The Aggregates Reclassified: What the 2020 M1 Redefinition Actually Changed

A Definition Changed, Not the Money Supply

In May 2020 the M1 money supply in the United States jumped by roughly 150 percent in a single monthly report. No one printed a wave of new currency and no bank suddenly created trillions in new deposits. What changed was the definition. The Federal Reserve reclassified savings deposits, moving them from M2 into M1. Understanding why that reclassification happened, and why it barely matters for the economy even though it matters enormously for reading a chart, is the real lesson buried in the aggregates.

What M1 and M2 Actually Count

Before 2020, M1 was the narrowest official aggregate: physical currency in circulation, plus demand deposits (ordinary checking accounts), plus a few other checkable deposit types. M2 was broader, adding savings deposits, small time deposits under 100,000 dollars, and retail money market mutual fund balances.

The distinction used to track something real: how easily a dollar could be spent right now. Currency and checking balances were considered fully liquid. Savings accounts were technically less liquid because federal rules under Regulation D limited the number of transfers or withdrawals a saver could make from a savings account each month, typically six.

That regulatory limit is what actually separated M1 from M2 for decades. It was not a description of how people used their money. It was a description of a banking rule.

Why the Line Moved in 2020

In April 2020, as part of the broader emergency response to the pandemic, the Federal Reserve's Board suspended the six-per-month transfer limit on savings accounts under Regulation D. That rule had originated in reserve requirement mechanics, and by that point in time reserve requirements themselves were being set to zero for all depository institutions.

Once the transfer limit was gone, the technical justification for keeping savings deposits out of M1 disappeared. The Fed's data staff reclassified savings deposits into M1 starting with the May 2020 report, backfilling the series historically so analysts could compare like with like. M2 barely changed, since savings deposits had already been part of it. M1 exploded, on paper, because it absorbed an enormous pool of savings balances that had existed all along.

This is the point worth sitting with: the redefinition tells you something about banking regulation and statistical bookkeeping. It tells you almost nothing about how much spendable money existed in the economy on either side of the reporting change.

What Velocity Is Actually Measuring

Velocity of money is calculated as nominal GDP divided by a money stock, usually M2. It is often described as how many times a dollar changes hands in a year, but that is a loose way of putting it. It is really a ratio: how much economic output the recorded money stock is associated with, given how that stock is defined.

Because velocity is a ratio, anything that changes the denominator without changing real economic activity will move velocity mechanically, with no economic meaning attached. The 2020 M1 redefinition did exactly this to M1 velocity, which fell sharply that month for pure accounting reasons. M2 velocity was affected far less by the redefinition itself, but M2 velocity had already been declining for decades before 2020, and that decline was real, not an artifact of reclassification.

Why Velocity Fell for Decades

From the early 1980s through the 2010s, M2 velocity trended downward with a few interruptions. Several structural forces drove this. Interest rates on savings and money market instruments fell over that period, especially after the 2008 financial crisis and again after 2020, which reduced the opportunity cost of holding money in low-yielding deposits rather than spending or investing it elsewhere. Banks and money funds also became more efficient at offering interest-bearing instruments that still counted as "money" under M2's definition, so more of the aggregate sat idle relative to GDP. Financial deepening, an aging population with higher savings rates, and a long stretch of subdued inflation expectations all pushed in the same direction: more money held per dollar of annual output, meaning lower velocity.

The practical result is that the same dollar of M2 was associated with steadily less nominal GDP as the decades passed. A snapshot of the money stock alone, without velocity, told you increasingly little about spending in the economy.

Money Growth and Inflation: The Mechanics, Not the Slogan

The quantity equation, MV equals PY, where M is the money stock, V is velocity, P is the price level and Y is real output, is an accounting identity, not a causal law. It is always true by construction. The slogan version, "more money printed always means more inflation," treats V and Y as constants and assumes M drives P directly. Neither assumption held up well across the actual data of the last several decades.

Between 2008 and 2015, the Federal Reserve's balance sheet and bank reserves expanded enormously through quantitative easing, yet consumer price inflation stayed persistently below the Fed's 2 percent target for most of that period. Velocity fell hard enough, and banks' willingness to lend against those new reserves stayed subdued enough, that the extra base money did not translate into a proportional surge in broad money spent on goods and services.

By contrast, 2020 to 2022 combined very large increases in M2, driven by pandemic fiscal transfers that were deposited directly into household and business bank accounts, with a temporary collapse and then partial rebound in velocity, and with real supply constraints on goods, energy and labor. Inflation did rise sharply, peaking in 2022, but attributing that entirely to "money growth" ignores the fiscal transmission channel, the supply-side shocks, and the shift in velocity that occurred as spending patterns normalized after lockdowns eased. Money growth was a necessary part of the story, but it was not a standalone predictor.

What to Take Away

The M1 and M2 aggregates are accounting categories built around banking regulations and liquidity conventions, and those conventions changed materially in 2020 when Regulation D's transfer limit was suspended. Velocity is not a fixed constant; it moves for real economic reasons, like interest rates and savings behavior, and it can also move for purely definitional reasons, as the 2020 M1 series shows. The quantity equation is always arithmetically true, but using it to forecast inflation from money growth alone requires assuming velocity and output are stable, which the historical record from 2008 through 2022 does not support. Reading the aggregates usefully means checking what changed in the plumbing before drawing conclusions about what changed in the economy.