Money supply is the stock of money in an economy measured under a specified definition, usually called a monetary aggregate. Depending on that definition, it includes currency and selected deposits or other liquid claims. For a quant using money growth as a macroeconomic signal, the first task is to identify exactly what the series counts: a change in its boundary can resemble an economic change. The Federal Reserve’s overview describes several measures rather than one unique total.
What belongs inside the boundary?
An aggregate combines eligible instruments held by eligible holders and issued by eligible institutions. A bank deposit is the customer’s asset and the bank’s liability; adding both sides would count the same claim twice.
Before comparing two money series, check:
| Boundary | Question to answer |
|---|---|
| Instruments | Which deposits, maturities and other liquid claims qualify? |
| Issuers | Which banks, funds or other institutions are covered? |
| Holders | Which household, business, government, financial and nonresident holdings qualify? |
| Geography and currency | Is coverage based on residence, and which denominations are included? |
“Narrow” generally emphasises means of payment; broader measures include additional money-like assets. Labels such as M1, M2 and M3 are jurisdiction-specific definitions, not universal buckets. For example, the Bank of England’s M4 definition specifies holdings by the UK private sector other than monetary financial institutions, and sterling instruments. Reading only “broad money” would miss those boundaries.
Two totals from the same fictional economy
At one date, suppose eligible households and businesses hold the following nonoverlapping stocks, all in millions of currency units. Issuer, holder and currency coverage are fixed throughout this illustration.
| Eligible holding | Amount |
|---|---|
| Currency held by the public | 60 |
| Transferable bank deposits | 140 |
| Selected additional deposits, excluded from the narrow measure | 100 |
Define our fictional narrow aggregate as public currency plus transferable deposits:
Define our fictional broader aggregate by including the additional deposits:
These are illustrative definitions, not official M1 and M2. The broader measure already contains the narrow measure: adding 200 and 300 would double-count the first two components. Nor is 300 versus 200 money growth: both measure the same date using different boundaries.
Original fictional example at one observation date, in millions of currency units. The two bars use the same holder, issuer and currency coverage; only the included instruments differ. They are alternative aggregates, not amounts to add together.
Public money and the monetary base overlap differently
The monetary base measures central-bank money under its own statistical definition. Bank reserve balances at a central bank are distinct from customers’ deposits at commercial banks. They are not an extra household spending balance to append to a broad-money total.
Cash location matters too. Currency held in a bank’s vault is not currency held by the public. In the US H.6 release, the currency component of M1 excludes depository-institution vaults. Its monetary-base currency component includes currency outside the Treasury and Federal Reserve Banks, including vault cash; reserve balances are added separately. Thus the base and public-money aggregates overlap, but neither is simply a pile of extra components to add to the other.
Commercial banks can create deposits when they lend. That does not make lending unconstrained: capital, risk, funding and settlement requirements matter. The Bank of England’s explanation distinguishes deposit money from the reserves banks use between themselves. A rise in reserves does not mechanically imply a fixed multiple of new public deposits.
A definition break can look like a monetary surge
US M1 provides a concrete warning. Beginning with the May 2020 observation, savings deposits were included within M1’s “other liquid deposits.” The Federal Reserve identifies a break in M1 from that reclassification; M2 was unchanged by that particular change. A model that treats the discontinuity as entirely new deposits misreads the series. H.6 technical explanation.
Methods continue to evolve. The same Q&A documents a July 2026 change in how retirement-account balances are netted when constructing M2, applied retrospectively. Keep the release vintage and methodology with the data, rather than relying on a remembered formula.
For research, record the jurisdiction, aggregate, units, frequency, observation convention, seasonal adjustment, definition and release vintage. A stock can be reported as an end-period level or an average over a period; those are different observations. Compare growth only across a consistent basis and inspect breaks or revisions before interpreting it.
Money aggregates help describe liquidity and financial conditions. They are not measures of total wealth, bank lending or nominal GDP. Money growth alone does not establish a fixed change in spending or inflation: money demand, credit conditions, production and policy responses also matter. The Fed treats money supply as one part of a wider economic information set, rather than a standalone forecast rule. Fed overview.
Sources
- Federal Reserve: What is the money supply? Is it important?
- Federal Reserve: H.6 Money Stock Measures, 27 October 2026, Table 1 footnotes
- Federal Reserve: H.6 Technical Q&As, 2020 reclassification and July 2026 netting methodology
- Bank of England: Further details about M4 data, Definitions
- Bank of England: How is money created?