Liquidity risk: measuring market liquidity, liquidity premiums and the tools of open-ended funds

Advanced12 min read · 14 sources

Liquidity risk is the risk of not being able to sell a position quickly enough, or only at a significant discount, or, as a fund, of not being able to meet redemptions without harming the remaining investors. It rarely shows up in everyday trading and almost always in crises, when everyone wants to sell at the same time. Here you will learn how market liquidity is measured, why investors demand a premium for liquidity risk and which tools open-ended funds use, with the EU rules that have applied in Germany since April 2026 as the example.

Two kinds of liquidity

Market liquidity describes how easily a security can be traded: how narrow the bid-ask spread is, how much can be traded at that price (depth) and how quickly the price recovers after a large order. Funding liquidity describes how easily a dealer or fund can raise money: through credit, collateral or new investors. The two are linked. Brunnermeier and Pedersen (2009) show in a model that dealers only provide market liquidity if they have funding, and that their funding depends on market liquidity via margins. Under certain conditions, the two reinforce each other: liquidity can suddenly evaporate, moves together across securities and is related to volatility.

For investors, there is a third level: fund liquidity. An open-ended fund promises daily or monthly redemption but may hold assets that can only be sold slowly. This liquidity mismatch is harmless as long as inflows and outflows are small, and dangerous when many investors want to exit at the same time. Whoever leaves first gets the full price; the remaining investors bear the cost of the sales. The tools described below are meant to eliminate this first-mover advantage.

Measuring market liquidity: spread, Roll measure and Amihud measure

The most direct measure is the spread between the bid and ask prices. Where no order book data is available, it can be estimated from prices. Roll (1984) shows that trading costs create negative autocorrelation in successive price changes, because trades alternate between the bid and the ask. Amihud (2002) measures illiquidity as the price reaction per unit of money traded. This Amihud measure can be calculated from daily prices and volumes for almost any stock:

For large orders, market impact matters in addition to the spread: empirically, it grows roughly with the square root of the order size relative to daily volume, market impact ≈ Y × σ × √(Q / V), as the article order types and execution shows with examples (Tóth et al. 2011). For the liquidity risk of a portfolio, this means that liquidation costs rise less than proportionally with position size, but they do rise, and in periods of stress σ shoots up, and with it the costs.

Liquidity as a risk factor

The idea that illiquid assets must offer higher returns is an old one. Amihud and Mendelson (1986) model investors with holding periods of different lengths: expected returns rise with the spread, but in a concave way. Long-term investors hold the illiquid securities and spread the costs over many years. Amihud (2002) also shows over time that expected market illiquidity raises the expected excess return, while unexpected illiquidity lowers contemporaneous prices, especially for small companies. So the premium compensates not only for the costs but also for the damage in liquidity crises.

Pástor and Stambaugh (2003) take the second step: what is rewarded is not the level of liquidity but the sensitivity to fluctuations in market-wide liquidity. From 1966 to 1999, stocks with high sensitivity returned 7.5% per year more than those with low sensitivity, adjusted for market, size, value and momentum; their liquidity factor explains half of momentum profits (see momentum crashes). Acharya and Pedersen (2005) combine this in a liquidity-adjusted CAPM. With c as illiquidity costs and betas each divided by Var(r_M − c_M), the following holds:

An asset is therefore doubly costly to hold if it becomes illiquid or falls precisely when the whole market is illiquid. In private markets, the same principle is called the illiquidity premium; how smoothed valuations conceal the risk is described in measuring private markets.

Fund liquidity: swing pricing, gates and suspension

EU law shows how regulators set out these tools. Directive (EU) 2024/927, known as AIFMD II, revised the rules for open-ended funds; in Germany, the amendments made by the Fund Risk Limitation Act (Fondsrisikobegrenzungsgesetz), which implements the directive among other things, have applied since April 16, 2026. Under the new Section 30a of the KAGB (German Capital Investment Code), a management company must select at least two appropriate liquidity management tools from the lists in the UCITS or AIFM Directive for each open-ended fund; the selection may not consist only of swing pricing and dual pricing, and for money market funds one is sufficient (KAGB 2026). ESMA, the EU's securities regulator, recommends combining, where possible, a quantity-based tool with an anti-dilution tool (ESMA 2026).

Suspension is the last resort. According to ESMA, suspensions should only be used in exceptional cases and temporarily, for example in the event of valuation problems or market closures. Management companies should derive the swing factor from the estimated liquidity costs, including significant market impact where appropriate (ESMA 2026); IOSCO (2023) takes the same view for all anti-dilution tools, with explicit and implicit costs combined. In May 2025, IOSCO revised its recommendations on the liquidity risk of open-ended funds into 17 recommendations in six areas (IOSCO 2025).

Does it work? Using investor data from UK bond funds, Jin et al. (2021) show that swing pricing eliminates the first-mover advantage and significantly reduces outflows in periods of stress. However, swing pricing only protects against dilution, not against the sale of illiquid assets itself; that is what gates, notice periods and, in extreme cases, suspension are for. Another worked example of swing pricing can be found in the article index providers and valuation points.

ETFs: primary and secondary market in periods of stress

With ETFs, there are two levels of liquidity. On the secondary market, investors trade with each other on the exchange; on the primary market, authorized participants create or cancel units, usually against a basket of the securities (creation and redemption). Arbitrage keeps the market price close to the net asset value as long as dealers have the balance sheet and risk appetite to trade the baskets. In March 2020, corporate bond ETFs traded at significant discounts to their net asset values, driven by high volatility and reluctant dealers (Aramonte and Avalos 2020); see premium and discount.

Two readings are possible, and both are partly right. Either the discount reflects the ETF's illiquidity. Or it shows that the market price reflects the actual selling prices of the bonds faster than a net asset value based on stale prices. For you, this means that an ETF is as liquid as its exchange trading in a calm market, but in stress it is no more liquid than the securities it holds. Execution, spreads and trading at NAV are explained in the article order types and execution.

Advanced: liquidation horizon and liquidity-adjusted risk

In fund risk management, liquidity is often measured in days: how long does it take to sell a position if you want to trade at most a fixed share of daily volume? On top of this come costs that rise with speed, and a spread that widens under stress. Both can be built into a liquidity-adjusted value at risk: price risk is supplemented by the half spread incurred on the sale and, for large positions, by market impact (see value at risk).

For funds, regulators require liquidity stress tests on both sides of the balance sheet: how much can be sold within what period and at what cost, and how high could redemptions be in a crisis? ESMA explicitly links the selection and calibration of the tools to the fund's investment strategy, redemption policy and liquidity profile (ESMA 2026). A common mistake is to stress only the asset side: redemptions come in waves, and large individual investors make them less predictable.

Limitations and common mistakes

• Measuring liquidity in a calm market: spreads and volumes on normal days say little about crisis days, when they can multiply.
• Confusing measures: the Amihud measure mixes trading pressure and information; across markets and currencies, it is only comparable after conversion.
• Equating fund liquidity with asset liquidity: a fund with daily redemption can hold illiquid assets. The tools decide who pays in a period of stress.
• Mistaking gates for protection: they protect the remaining investors, not the one exiting, whose money arrives later and at an unknown price.
• Backtests without liquidity: prices of illiquid securities are often stale; smoothed returns underestimate volatility and correlation.
• Not checking the legal terms: which tools a fund uses is set out in its fund rules. This article is not legal or investment advice.

How Portfolio Lab calculates

Portfolio Lab does not calculate liquidity measures such as the Amihud measure, nor market impact. In the Backtest, you enter trading costs as you would with a broker: a bid-ask spread as a percentage of the price and an order fee made up of a fixed amount and a percentage. Purchases are made half a spread above the closing price, sales half a spread below it. The spread is fixed; the app does not model its widening under stress. Every position is assumed to be tradable at the closing price, and tools such as swing pricing, gates or notice periods do not exist in the backtest. More in the backtest methodology.

Under Securities, the screener filters funds by their fund size (from 0, 50, 100, 250, 500, 1,000 or 5,000 €m). This is a rough indication of closure risk and tradability, not a liquidity measure. The app does not cover open-ended property funds as a series; you will find listed real estate under Asset classes. The Sentiment page shows, among other things, the Fed's net liquidity in the “Rates and liquidity” group. This is a macroeconomic quantity, not the market liquidity of a security.

This article explains a method; it is not investment or tax advice. Worked examples are simplified and rounded. All terms in the glossary.