The Illiquidity Discount: What the Evidence Actually Supports

The illiquidity discount is the most widely used and least documented input in private valuation. It appears in almost every model as a round number, and in almost no file as a sourced one.

The evidence supports a discount. It does not support the way most people apply it.

What the evidence actually shows

Two broad families of study inform the number. Restricted stock studies compare the price of shares subject to transfer restrictions against the same company’s freely traded shares. Pre-IPO studies compare private transaction prices against the subsequent listing price.

Both approaches produce a discount. Both also produce a wide range, and the range moves with the period studied, the method chosen and the sample. Pre-IPO studies in particular are affected by selection: companies that reach a listing are not a random sample.

The honest summary is that the direction is well supported and the magnitude is not a single number.

Size it to the specific facts

Four things move the right discount for a given holding.

Expected time to liquidity. A company with a live sale process is closer to liquid than one with no exit in view.

Transfer restrictions. Drag, tag, rights of first refusal and consent requirements each narrow the set of possible buyers.

Secondary market. Some private companies have an active secondary; most do not. Where one exists, it is evidence.

Size and share class. A minority stake without information rights is harder to sell than a controlling one.

The double-count to avoid

If the peer set is already composed of small, thinly traded listed companies, part of the illiquidity is in the observed multiple. Applying a full discount on top charges twice for the same characteristic.

The same applies to a small-company risk premium in a discount rate used alongside a size discount on the multiple. Pick where the effect lives and be consistent.

This is one of the things an auditor tests directly, and it is easy to show you have not done it if the adjustments are laid out as an explicit bridge.

What belongs in the file

State the discount as a range with a midpoint, the facts that put it there, and the sensitivity of the valuation across that range.

That is more defensible than a single number, more useful to a reader, and no more work once the habit is in place. It also makes the annual review straightforward: the facts change, the range moves, and the reasoning is already written down.