1. Liquidity: the price you can really get out at

Risk 7: The risk that is not in the prices

The previous six levels deal with the same risk from different angles: the risk in the price. Everything you have measured, sized, stressed and budgeted comes out of a series of returns. This level deals with what is not in that series and what ruins people who were running their portfolio perfectly well. We start with the item closest to the price and, even so, invisible in it.

That number on the screen is a fact about somebody else

The price you see quoted is the price of the last trade crossed, or the best offer available right now for a specific size that almost nobody looks at. It is true information about a transaction in a hundred shares carried out by someone else. It is not a promise about the forty thousand you are holding.

Your portfolio is not worth the screen price multiplied by your holdings. It is worth what someone would pay you for the whole of it, over the period in which you would need to sell it, in the market conditions in which it would occur to you to do so. Those three conditions tend to show up together and in their worst version.

The Costs course already explained market impact: how your own order pushes the price against you and how many basis points that costs. Take it as read. Here it is not a line of expenditure: it is a risk, because its magnitude is uncertain, it changes regime without warning and it multiplies precisely on the day you decide to use it. A cost is budgeted for; this is not.

The only liquidity figure that matters: how many days you are

There is a simple way of turning liquidity into a number that means something. Divide the value of your position by the average daily volume traded in the security, and then divide again by the fraction of that volume you are prepared to account for.

With figures. You hold 90,000 euros in a company that trades an average of 150,000 euros a day. If you accept the usual convention of not exceeding 15% of a session's volume so as not to move the price perceptibly, you can place 22,500 euros a day. Getting out takes you four sessions, and that is in a market behaving as it has behaved so far.

That figure is more useful than almost everything you have calculated up to now, because it is a hard limit and not a statistical estimate:

• Less than one day of volume. You can unwind the position in a single session without attracting attention.
• Between one and five days. An orderly exit, but you need a plan and several sessions.
• Between five and twenty days. Your size depends on the market playing along. You are not the one who decides when you get out.
• More than twenty days. You do not have a position: you have a stake, and the position has you.

Notice what that implies. Level 2 sized by edge and volatility, and level 5 by risk budget. Liquidity imposes a ceiling above both: it does not matter how good the idea is if you cannot execute the exit your thesis requires.

Where liquidity is worse than it looks

Three places where the number misleads systematically.

• Small capitalization stocks. Low volume, a wide spread and a handful of participants. On top of that, the published average volume is usually inflated by a few exceptional sessions, so the median is a good deal worse than the mean.
• Thinly traded exchange-traded funds. Here both camps get it wrong. Dismissing a product because of its on-screen volume ignores the fact that an authorized participant can create new units. But assuming that mechanism solves everything ignores the essential point: the real liquidity of an exchange-traded fund is that of the assets it holds inside. A product on high yield bonds or on small emerging market shares inherits the illiquidity of the underlying however much money it gathers.
• Open-ended funds holding assets that cannot be sold in a day. Property, unlisted companies, private credit, the debt of small countries. They promise daily redemption on things that take months to sell. It is called a liquidity mismatch and it is not an anomaly: it is the design of the product.

Liquidity evaporates on the very day you need it

Liquidity is not a property of the asset but of the state of the market, and its correlation with stress is negative and brutal. When the blow lands, spreads widen, the depth of the book vanishes, whoever was quoting stops quoting and everybody who wants to sell wants to sell the same thing. In March 2020 not even US government debt, the asset used as the very definition of liquidity, traded normally for a few days.

There is a second mechanism, subtler and specific to funds. When the redemptions start, the manager meets them by selling what can be sold: the most liquid holdings. Those who leave first are paid at the full net asset value and whoever stays inherits a more illiquid portfolio than the one they bought. The exit queue has an order, and you are not first in it.

A real case: daily redemption on assets that would not sell

In June 2019 the LF Woodford Equity Income fund suspended redemptions. It had at one point run something of the order of ten billion pounds at its peak and was around three billion seven hundred million on the day of the suspension, after months of continuous outflows. The reason was not a market fall: it was that the portfolio had been filling up with unlisted companies and very small stocks while the money that remained walked out of the door, so that the illiquid part weighed more and more heavily on what was left. The fund never reopened, it was wound up in pieces and the unitholders were paid in several installments over a period of years, recovering considerably less than the net asset value showed on the day they wanted to leave.

It is not an isolated or exotic case. In June 2016, in the days following the British referendum, a very considerable part of the United Kingdom's open-ended property fund sector suspended redemptions almost simultaneously, and the episode repeated itself in 2020. Investors were not wrong about the price of the properties: they were wrong in believing they could get out whenever they liked of something that takes months to sell.