The BlockPhi Global Liquidity Model.

M2 counts bank deposits, but much of today’s credit runs on government bonds. A measure that counts them is usually priced for funds, so we built one for individual investors too.

Open a free accountLicense the data
Sources of Global Liquidity at each year end, US$ trillions, and the latest week.
Week endingGlobal LiquidityCollateral baseCentral bank money
25 December 2020$154.80$92.22$29.72
31 December 2021$167.65$98.15$32.51
30 December 2022$162.13$91.03$29.36
29 December 2023$168.48$93.03$27.93
27 December 2024$170.65$92.87$25.63
26 December 2025$187.13$101.29$26.00
9 October 2026$198.20$104.84$26.06
Source: BlockPhi Global Liquidity Model9 October 2026
US$ trillions9 October 2026
Global Liquidity
$198.20
Collateral base
$104.84
Central bank money
$26.06
Sovereign debt
$78.77
Collateral multiplier
1.8905

The problem

The standard read of liquidity misses most of what credit runs on.

Partial. Global M2 is a common proxy for liquidity, and it is easy to download. M2 counts cash and deposits. It leaves out the government bonds that banks, funds and dealers borrow against: about $80 trillion of a collateral base of about $107 trillion on 18 September 2026, a model estimate. Fed net liquidity covers one central bank.

Since January 2020, that sovereign debt has added far more to the collateral base than central bank balance sheets have.

Out of reach. A full Global Liquidity series usually sits behind an institutional licence, which leaves individual investors with the partial read.

Why it matters. If you weigh an asset against liquidity, the measure you choose shapes the answer. M2 leaves out the government bonds that much of today’s credit is borrowed against.

Why Global Liquidity is not Global M2
Cumulative change in US dollars since 3 January 2020 for central bank balance sheets and for sovereign debt, two lines, weekly to 18 September 2026. By 18 September 2026: central banks plus $5.76 trillion, sovereign debt plus $27.84 trillion.
Central banks vs sovereign debt, cumulative change since 3 January 202018 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model
Two lines in US dollar trillions, from zero: the BlockPhi Global Liquidity Model, weekly from 3 January 2020 ($123.1T) to 18 September 2026 ($200.3T), and Global M2, the M2 or broad money of 33 economies (United States, Euro area, China, Japan, United Kingdom, India, South Korea, Hong Kong, Australia, Taiwan, Canada, Russia, Brazil, Switzerland, Mexico, Saudi Arabia, United Arab Emirates, Poland, Thailand, Singapore, Malaysia, Turkey, Indonesia, Sweden, Norway, Czechia, Philippines, South Africa, Denmark, Egypt, New Zealand, Colombia and Chile), each converted to US dollars and summed, monthly from January 2020 ($91.2T) to July 2026 ($137.5T). On 31 July 2026 Global Liquidity stood at $193.5T.
Global Liquidity vs Global M2, in US dollars18 September 2026. Global Liquidity is a model estimate.Sources: BlockPhi Global Liquidity Model; Global M2: the M2 of 33 economies, about 95% of the world’s, from their central banks and statistics offices, converted to US dollars and summed by BlockPhi.

The solution

The collateral that M2 leaves out, counted.

The Global Liquidity Model is a composite that measures the balance-sheet capacity behind global credit: the balance sheets of five central banks plus 17 government debt markets, revalued every Friday in US dollars and multiplied by a collateral multiplier keyed to the US ten-year Treasury yield.

The collateral base by component, 5 central banks and 17 government debt markets stacked by size and coloured by currency, with their total line, and Global Liquidity drawn above it, in US$ trillions, weekly from 3 January 2020 to 18 September 2026. In the week of 18 September 2026, Global Liquidity was $200.31 trillion and the collateral base $106.56 trillion.
Collateral base by component, with Global Liquidity18 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model

In the terminal

A free account opens the long-term charts. BlockPhi Gold adds all 22 components and what moved Global Liquidity each week, in dollars.

Open a free account

Under licence

For use inside a firm: the whole series, by API or CSV, set up once terms are agreed.

License the data

Every component, in plain view

Table of the collateral base components in the week of 18 September 2026, 18 rows with the central bank and sovereign debt of the United States, China, Japan and the United Kingdom combined: the latest level and the change over 11 weeks, and on wider screens a two-year trend line, the change over 1, 4 and 52 weeks, the share of the collateral base and what each one moves on.
Table of the collateral base components in the week of 18 September 2026, 18 rows with the central bank and sovereign debt of the United States, China, Japan and the United Kingdom combined: the latest level and the change over 11 weeks, and on wider screens a two-year trend line, the change over 1, 4 and 52 weeks, the share of the collateral base and what each one moves on.
Table of the collateral base components in the week of 18 September 2026, 18 rows with the central bank and sovereign debt of the United States, China, Japan and the United Kingdom combined: the latest level and the change over 11 weeks, and on wider screens a two-year trend line, the change over 1, 4 and 52 weeks, the share of the collateral base and what each one moves on.
Component detail18 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model

What moved, and when

It says what moved

Each week’s move can be split into balance-sheet and issuance changes, bond prices, currencies and the collateral multiplier, so the dollar’s part can be separated, not hidden inside one line.

Bars attributing the change in Global Liquidity over the 11 weeks to 18 September 2026 to its drivers: Balance sheets and debt plus $2.30 trillion; FX translation plus $2.96 trillion; Bond prices minus $1.33 trillion; Collateral multiplier plus $1.73 trillion. Together they sum to the change in Global Liquidity, plus $5.66 trillion.
What moved Global Liquidity over the past 11 weeks, by driver18 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model
Bars of the Friday-to-Friday change in the collateral base, including FX translation and bond prices, in US dollars, rises in green and falls in red, weekly from 10 January 2020 to 18 September 2026. The week of 18 September 2026: plus $1.86 trillion.
Weekly change in the collateral base18 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model

The collateral base

Stacked areas of the 22 collateral base components, 5 central banks and 17 government debt markets, largest at the bottom and coloured by currency, with the collateral base total line, in US$ trillions, weekly from 3 January 2020 to 18 September 2026, when the total was $106.56 trillion.
Collateral base by component18 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model
Ranked bars of the change in each of the 22 collateral base components over the 11 weeks to 18 September 2026, in US dollars, rises in green and falls in red. Largest rise: China, plus $762 billion. Largest fall: European Central Bank, minus $116 billion.
What moved the collateral base over the past 11 weeks18 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model

Growth and the dollar

Global Liquidity growth: the 3-month annualised and 12-month change, weekly from 3 January 2020 to 18 September 2026, with a zero line. In the week of 18 September 2026 the 3-month annualised rate was 9.89 per cent and the 12-month rate 6.92 per cent.
Global Liquidity growth, 3-month annualised and 12-month18 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model
Cumulative FX translation on every non-dollar collateral base component since 3 January 2020, in US$ trillions, weekly to 18 September 2026, shaded from zero. By 18 September 2026 it stood at minus $4.06 trillion.
Dollar translation effect on the collateral base, cumulative since 3 January 202018 September 2026. Model estimate.Source: BlockPhi Global Liquidity Model

Documentation