"Liquidity" is one of the most-used and most-abused words in macro. In the context of how it drives risk assets (equities, credit, crypto, commodities), it usually means monetary/system liquidity — the quantity of cash and near-cash balances sloshing through the financial system that can be allocated to assets — not the narrow trading sense of "how easily can I sell this without moving the price." The core idea is that the price of risk assets has two parts: the pool of liquidity available, and how willing investors are to push that pool into risk. When the pool grows and risk appetite rises, valuations expand even if earnings do not; when the pool shrinks, multiples compress and the marginal, leveraged buyer disappears first. The central tension is that this relationship is genuinely powerful at cycle scale but is routinely over-fitted into precise, mechanical "net liquidity" trading signals that do not hold up.

What "liquidity" actually means here

Three distinct concepts get collapsed into one word; keep them separate (the ECB and Fed both stress this distinction):

• Monetary / system liquidity — central-bank reserves plus broad money and credit creation by the banking system. This is the cycle driver for asset prices. Michael Howell's "Global Liquidity" framework expands it further to include cross-border flows and the gross funding/refinancing capacity of the financial system, which he describes as a pool on the order of $130 trillion of "footloose" credit and capital (Howell, Capital Wars).
• Funding liquidity — the ease with which leveraged players (banks, dealers, hedge funds) can borrow cash against collateral. Visible in repo rates, SOFR-vs-IOR spreads, FX swap bases, and cross-currency basis.
• Market liquidity — the ability to transact size without moving price (bid-ask spread, depth, immediacy). This is a transmitter and amplifier, not the cause: when funding liquidity dries up, dealers pull back, market depth thins, and price moves get violent.

The danger is borrowing one concept's credibility for another. The academic evidence that QE funding/portfolio-balance effects move bond yields is strong; the claim that a weekly "net liquidity" arithmetic predicts the S&P with 95% accuracy is a different, much weaker claim.

The transmission channels

1. Portfolio-rebalancing / hunt for yield. When the central bank buys bonds (QE) or reserves grow, it removes safe, yielding assets from the market and forces holders to substitute toward riskier assets, bidding up equities and compressing risk premia. This is the best-documented channel in the QE literature (NBER, ECB working papers). 2. Cost and availability of leverage. Ample funding liquidity lets levered investors carry larger risk positions cheaply; tightening funding (repo stress, widening basis) forces de-grossing, which hits the most crowded/leveraged risk assets first. 3. Discount rate. More liquidity is typically associated with lower real rates and lower term/risk premia — see the sibling nodes [How Interest Rates Affect Stocks] and [How Bond Yields Affect Equity Multiples]. 4. Refinancing / debt-rollover. Howell emphasizes that the system is dominated by refinancing existing debt, not net new borrowing; when gross liquidity is ample, rollover is frictionless and risk appetite stays high.

How it's used in practice

Macro practitioners track liquidity rather than treating it as a tradeable trigger:

• The "net liquidity" proxy (popularized by Max Anderson): Fed balance sheet − Treasury General Account (TGA) − ON reverse repo (RRP). The logic is that money parked in the TGA or RRP is sterilized — out of the system — while drawdowns of either release reserves. TGA rebuilds (e.g., after a debt-ceiling resolution) drain reserves and tighten conditions; TGA drawdowns and RRP drains add reserves. The Fed's own FEDS Notes confirm the mechanical accounting: a TGA increase, all else equal, reduces reserves.
• Global liquidity indexes (Howell/CrossBorder, plus aggregated G4 central-bank balance sheets and credit) used to gauge the cycle. Howell argues the global liquidity cycle runs roughly 5–6 years and leads asset prices.
• Funding-stress gauges — repo/SOFR spikes, FX swap basis, cross-currency basis, GC repo vs. IORB — watched as early-warning tripwires for forced de-risking.
• Financial conditions indexes (Chicago Fed NFCI, Goldman/Bloomberg FCI) bundle credit spreads, rates, the dollar, and volatility into a single "easy vs. tight" read.

For tactical positioning the practical use is regime context and tail-risk awareness, not entry timing: rising liquidity supports a "buy dips" posture; draining liquidity raises the odds that a normal pullback turns into a forced-selling cascade.

Adoption, debate & evidence

The cycle-scale claim is widely held and reasonably supported: Howell's work shows tight correlation between global liquidity and aggregate world asset wealth, and he argues that barely one-fifth of decades-long Wall Street gains came from earnings — the rest from rising liquidity and investors' appetite for risk (Howell's own characterization; the precise share is his estimate, not an independently audited figure). Episodes broadly consistent: the 2020–21 QE melt-up, the 2018 QT/repo stress, and the September 2019 repo spike all line up with the funding-liquidity story.

The precise, short-horizon version is contested and probably over-fitted:

• The famous 0.95 correlation of net liquidity with the S&P was an in-sample, level-on-level fit. Both series trend strongly upward, so correlations of trending non-stationary series are notoriously inflated (spurious-regression problem). On changes, the relationship is far noisier, and the signal has visibly failed for stretches (e.g., parts of 2023, when stocks rose while net liquidity fell).
• Net liquidity ignores velocity, bank credit creation outside reserves, and the rest of the world — it is a US-reserves proxy, not "the money supply for stocks."
• Causality is hard to pin down: liquidity, rates, and risk appetite move together, so attributing equity moves to the liquidity line alone is fragile.

Honest summary: liquidity is a real, important conditioning variable for risk assets at quarterly-to-multiyear scale; it is not a reliable mechanical timing indicator at weekly scale, and the headline correlations quoted by retail liquidity-trackers are overstated.

Strengths & limitations

• Works best at cycle turns and during stress, when liquidity dominates fundamentals — major QE/QT inflections, funding seizures, debt-ceiling TGA swings.
• Fails as a precise short-term signal: leads and lags are unstable (often quoted as ~2 weeks to ~3 months, but inconsistent), and equities frequently decouple for months.
• The #1 misuse is treating the net-liquidity chart as a mechanical buy/sell trigger and citing the spurious level-correlation as proof. The second is conflating market liquidity (bid-ask) with monetary liquidity.
• Regime dependence is total: the channel is strongest when rates are near zero and QE is the active policy tool, and weaker when the cycle is driven by inflation/earnings.

Sources

• Michael Howell, Capital Wars: The Rise of Global Liquidity (2020); Capital Wars Substack — "Think Correctly About 'Liquidity'", "The Global Liquidity Cycle".
• Federal Reserve, FEDS Notes — "Fluctuations in the Treasury General Account and their effect on the Fed's balance sheet" (2025).
• Federal Reserve Bank of New York — "Money Market Conditions and the Federal Reserve's Balance Sheet" (2025).
• ECB Financial Stability Review — "Gauging the interplay between market liquidity and funding liquidity"; ECB Working Paper 2399 (QE and the price-liquidity trade-off).
• NBER WP 17555 — "The Effects of Quantitative Easing on Interest Rates" (portfolio-balance channel).
• GuruFocus / Investing.com — "Fed Net Liquidity vs S&P 500" and critiques of the net-liquidity narrative (origin of the 0.95 figure; Max Anderson construction). Flagged: the 0.95 correlation is an in-sample level fit; treat as contested/overstated.