The Velocity Collapse

Here is the question that has haunted economists since 2009: if the Federal Reserve created trillions of dollars through quantitative easing, why didn’t prices explode? Between 2008 and 2014, the Fed’s balance sheet grew from $900 billion to $4.5 trillion. The monetary base quintupled. And yet inflation averaged just 1.6% during that period — below the Fed’s 2% target. Peter Schiff predicted hyperinflation on Fox News. Ron Paul held up gold bars in congressional hearings. Neither prediction came true.

The answer is velocity — specifically, the velocity of M2, tracked by FRED as M2V. Velocity measures how many times each dollar of money supply is spent to buy final goods and services in a year. It’s calculated simply: GDP divided by M2. If GDP is $28 trillion and M2 is $22 trillion, velocity is about 1.27 — meaning each dollar changes hands roughly 1.3 times per year in the production of goods and services.

That might seem surprisingly slow. But remember, M2 includes savings accounts, money market funds, and CDs — money that’s sitting still. Not every dollar is in active circulation at any given moment. The velocity number captures the average turnover of all money, active and dormant alike. And for decades, that average has been falling.

Era 1: The Stable Range (1959–1987). For nearly three decades, M2 velocity oscillated in a narrow band between 1.65 and 1.85. It was remarkably stable — so stable that Milton Friedman built his entire monetary framework on the assumption that velocity was essentially constant. If velocity is constant, then the quantity equation (MV = PQ) means that increasing the money supply must increase either prices or output. Friedman’s monetarism won him the Nobel Prize in 1976, and it worked beautifully as long as velocity cooperated.

During this era, the money supply and GDP grew at roughly similar rates. When the Fed pumped money in during the inflationary 1970s, velocity stayed put, and prices rose almost in lockstep. When Volcker slammed the brakes in the early 1980s, restricting money supply growth, the economy contracted predictably. The textbook worked.

Era 2: The Rise and Peak (1988–1997). Starting in the late 1980s, velocity broke out of its historical range and started climbing. It rose from 1.77 in Q1 1988 to a peak of 2.20 in Q3 1997 — the highest level ever recorded. Financial innovation was the driver. Money market mutual funds gave savers higher yields, which meant less idle cash. Electronic banking accelerated transactions. The booming stock market pulled money out of bank accounts and into brokerage accounts (which aren’t counted in M2), effectively reducing M2 relative to economic activity.

This was the era when the “velocity is constant” assumption broke down, and the Fed quietly stopped targeting money supply aggregates. Greenspan famously told Congress in 1993 that the relationship between M2 and economic activity had become “unreliable.” The Fed shifted to interest rate targeting — setting the federal funds rate directly rather than trying to control the money supply. It was an implicit admission that Friedman’s framework, while theoretically elegant, no longer described reality.

Era 3: The Great Decline (1998–present). After peaking in 1997, velocity began a decline that has continued for nearly three decades. It fell from 2.20 to 1.97 by 2001 (the dot-com bust pushed money back into safe deposits), then stabilized around 1.95–2.00 for several years. The 2008 financial crisis accelerated the decline: velocity dropped from 1.97 in Q1 2008 to 1.73 in Q1 2009 as the economy contracted and the Fed flooded the system with reserves.

But the crisis decline was just the beginning. QE created trillions of dollars that sat in bank reserves at the Fed, never entering the real economy. Velocity kept falling — 1.64 by 2012, 1.52 by 2015, 1.43 by 2018. Each round of QE pushed M2 higher while GDP grew at a modest 2% pace, driving the ratio inexorably down.

COVID delivered the final blow to any remaining faith in velocity’s stability. In Q2 2020, velocity crashed to 1.128 — by far the lowest reading in the 67-year history of the series. The denominator (M2) was surging as the Fed bought $3 trillion in bonds and Congress mailed stimulus checks. The numerator (GDP) was collapsing as lockdowns shuttered businesses. The result was a ratio that would have been unimaginable a generation earlier.

What does a velocity of 1.13 actually mean? It means that the average dollar in M2 was spent on final goods and services only slightly more than once per year. The entire $18 trillion money supply was turning over at a glacial pace. Americans were saving at rates not seen since World War II. The personal savings rate hit 33.8% in April 2020 — the highest ever recorded. Stimulus checks arrived but had nowhere to go: restaurants were closed, flights were cancelled, retail stores were shuttered. Money piled up.

This is why the $6.2 trillion in money creation didn’t cause immediate hyperinflation. The money existed, but it wasn’t moving. It was sitting in checking accounts, savings accounts, and money market funds, waiting. The velocity collapse acted as a shock absorber, temporarily neutralizing the inflationary impact of the largest money supply expansion in history.

The word “temporarily” is doing a lot of work in that sentence. Because when the economy reopened in 2021, the savings started to flow, velocity ticked up from 1.13 to 1.15 to 1.19, and suddenly $6 trillion in excess money supply met a supply chain that couldn’t keep up with demand. CPI hit 7% in December 2021 and peaked at 9.1% in June 2022. The inflation wasn’t missing — it was delayed.

1. The savings glut. Americans — particularly older Americans — hold far more in savings and money market accounts than they did in the 1990s. Baby boomers accumulated retirement savings throughout the 2000s and 2010s. This money sits in M2 but turns over slowly, dragging down the average. A retiree with $500,000 in a money market fund might spend $40,000 per year, giving that money a personal velocity of 0.08. Mix enough retirees into M2, and the aggregate velocity falls.

2. The financialization of the economy. Financial assets — stocks, bonds, derivatives — generate enormous amounts of “activity” that GDP doesn’t capture. When BlackRock buys $100 million of Treasury bonds, that transaction doesn’t appear in GDP (GDP measures production, not financial trading). But the money used in that transaction sits in M2 before and after the trade. As the financial sector grew from 5% of GDP in 1980 to nearly 9% today, more money went into activities that M2 captures but GDP doesn’t, pushing their ratio down.

3. Excess reserves at the Fed. After 2008, banks parked trillions in reserves at the Federal Reserve rather than lending them out. These reserves are part of the monetary base (and indirectly push up M2 through the banking system) but never enter the real economy. They sit on balance sheets, earning a modest interest rate from the Fed, contributing nothing to GDP. It’s money that exists but doesn’t circulate.

4. Low interest rates suppressed opportunity cost. When interest rates are near zero, there’s almost no penalty for holding cash. In the 1990s, a savings account might pay 5% — enough to encourage people to put money to work. With rates at 0.01% from 2009 to 2022, the incentive to move money out of liquid accounts vanished. People held more M2 because holding M2 cost them nothing.

5. Global dollar demand. Foreign governments, corporations, and individuals hold enormous amounts of U.S. dollars — an estimated 60% of all physical currency and substantial deposits in dollar-denominated accounts. This money is counted in M2 but generates GDP elsewhere (or nowhere, if it’s hidden under a mattress in Buenos Aires). The internationalization of the dollar has permanently depressed domestic velocity.

Velocity has risen modestly from its COVID low of 1.128 in Q2 2020 to 1.41 in Q4 2025. That’s a meaningful recovery in percentage terms — about 25% — but it remains far below the pre-crisis level of 1.94 in 2008, let alone the 1997 peak of 2.20. The recovery has been driven by three forces: the reopening spending boom (2021–2022), the Fed’s interest rate hikes (which raised the opportunity cost of holding cash), and the drawdown of pandemic savings (Americans spent down their excess savings through 2023).

But structural forces suggest velocity won’t return to historical levels. The aging population will keep savings balances high. Financial markets will continue to absorb money that doesn’t appear in GDP. And the Fed’s “ample reserves” framework, adopted in 2019, explicitly commits to keeping excess reserves in the banking system. The old world of scarce reserves and high money multipliers is gone.

The practical implication is profound: the Fed can create far more money than it once could without triggering proportional inflation, because much of that money will sit idle. But this creates a different risk — asset inflation. When trillions of dollars pile up in savings and money market accounts with nowhere productive to go, they chase financial assets instead. Home prices, stock prices, and bond prices all benefited from the combination of abundant money and low velocity. The inflation wasn’t in consumer prices (at least not until COVID). It was in asset prices.