Funding Liquidity Risk

A fund can hold assets that are perfectly sellable, at a fair price, with willing buyers on the other side, and still fail. What kills it is not the asset, it's the calendar: cash is owed today, and the assets can't be converted into cash fast enough, or the people who normally lend against those assets stop lending. That gap between "I have value" and "I have cash right now" is funding liquidity risk, and it is a different animal from the risk that an asset itself is hard to sell.

Two kinds of liquidity, kept separate on purpose

Market liquidity is about the asset: can you sell it quickly, in size, without moving the price against yourself? A thinly traded small-cap stock has poor market liquidity.

Funding liquidity is about you: can you raise cash, by borrowing, by selling, or by drawing a credit line, in time to meet an obligation? A fund can hold a highly liquid asset like a Treasury bond and still face funding liquidity risk if the specific channel it relies on for cash, say a repo lender, pulls back all at once.

The distinction matters because they're often confused for the same thing, but a portfolio manager can fix a market liquidity problem by holding more liquid assets, while a funding liquidity problem is about the relationships and terms a fund depends on, which can vanish even when the assets themselves are fine.

How the two amplify each other

This is where things get dangerous. Suppose a leveraged fund's assets fall in value. Lenders react by raising the margin they require, or cutting the amount they'll lend per dollar of collateral (a "haircut" increase). The fund now needs more cash than before just to hold the same position, precisely when its assets are worth less. If it can't raise that cash, it's forced to sell into a market that is already falling, which pushes prices down further, which triggers the next round of margin increases across every other leveraged holder of similar assets.

Worked example: the repo mechanism

A fund holds $100m of mortgage bonds and funds them via overnight repo with a 5% haircut, meaning it can borrow $95m against them and must fund the remaining $5m with its own capital. If the bonds' market value drops 10% to $90m and the lender simultaneously raises the haircut to 10% (a common reaction when an asset class looks risky), the fund can now only borrow $90\text{m} \times 0.90 =$ $81m. It must find $14m in cash almost overnight — $95m it previously had, minus $81m it can now get — even though it hasn't sold anything. If dozens of funds hold the same bonds and face the same haircut increase simultaneously, this is exactly the mechanism behind the 2007–08 repo market freeze.

What this means in practice

• Diversify funding sources, not just assets. A fund that relies on a single prime broker or repo counterparty has concentrated funding risk even if its asset book is well diversified.
• Term-match where possible. Funding a long-horizon, hard-to-sell asset with overnight borrowing is the classic setup for a funding liquidity crisis — the loan can be pulled long before the asset needs to be sold. This is the same logic that fuels Margin Calls and Forced Liquidation.
• Stress-test the loop, not just the position. A risk model that only asks "what if the asset falls 10%" misses the second-round effect of haircuts rising at the same time. See Counterparty Credit Risk for the mirror-image question of what happens when the lender itself is under stress.