For nearly a century, the velocity of money—the rate at which a dollar changes hands in the economy—was a cornerstone of monetary theory. When velocity rose, inflation followed. When it fell, deflation loomed. Then, after 2008, velocity collapsed by half and kept falling. The textbooks predicted chaos. The economy shrugged.
The velocity of money, denoted V in the equation MV=PQ, measures how many times each dollar circulates through the economy in a given period. If you earn a hundred dollars and spend it at a restaurant, and the restaurant pays its supplier, and the supplier pays its workers, that single hundred dollars has high velocity. If you earn it and leave it in a savings account, velocity drops.
From the 1960s through 2007, velocity in the United States hovered between 1.5 and 2.2, remarkably stable. The Federal Reserve could predict inflation by watching the money supply and velocity together. Then the financial crisis hit. Velocity peaked at 2.2 in mid-2007 and began a relentless decline. By 2020 it had fallen below 1.1. The Fed had printed trillions in new money—quantitative easing expanded the monetary base fivefold—yet inflation remained stubbornly low for over a decade. The equation that anchored monetary policy for generations had stopped working.
The structural break nobody saw coming
The collapse reflects three deep changes in how money moves through modern economies. First, the explosion of financial assets. When the Fed creates reserves, commercial banks can hold them as excess reserves rather than lending them into the real economy. Before 2008, banks held almost no excess reserves—they lent every dollar they could. After 2008, they parked trillions at the Fed, earning interest with zero risk. Money that sits in a reserve account has velocity of zero.
Second, the rise of what economists call the "portfolio channel." Quantitative easing pushed investors into riskier assets—stocks, corporate bonds, real estate—but those transactions often occur between financial institutions, not between a bank and a household buying groceries. A pension fund selling Treasuries to the Fed and buying equities creates no velocity in the consumer economy. The money circles within finance, inflating asset prices but not the price of bread.
Third, inequality. High earners save a larger share of income than low earners. As wealth concentrated at the top over four decades, an increasing share of each new dollar created stayed in savings, investment accounts, and real estate rather than circulating through consumption. Velocity measures transactions; if the rich hold money rather than spend it, velocity falls even as the money supply grows.
Why the Fed stopped caring
By 2015, Federal Reserve officials had quietly abandoned velocity as a forecasting tool. Internal models that relied on stable velocity relationships were replaced by frameworks focused on unemployment, capacity utilization, and inflation expectations. The equation MV=PQ still holds mathematically—it is an identity, true by definition—but V had become too unstable to be useful.
The collapse also explains why inflation remained low despite massive monetary expansion after 2008, and why it surged so quickly in 2021. The pandemic stimulus checks and expanded unemployment benefits put money directly into the hands of households, bypassing the financial system. Velocity began to recover. Supply chains broke. Suddenly the money was chasing goods again, and the old relationship between money supply and prices reasserted itself—but only because velocity had started moving in the opposite direction.
Our take
The velocity collapse is a reminder that economic models are not laws of nature. They are descriptions of human behavior, and when behavior changes, the models break. The Fed learned this the hard way, but the lesson applies beyond central banking. Any framework that treats a behavioral variable as a constant is a crisis waiting to happen. Velocity fell because the structure of the economy changed—more finance, more inequality, more caution. Those forces have not reversed. The next time someone invokes MV=PQ to predict inflation, ask them what they assume about V, and why. The answer will tell you whether they learned anything from the last fifteen years.
