Global net liquidity — the supply behind the price
Everyone talks about liquidity. Almost nobody measures it.
Five central banks, one line, updated weekly. Read it yourself, then
see why the usual version of this chart is not measurement.
5 banks
Fed, ECB, Bank of Japan, PBoC and Bank of England, converted to one currency.
1 line
Net of the Treasury account and reverse repo, weekly, against BTC, the S&P 500 and gold.
WHAT THE CYCLE LABEL MEANS
EXPANSION: Central-bank liquidity rose this week.
NEUTRAL: The week-on-week change is inside the flat band.
CONTRACTION: Central-bank liquidity fell this week.
This describes liquidity, not markets. It is not a signal to buy or sell anything.
- Ten-year and full history
- Macro overlays: unemployment, DXY, Fed rates, CPI and M2
- The full data table, and CSV export
- Chart image and PDF report
Educational content only — not financial advice. Central-bank balance sheet data from FRED and the publishing central banks; asset prices from Yahoo Finance. Figures are as reported and may be revised at source. Nothing here is a recommendation to buy or sell any asset.
Everyone talks about liquidity. Almost nobody measures it.
“Liquidity is drying up” is one of the most repeated sentences in financial media, and it is almost never attached to a number, a source or a definition. It is a mood presented as a market condition.
A sentence
Liquidity is tightening, or loosening, asserted with confidence and no arithmetic behind it. You cannot check it, reproduce it, or tell when it stops being true.
A sum you can audit
Five published balance sheets, two published drains, one conversion, one subtraction. Every input has a series code and every step is on this page.
There is a second problem, and it is the more expensive one. Most versions of this chart quietly show the Fed only and call it global. That was defensible for a while and stopped being defensible some time ago: the euro area, Japan and China are not rounding errors, and there have been long stretches where the Fed was draining while the rest of the world was adding. A US-only line through one of those stretches does not merely lose precision. It points the wrong way.
The third problem is subtraction. A central bank’s total assets are not the same as the money actually available to markets, because some of it is parked where it cannot do anything — sitting in the Treasury’s account at the Fed, or lent back overnight through reverse repo. A chart of gross assets treats those as if they were still in circulation. This one nets them out and plots both lines, so you can see how much of the movement is the balance sheet and how much is the plumbing.
None of that makes this a forecast, and it is worth being blunt about that before you scroll any further. The line tells you what the supply of money has done. It does not tell you what any asset will do next, it has no predictive claim attached, and the correlations printed under the chart are there so you can judge how loose the relationship actually is rather than take our word for it.
Five balance sheets in, two drains out, one line.
This is the whole calculation. There is nothing proprietary in it, which is the point — you should be able to rebuild this in a spreadsheet and get the same answer.
FRED: WALCLFRED: WTREGENFRED: RRPONTSYDFRED: ECBASSETSWFRED: JPNASSETSThe chart also carries a gross line, and the difference between the two is the part worth watching. Gross is the sum of the published balance sheets with nothing taken out. Net removes the Treasury account, the reverse repo facility and the policy adjustments. When the two lines move together, the balance sheets are driving it. When they diverge, the plumbing is — and a drain that large has moved markets before while every headline number stayed flat.
You can also split it. US shows the Fed leg on its own, ROW shows the other four combined, and GLOBAL is the sum. That split exists because the interesting weeks are the ones where the two halves disagree, and a single global line hides exactly those.
One thing the tool will not do is fill a gap. If a source does not publish, or the components sum to something impossible, the affected week is left blank and the outage is named on the chart. It does not carry the last value forward and it does not smooth over the hole. A flat line through a break looks like calm and is not.
Read it in about a minute, once a week.
This is not a screen to sit in front of. The underlying data updates weekly, so checking it more often than that is checking nothing.
Look at the direction, not the level
The absolute total is a large number that means very little on its own. What carries information is whether it is rising or falling, and how fast — which is what the weekly change under the headline and the momentum panel below the chart are for.
Check whether the two halves agree
Switch between US and ROW. When the Fed is draining and the rest of the world is adding, a US-only chart tells you the opposite of what is happening. Weeks where the halves disagree are the ones worth a second look.
Put an asset on it and read the correlation
Turn on Bitcoin, the S&P 500 or gold. The rolling 90-day correlation prints under the chart. Read it honestly — it moves around a lot, it goes negative, and knowing that is more useful than a story about how liquidity drives everything.
Then go and do something else
There is no alert, no notification and no account. The data changes weekly and the tool is here when you want it. Nothing here needs you to watch it, which is the entire design intent.
Free
- The live net and gross liquidity lines
- Three years of weekly history — YTD, 1Y and 3Y
- US, rest-of-world and global splits
- Bitcoin, S&P 500 and gold overlays
- US recession bands
- Rolling 90-day correlations
- Log scale, USD, EUR and JPY, momentum panel
Pro — $49 once
- The full history, back roughly thirty years
- The ten-year view
- Unemployment, dollar index, Fed rates, CPI and M2 overlays
- The week-by-week data table
- CSV export of the whole series
- Chart image and PDF report
What this tool cannot do.
Listed here rather than buried, because a tool that only advertises its strengths is not a measuring instrument, it is an argument.
Objection
“Liquidity does not actually drive asset prices.”
Sometimes it plainly does not, and the tool will show you that rather than argue with you. Turn on an asset and watch the rolling correlation: it is not stable, and it goes negative. There are long stretches where liquidity rose and the asset fell.
That is why the correlation is printed on the chart instead of asserted in the copy. The honest claim is narrow: this is the supply side, and the supply side is one input among several. Anyone telling you it is the master key is selling something.
Objection
“Two of your five central banks are scraped off a web page.”
Correct, and it is the weakest part of the tool. The Fed, ECB and Bank of Japan come from FRED and are solid. The PBoC and the Bank of England do not publish a comparable machine-readable series, so those two are read from the published tables, with a manual override in Settings for when a source redesigns its page.
When a scrape misses, the tool falls back to the stored figure and flags it on the admin screen rather than silently substituting something. It is a real limitation and you should weigh the China and UK legs accordingly.
Objection
“Is this going to tell me when to buy?”
No, and it is not built to. There is no signal, no score, no threshold and no alert anywhere in it. It reports a measurement and stops. The word EXPANDING or CONTRACTING under the headline is the sign of the weekly change in central-bank balance sheets — a fact about liquidity, not a view on any asset.
If you want the risk-first framework this site actually runs on, that lives in the weekly issue and in the tools built for it. This one is context, not a trigger.
Objection
“The number changed since last week and it was not just new data.”
Central banks revise. The Fed restates, the ECB restates, and a figure published on a Thursday is not always the figure that stands a month later. This tool shows what the source currently publishes, so a revision upstream moves the history here too.
Exchange rates add a second wobble: the euro and yen balance sheets are converted to dollars, so part of any week’s move is currency rather than policy. When the live rate feed is unavailable the page says the totals are approximate rather than pretending otherwise.
Context for a plan you already have.
This is the least actionable tool on the site, and that is the correct description of it. It answers one question well and refuses the others.
Worth your time if
- You keep reading that liquidity is tightening or loosening and you want to see the series behind the sentence.
- You already run a systematic plan and want the macro backdrop as context rather than as a trigger.
- You want to know how tightly Bitcoin or the S&P has actually tracked liquidity, including the stretches where it did not.
- You would rather have a number with a published source code attached than a confident paragraph.
- You check in once a week and then get on with your life.
Skip it if
- You want an entry signal. There is none here, and adding one would make the tool worse.
- You want a forecast of where liquidity goes next. It measures what happened, and nothing else.
- You trade intraday. The underlying data is weekly, so this is the wrong resolution entirely.
- You want one number that explains the market. This is one input, it is loosely correlated, and it is openly presented that way.
- Three years of history already answers your question — in which case use the free version and keep your $49.
What the number is, and what it is not.
Answered against what the tool actually does, not against what would be convenient to claim.
What is global net liquidity?
The combined balance sheets of the world’s major central banks, minus the money that exists but is not circulating. Here that means the Fed, the ECB, the Bank of Japan, the People’s Bank of China and the Bank of England, added together in US dollars, with the US Treasury General Account and the Fed’s overnight reverse repo facility subtracted.
The subtraction is the part that makes it “net”. A central bank’s total assets tell you what it has bought. They do not tell you how much of that is actually loose in the financial system, because cash sitting in the government’s account at the Fed, or lent straight back to the Fed overnight, is not doing anything. Gross tells you what the balance sheets say; net tells you what is available.
How is global liquidity calculated?
As a plain sum, and the whole of it is on this page. Fed total assets − Treasury General Account − reverse repo, plus the ECB, Bank of Japan, PBoC and Bank of England balance sheets converted into dollars, with each of the last four adjusted for its own published policy operations.
The US figures come from the St. Louis Fed’s FRED database — WALCL, WTREGEN and RRPONTSYD — alongside ECBASSETSW and JPNASSETS. There is nothing proprietary in the method, which is deliberate: you should be able to rebuild this in a spreadsheet from the same public series and get the same line.
Does global liquidity affect Bitcoin and the S&P 500?
The relationship is real, loose, and much less reliable than it is usually presented. Rather than assert a number, this tool prints the rolling 90-day correlation on the chart and lets you watch it move. It does move — a lot — and it spends time negative.
What that means in practice: liquidity is one input into asset prices, not the explanation for them. There have been long stretches where liquidity expanded and risk assets fell, and stretches where the opposite happened. Anyone presenting this as a mechanical relationship is overstating it. Turn on Bitcoin, then turn on the S&P, and judge the strength of it for yourself instead of taking anybody’s word for it, ours included.
Which central banks are included, and how reliable is each one?
Five. The Fed, ECB and Bank of Japan come from FRED and are as solid as this kind of data gets. The PBoC and the Bank of England are read from their published tables, because neither offers a comparable machine-readable series, and both have a manual override for when a source redesigns its page.
That is a genuine weakness and it is worth knowing which legs it applies to. When one of those reads fails, the tool falls back to the stored figure and records it rather than quietly substituting something plausible. It never fills a gap with an estimate.
How often does it update?
The underlying data is weekly, because that is how the central banks publish it. The tool refreshes in the background through the day so the page is never waiting on a feed, but checking it more than once a week is checking the same numbers again.
That cadence is the point rather than a limitation. This is a once-a-week context check, not a screen to sit in front of, and there is no alert or notification anywhere in it to suggest otherwise.
Is it a buy signal?
No. There is no signal, score, threshold or alert anywhere in this tool, and adding one would make it worse. It reports a measurement and stops.
The word EXPANDING or CONTRACTING under the headline is the sign of the week-on-week change in central-bank balance sheets — a fact about liquidity, not a view on any asset. If you are looking for the risk-first framework this site actually runs on, that lives in the weekly issue and in the tools built around it. This one is background, not a trigger. Educational content only, and not financial advice.
What is the difference between the free and Pro versions?
History and export. The free version is the working tool, not a preview: the live net and gross lines, the US and rest-of-world split, Bitcoin, the S&P 500 and gold, recession bands, the rolling correlations, and three years of weekly history.
Pro opens the full series back roughly thirty years, adds the ten-year view, and lays unemployment, the dollar index, Fed rates, CPI and M2 over the same axis. It also unlocks the week-by-week table and exports — CSV, chart image and PDF. It is $49 once, not a subscription. If three years answers your question, it answers it.
Do I need an account, and is my data saved?
No account and no email gate for the free version. You are not entering any personal data into this tool — there is nothing to enter. It reads public central-bank series and draws them.
The only thing kept in your browser is your chart preferences, such as light or dark. Buying Pro creates an account so the purchase can be attached to something, which is the only reason it exists.
Three years is enough to see the shape. Thirty is enough to test it.
Pro opens the full history back to the 1990s, adds the ten-year view, and puts unemployment, the dollar index, Fed rates, CPI and M2 on the same axis as the liquidity line. It also unlocks the week-by-week data table, and exports the whole series as CSV, the chart as an image, and the page as a PDF.
Paid once. Not a subscription, not a bundle.
The free version is the working tool, not a screenshot of one: the live liquidity line, the US and rest-of-world split, Bitcoin, the S&P 500 and gold on the same chart, recession bands, and the rolling correlations. If three years answers your question, it answers it, and you do not need this. Educational content only — not financial advice.