M2 is at $23.3 trillion. Strip out retail money market funds and the rest of the money supply is smaller than in January 2022.

M2, the Federal Reserve's broad money aggregate, reached $23,342.8 billion in August 2026, per the H.6 Money Stock Measures release of September 22, 2026. Take out retail money market funds, and the rest of M2 is $35.7 billion smaller today than it was in January 2022, before the Federal Reserve began raising rates that March; the first increase came March 16, 2022. Retail money market fund balances — the RMFSL component of M2, held in funds accessible to individual investors — stood at $3,073.9 billion in August 2026, up from $1,342.8 billion in January 2022. The gain: $1,731.1 billion, a 128.9 percent increase over four and a half years. Retail MMFs now represent 13.2 percent of M2, up from 6.2 percent in January 2022 — a share that more than doubled. Subtract RMFSL from M2SL and the remainder — M1 plus small-denomination time deposits — was $20,304.6 billion in January 2022 and $20,268.9 billion in August 2026: a decline of $35.7 billion, or 0.18 percent. M2's total gain above the January 2022 baseline was $1,695.4 billion; the retail MMF component grew $35.7 billion more than that. The Federal Reserve began cutting rates in September 2024, lowering the target range from 5.25–5.50 percent to 3.50–3.75 percent. In September 2024, when the Fed made its first cut, the RMFSL series read $2,540.0 billion. By August 2026, it had grown another $533.9 billion — a 21.0 percent increase from September 2024 through August 2026.

The following is analysis, not fact. The aggregate M2 figure carries a hidden composition. The rest of M2 — everything other than retail money market funds — has changed little on net since January 2022: summed together, those components are $35.7 billion below where they began. The M2 headline is above its January 2022 level only because retail MMF balances more than doubled. The less obvious part of the picture: the flow persisted through rate cuts. By a yield-driven logic, falling yields should weaken the incentive to hold MMFs: as the Fed lowered rates, MMF yields fell with them. Yet balances grew 21 percent in the same 23 months. One consistent interpretation: households that set up or expanded MMF positions during the hike cycle have not unwound them — whether from inertia, the still-positive spread over bank savings rates, or a more durable reassessment of how to hold liquid savings. ^sspeculative If that shift is structural rather than cyclical, the conventional reading of M2 as a proxy for broad-based liquidity obscures how much of that liquidity now sits in an instrument with different redemption dynamics from a bank deposit's. ^sspeculative

The counter to the "wall of cash" narrative is mechanical. Buying equities does not consume money: when an investor redeems money-market fund shares to buy stock, the redemption shrinks the fund and the cash passes to the seller, who then holds it. The total money supply is unchanged. Aggregate money-market balances can stay high even as equity valuations climb, as Wolf Street's Wolf Richter argued in January 2026, because the system's money does not drain into the market. That makes the "waiting to flow in" frame a misdescription of how money moves. A separate counterpoint: real MMF yields — nominal fund yields less measured inflation — have at points in 2026 turned negative, as Wolf Street documented in June 2026. A sustained negative-real-yield environment would reduce the incentive to hold cash funds even if nominal balances have so far been stable. ^sspeculative