rise in debt.” In the twentieth century, economist John Maynard Keynes challenged this strict balanced-budget view. He argued that during recessions governments should run deficits to support demand and reduce unemployment. However, he also maintained that governments should run surpluses during periods of economic growth, so that budgets would balance over time.

In the heart of New York City, a glowing digital billboard tracks the total U.S. debt, the population, and the debt per capita. As of early 2026, that per-person figure hovers around $115,000. The National Debt Clock was erected to raise awareness about government borrowing. It implies that every American will ultimately pay the price, one way or another. It frames the national debt as a burden on future generations.

Excessive debt is sometimes portrayed as an existential threat. A common fear is that the government may be unable to refinance maturing debt, leading to default and collapse of the national currency. To prevent such outcomes, authorities might resort to “printing money,” fueling inflation and eroding people’s savings. Moreover, as debt grows, a larger share of the budget goes toward interest payments rather than public services. Some argue that heavy reliance on foreign lenders could result in creditors effectively “owning” the country. In more extreme narratives, a looming debt crisis is described as deliberately engineered to reset the financial system. It would be used as a pretext, they claim, to replace cash with a central bank digital dollar in order to expand surveillance and restrict how people use their money.

Modern Monetary Theory

In 1993, Warren Mosler published Soft Currency Economics, which he described as “an effort to provide insight into the fiat monetary system”. Working as a bond trader, Mosler struggled to reconcile how financial markets operated with the way politicians talked about government budgets. He observed that the federal government never struggled to finance wars. It simply authorized spending, and funds were credited to the accounts of beneficiaries. Mosler asked why there is persistent rhetoric about an inability to fund essential public services? He concluded that the problem is widespread misunderstanding of how modern fiat currency systems function. This led him to develop the framework that later became known as Modern Monetary Theory, or MMT.

MMT begins from the premise that a sovereign government is the issuer of its own currency. As the monopoly issuer, it does not need to raise money through taxation before it can spend, nor can it run out of its own currency. Just as a soccer referee can always award more points, a currency-issuing government can always create more money.

Federal taxes therefore do not finance infrastructure, defense, scientific research, and Social Security. Their main function is to create demand for the national currency. Taxes play another important role. When the government spends, it injects money into the economy. This is not an episodic occurrence, but a structural dynamic: public salaries, entitlements, contracts, etc., all are paid with newly issued money. Taxing some of this spending reduces the supply of money. One can imagine bringing a bag of cash to Washington D.C. to settle a tax bill. A Treasury clerk accepts the payment, records it, and then shreds the bills. This may sound bizarre, but the effect of taxes is extinguishing money from the system. A useful analogy is a department store that issues its own gift cards. When customers redeem the cards, the store simply destroys them. In the hands of the issuer, these cards are worthless paper.

In MMT’s framing, budget deficits are not seen as a problem. A deficit simply means the government has spent more money into the economy than it has taken back in taxes. The difference remains in the hands of households and businesses. Mechanically, when the government runs a deficit, it covers the gap by selling bonds. These bonds serve two purposes. First, they drain the money that has not been collected back in taxes, reducing the amount of cash circulating. Second, they provide a safe place for people to earn interest. Should their repayment be a source of concern? Within this framework, not particularly. A government bond can be understood as money moved from an account that does not pay interest into one that does. When the bond matures, the government simply flips the funds to a regular account. Accordingly, government debt is not a burden, but a record of accumulated savings held by households, retirement funds, insurance portfolios, and institutions. Seen this way, the National Debt Clock could just as well be called the National Wealth Clock.

Government finance is commonly understood as a “tax-and-spend” system, in which the state must first collect revenue before it can spend. MMT proposes instead a “spend-and-tax” framework. This change in perspective has significant policy implications. From this standpoint, unemployment signals that the government has not spent enough to mobilize available labor. The proposed shift reframes public debate. Instead of asking, “How will we pay for it?” the central question becomes, “Do we have the real resources?” Ultimately, MMT argues that the true limit on government spending is available labor and raw materials, not tax revenue. Mosler wrote,

We are told that national health care is unaffordable, while hospital beds are empty. We are told that we cannot afford to hire more teachers, while many teachers are unemployed. And we are told that we cannot afford to give away school lunches, while surplus food goes to waste.

Does MMT encourage reckless spending? MMT’s own answer is no. The real constraint on government spending is not the size of the deficit or the level of debt. It is inflation. If the government spends into an economy that has idle workers and unused capacity, that spending can increase output without pushing up prices. More money meets more goods and services. But if the economy is already operating near its limits or if it cannot mobilize resources (e.g., as during COVID-19 lockdowns), then additional spending does not create more products. It increases demand for what already exists, driving prices up. MMT therefore frames fiscal discipline as a balancing act: spend to achieve full employment, reduce spending if inflationary pressure builds.

It is important to emphasize that MMT principles apply to sovereign currency issuers that borrow in their own currency. Local governments do not create money; they use it. Provinces, states, and cities must raise revenue through taxes, fees, or borrowing in order to fund public services. If they fail to manage their budgets responsibly, they can default, much like a household or a business.

Warren Mosler developed his ideas alongside Randall Wray and Bill Mitchell. They described MMT as “a factual account of how sovereign money creation actually works.” MMT remains an influential heterodox framework, but it has not become a mainstream macroeconomic theory. While many economists do accept that government spending creates new reserves and deposits, they push back on the claim that this amounts to largely unconstrained money creation. They reject the view that large deficits are normal and generally safe, arguing that MMT underestimates both the political and the practical difficulty of controlling inflation. They are also critical of how it treats government spending and taxation as the primary tool for managing employment and price stability.

In the mainstream view, the economy is guided mainly through central bank policy rather than through government deficits. The government’s role is to provide public services and to keep the budget under control. The central bank is usually given a dual mandate: to support full employment and keep inflation low and stable. It tries to do this mainly by setting interest rates to affect borrowing, investment, and private sector hiring.

Putting It Together: A Toy Economy from Scratch

We can now combine insights from Modern Monetary Theory and credit creation theory to show how a new economy could get off the ground without any pre-existing stock of money. From MMT, we take the idea that the government is the initial mover, bringing money into existence through spending and generating demand for it by requiring taxes to be paid in its own currency. Banks then create additional money through lending. This privately created money is seamlessly integrated into the economy because it can be used across the banking system and settled in the government’s currency.

Day 1: No Prior Money in the System. A central bank is established, and several commercial banks are chartered. The government then begins spending, hiring employees and paying them through accounts at commercial banks. These payments create deposits for the recipients at their commercial banks and corresponding reserves for those banks at the central bank.

Day 2: Commercial Bank Activity Begins. Commercial banks start issuing loans, creating new customer deposits in the process. These loans represent claims on future economic activity, linking money creation to the production of goods and services. Bank money is elastic because it can expand and contract in response to demand for credit.

Day 3: Central Clearing Is Established. When two customers of the same bank pay one another, the bank can simply adjust their account balances internally. But when a payment crosses from one bank to another, the banks must ultimately settle in reserves. Since deposits are far larger than reserves, banks cannot settle every interbank payment on a gross, one-by-one basis. Instead, they offset payments flowing in opposite directions and settle only the net differences in reserves. A clearing house coordinates this process by recording, matching, and netting payment obligations among banks before final settlement in central bank money.

In recent data around 2025, U.S. banks held roughly $3 trillion in reserves at the central bank alongside about $18 trillion in customer deposits across some 4,400 banks. If one bank ended the day short of reserves, it could borrow them from another bank. When the whole banking system needs more reserves, the central bank would supply them with a few keystrokes. The lubricant of reserves helps keep the payment system running smoothly and ensures that one dollar is treated as one dollar, no matter which bank it comes from. This singleness of money is an important design feature. Without it, money stops being truly money and starts behaving like a collection of risky private promises.

Deconstructing Bernanke’s Remarks

We can now return to Bernanke’s claim that the rescue of the banking system involved neither taxpayers’ money nor new money in circulation. One prominent example was the $700 billion authorized under the Troubled Asset Relief Program (TARP). So where did that money come from?

First, the U.S. government issued bonds and sold them to investors such as banks, pension funds, and corporations. These investors paid for the bonds out of their commercial bank deposits. Their banks debited those deposits, and the payments were settled through transfers of reserves to the Treasury General Account, or TGA, at the Federal Reserve. This moved reserves out of the banking system and into the TGA. The government then spent from the TGA to purchase preferred shares in troubled banks. As it did so, reserves moved back into the banking system.

Net Result (System Level):

Investors converted bank deposits into safer government securities.
There was no change in reserves: reserves drained into the TGA during debt issuance and returned to banks during TARP spending.
Banks received fresh capital.
No new money (neither deposits nor reserves) was created, and no taxpayer funds were used.

Bernanke’s remarks no longer seem paradoxical once money is understood not as a fixed stock that governments and banks merely distribute, but as a layered system of liabilities that can be issued, transferred, and extinguished in different ways.

Conclusion

We are trying to clarify three questions: what money is, how it began, and what makes something function as money. We may never know with certainty how money first emerged in human societies. But whatever its origins, money is best understood functionally as a credit relation. In modern economies, institutions such as commercial banks, central banks, and governments give this credit-based money trust, scale, and stability. We next turn to emerging forms of money, including cryptocurrencies, stablecoins, and central bank digital currencies, and ask how they fit, or fail to fit, within this framework. Finally, we return to the deepest question of all: what money is, philosophically speaking.