The Quantity Theory as Organizing Principle: MV=PY, Seriously

An Identity That Keeps Showing Up

Every macroeconomics textbook, somewhere in its first hundred pages, prints the same four letters: MV = PY. M is the money stock. V is velocity, the average number of times a unit of money changes hands in financing transactions over a period. P is the price level. Y is real output. Multiply M by V, and you get nominal spending. Multiply P by Y, and you get nominal GDP. The equation holds by definition, since velocity is calculated as PY divided by M. As a matter of pure accounting, it cannot be wrong.

This is precisely why many students dismiss it. If MV = PY is true by construction, what work can it do? You might as well write “revenue equals price times quantity” and call it a theory of the firm. The monetarist answer, developed over more than a century of argument, is that an identity becomes a theory when you add a story about which variables move first, which ones adjust, and which ones are approximately stable. The equation of exchange is not the theory; it is the organizing frame into which the theory is poured.

Understanding how that frame works, where it came from, and why serious economists still argue about it takes us through some of the most consequential debates in monetary economics.

Fisher’s Transaction Version

The modern form of the quantity equation owes most to Irving Fisher, the Yale economist whose 1911 book The Purchasing Power of Money gave the identity its canonical algebraic dress. Fisher wrote it as MV = PT, where T stood for the total volume of transactions, not just final-goods output. Every time a used car changed hands, every time a bond was traded, every time a worker cashed a paycheck, that counted toward T.

Fisher’s version had a clear causal story. The money stock M was determined by the banking system and ultimately by the monetary authority. Velocity V was pinned down by institutional features of the payments system: how often workers were paid, how quickly checks cleared, how much credit was available for bridging gaps between receipts and expenditures. These institutional features changed slowly. Real transactions T were determined by the real economy, by technology and labor and capital. If M, V, and T were all determined independently of the price level, then P had to do all the adjusting. Double the money supply, and prices double. This was the strict quantity theory: money is a veil over real activity, and inflation is always and everywhere a monetary phenomenon, though Fisher would not have used that exact phrase.

The elegance was seductive. But the cracks were visible even in Fisher’s own time. Velocity was not literally fixed; it rose during booms and fell during panics, precisely when you most wanted a stable framework. The boundary between “transactions” and “output” was fuzzy. And the causal claim that M moved independently of P required you to believe that the banking system did not respond endogenously to the price level, a claim that banking theorists from Henry Thornton onward would have questioned.

The Cambridge Cash-Balance Approach

Across the Atlantic, a different formulation was brewing at Cambridge. Alfred Marshall, A.C. Pigou, and later John Maynard Keynes recast the relationship not in terms of the flow of transactions but in terms of the stock of money people wanted to hold. The Cambridge equation is usually written as M = kPY, where k is the fraction of nominal income that people wish to hold as money balances. Algebraically, k is just the reciprocal of velocity (k = 1/V), so the two formulations are mathematically identical. But the intellectual flavor is different.

Fisher asked: how fast does money circulate? Cambridge asked: why do people hold money at all, and how much do they want to hold? The Cambridge approach opened the door to thinking about money demand as a choice variable, influenced by interest rates, uncertainty, convenience, and habit. If k were truly constant, the two approaches would generate the same predictions. But if k responded to the interest rate, as Keynes would later argue forcefully in the General Theory, then the relationship between M and PY became much less mechanical. An increase in M might be absorbed into idle balances rather than pushing up spending and prices.

This was the seed of the liquidity preference revolution that would challenge quantity-theoretic thinking for decades. But it was also, in a quieter way, the seed of the monetarist response. If you could model money demand carefully enough, with the right variables and the right functional form, you could rescue the predictive power of the quantity relationship even in a world where k was not literally a constant.

Friedman’s Restatement

Milton Friedman’s 1956 essay “The Quantity Theory of Money: A Restatement” is one of the most influential pieces in monetary economics, and one of the most carefully misread. Friedman did not claim that velocity was constant. He did not claim that the money supply mechanically determined the price level in every quarter. What he claimed was that the demand for money was a stable function of a small number of variables: permanent income, the expected returns on alternative assets (bonds, equities, physical goods), and a few institutional and taste parameters.

“Stable” is the key word, and it does not mean “constant.” A stable money demand function means that if you know the values of the relevant arguments, you can predict the quantity of real money balances people want to hold with reasonable accuracy. This is an empirical claim, not a definitional one, and it is testable. If money demand is stable, then changes in the money supply that are not matched by changes in the demand-determining variables will show up as changes in nominal spending, which in the short run may split between changes in real output and changes in prices, and in the long run will predominantly affect prices.

Friedman was explicit that his restatement was a theory of money demand, not a complete theory of income determination. He was also explicit that the short-run dynamics could be complicated. Money might affect output before it affected prices. The lags were “long and variable,” as he famously warned. But the long-run anchor was clear: sustained increases in the money supply, beyond what was needed to accommodate real growth, would produce sustained inflation.

This was a more sophisticated position than either Fisher’s mechanical version or the crude “money causes inflation, full stop” that critics liked to attack. It was also a position that generated testable predictions, which is why the monetarist research program produced so much empirical work on money demand, velocity trends, and cross-country inflation comparisons.

Velocity: Constant, Stable, or Unpredictable?

The entire edifice rests on what velocity does. If velocity is constant, the quantity theory is a machine: crank the money supply handle, and nominal GDP moves proportionally. If velocity is stable and predictable, the quantity theory is a useful forecasting tool: you need to model velocity, but the model works. If velocity is unstable and unpredictable, the quantity theory is an empty identity dressed up in theoretical clothing.

The historical record offers ammunition for all three positions, depending on which period and which monetary aggregate you examine.

From roughly the 1950s through the early 1980s, the velocity of M1 in the United States followed a smooth upward trend, driven largely by innovations in cash management and the spread of credit cards. This made M1 velocity reasonably predictable, and monetarist forecasts based on M1 growth performed respectably. The Federal Reserve, under Arthur Burns and then Paul Volcker, paid serious attention to monetary aggregates as guides to policy.

Then the relationship broke. Starting in the early 1980s, M1 velocity became erratic. Financial deregulation, the proliferation of interest-bearing checking accounts, and the spread of money market mutual funds blurred the boundary between “money” and “near-money.” People shifted funds between categories in response to small changes in relative interest rates, and the measured velocity of any particular aggregate became unreliable. The Fed quietly abandoned monetary aggregate targeting. By the Greenspan era, the federal funds rate was the acknowledged instrument, and money supply numbers were afterthoughts in the FOMC minutes.

Monetarists had several responses. Some argued that the problem was the wrong aggregate: if you used a broader or more carefully weighted measure of money (Divisia indices, for example, which weight components by their “moneyness”), velocity was better behaved. Others argued that the instability was a transition phenomenon tied to a specific wave of financial innovation, not a permanent refutation of the framework. Still others conceded that short-run velocity was hard to predict but maintained that over longer horizons, the quantity-theoretic relationship between money growth and inflation held up in cross-country data.

The COVID-era experience added a new chapter. Between early 2020 and early 2022, the M2 money stock in the United States grew by roughly 40 percent, an unprecedented peacetime expansion driven by fiscal transfers, Fed asset purchases, and bank balance sheet expansion. Monetarists predicted that this would produce significant inflation once velocity, which had collapsed during lockdowns, normalized. And inflation did surge, peaking above 9 percent in mid-2022. Critics countered that the inflation had more to do with supply disruptions, energy prices, and fiscal policy composition than with the money supply per se. The debate continues, but the correlation gave quantity theorists their strongest talking point in decades.

The Great Inflation Through the MV=PY Lens

The 1970s remain the most important test case for the quantity theory in practice. The Great Inflation, roughly 1965 to 1982, saw consumer prices in the United States more than triple. What caused it?

Through the MV = PY lens, the story is straightforward in outline if messy in detail. Money growth accelerated in the mid-1960s, partly to finance the Vietnam War and Great Society programs without politically painful tax increases, and partly because the Federal Reserve was reluctant to resist fiscal pressures. The acceleration continued, with fits and starts, through the 1970s. Oil shocks in 1973 and 1979 added supply-side pressure, but monetarists argued that the oil shocks could only produce a one-time shift in relative prices, not a sustained increase in the general price level, unless accommodated by continued money growth. The sustained character of the inflation was, in this reading, fundamentally a monetary phenomenon.

The disinflation under Volcker, beginning in 1979, fit the same frame: a sharp and sustained reduction in money growth, achieved through painfully high interest rates, broke the inflationary momentum at the cost of a deep recession. The short-run output costs were real and large, consistent with Friedman’s warning about long and variable lags. But the long-run result was a return to price stability.

Keynesian and structuralist accounts emphasize different elements: the breakdown of the Phillips curve, the role of wage-price spirals, the institutional collapse of incomes policies, the geopolitics of oil. These accounts are not necessarily incompatible with the monetarist narrative; they tell the story of why money growth accelerated and how the transmission mechanism worked in specific institutional contexts. But the monetarist insistence that you cannot have sustained inflation without sustained money growth remains a powerful disciplinary claim. It forces every inflation story to answer the question: where did the money come from?

A Discipline for Storytelling

This is perhaps the most underappreciated role of MV = PY in professional economics. The equation is not primarily a prediction machine. It is a consistency check. If someone tells you that inflation rose because of greedy corporations, the quantity equation asks: did the money stock or velocity change in a way that permitted higher nominal spending? If not, then higher prices in one sector must have been offset by lower prices elsewhere, or by reduced real output. The story might still be true as a distributional claim, but it is not a story about the general price level unless the monetary side cooperates.

Similarly, if someone tells you that a central bank “printed money” and therefore inflation must follow, the equation asks: what happened to velocity? If velocity fell by the same amount that money rose, nominal spending did not change, and neither did the price level. This is roughly what happened in many advanced economies after the 2008 financial crisis: massive expansions of central bank balance sheets were absorbed into excess reserves and depressed velocity, producing little inflation for years.

The equation does not tell you which story is right. It tells you which stories are internally consistent and which ones have a hole in the accounting. For a field where storytelling often outruns arithmetic, that is a genuinely useful function.

Critiques from the Keynesian and Post-Keynesian Camps

The most basic Keynesian critique is that the quantity theory gets the causation backward, or at least makes it too simple. In a world with an active banking system, much of the money supply is endogenous: banks create deposits when they make loans, and loan demand depends on economic activity, interest rates, and confidence. If money is largely demand-determined, then running from M to PY is like running from the thermometer reading to the weather. The thermometer is correlated with the temperature, but breaking the thermometer does not change the climate.

Post-Keynesians push this further. In their view, central banks do not and cannot control the money supply directly. They set interest rates, and the quantity of money is whatever the banking system creates at that rate of interest. The money supply is a residual, not a cause. In this framework, MV = PY is not wrong, but it is vacuous as a causal story. Velocity absorbs all the action, and the equation reduces to an elaborate way of saying “nominal spending equals nominal spending.”

There is also a measurement critique. Which M? At what frequency? Over what horizon? The answer matters enormously for whether the quantity relationship holds empirically, and monetarists have sometimes been accused of picking the aggregate and the time frame that make their case look best.

Finally, there is the instability critique. Even if money demand was stable in the 1950s and 1960s, it became unstable in the 1980s and 1990s due to financial innovation, and there is no guarantee it will re-stabilize. A theory that works only when the financial system is not changing much is a theory of limited practical value, since the financial system is always changing.

Modern Relevance: QE, M2, and the Search for a Nominal Anchor

The post-2008 era has been a strange laboratory for the quantity theory. Quantitative easing produced enormous increases in the monetary base without corresponding increases in broader aggregates or in inflation, at least initially. This seemed to vindicate the Keynesian view that the money multiplier was unstable and that base money was not the right M. But the COVID-era surge in M2, which did precede a surge in inflation, reopened the question.

One reading is that the transmission mechanism depends on where the money goes. QE after 2008 created reserves that largely stayed within the banking system, boosting asset prices but not consumer spending. Fiscal transfers during COVID put money directly into household bank accounts, boosting M2 and consumer spending simultaneously. The quantity theory works better when the money creation mechanism puts purchasing power in the hands of people who will spend it.

Another reading is that the quantity theory works over sufficiently long horizons, even if it is unreliable quarter to quarter. Cross-country data over decades still show a strong positive relationship between average money growth and average inflation, particularly for countries with high inflation. The relationship is noisier for low-inflation countries, where supply shocks, measurement error, and financial innovation generate more variance. But the basic pattern persists.

Central banks today do not explicitly target money supply growth. They target inflation directly, using interest rates as the instrument and forward guidance as the communication device. But the intellectual infrastructure of inflation targeting owes more to the monetarist tradition than is sometimes acknowledged. The insistence that inflation is ultimately a monetary phenomenon, that central bank independence matters, that expectations must be anchored, and that rules or rule-like behavior is preferable to pure discretion: these are all propositions that trace their lineage through the quantity-theoretic tradition, even if the specific policy instrument has changed.

What Survives

MV = PY is not a theory of everything. It does not explain recessions, financial crises, inequality, or technological change. It is not a forecasting tool that spits out next quarter’s inflation rate. It is an organizing principle that imposes a minimal consistency requirement on macroeconomic stories: if you are going to talk about the price level, you must account for what happened to money and velocity.

For monetarists, this discipline is the point. The equation of exchange is the skeleton on which empirical flesh is hung. Whether the flesh is Fisher’s mechanical causation, Friedman’s stable money demand, or a modern Divisia-aggregate approach, the skeleton holds the story together and prevents it from becoming a collection of anecdotes with no accounting backbone.

For critics, the skeleton is bare precisely because it has no muscles. An identity cannot be a theory, no matter how much institutional detail you drape over it. The real action is in the institutional, behavioral, and political mechanisms that determine how money is created, how it circulates, and how expectations form. These mechanisms are not captured by four letters on a blackboard.

Both positions have merit, which is why the quantity theory has survived its periodic obituaries for more than a century. It is not the last word on inflation. But it remains, for those who use it carefully, a remarkably durable first word.