Outgrace, home

Pricing a minority stake in a private company, and where the evidence runs out

Start from what a buyer with no tie to the owner would pay. This framework combines three lenses and marks where the public sources give no figures to lean on.

A minority stake in a private company has no screen price. The owner wants a high number, the buyer wants a lower one, and a board is asked to approve whichever figure the valuer delivers. This article sets out a framework for building that figure so it survives questions from an auditor or a counterparty. Call it a four-pillar method: price the buyer, build the cash flow view, test the result against comparable multiples, then apply discounts you can explain. It is a method to document and review, not a schedule of rates.

Before the method, a note on the evidence. We reviewed four sources: the International Accounting Standards Board's IFRS 13 page, the IRS homepage, the Federal Reserve homepage and a King's College London doctoral thesis. Only the first says how fair value is measured, and it does so in principle. None of the four gives a discount rate, a terminal growth assumption, a comparable multiple, a discount for lack of control or a discount for lack of marketability. Where this article makes a judgement without a source behind it, it says so.

The IFRS 13 standard is the natural anchor, and it is worth being exact about what it does. The IASB issued it in May 2011. It also takes the place of measurement rules that sat inside separate Standards. Fair value under the standard is an exit price: the amount you would realise by selling, not the amount you paid or hope to receive. Measurement rests on the assumptions a buyer in the market would make at the time, risk included. The standard applies to entities reporting under IFRS, so first check whether it binds your number at all. A financial reporting mark, a sale negotiation and a tax filing are three different jobs, and they seldom produce the same figure.

Chapter one: 1. Price the buyer, not the owner

Start by naming the buyer before building any model. A financial investor tends to ask about exit timing and the cash it can take out. A strategic buyer may ask what it can combine with its own business. Each assumes different growth and different risk. Pick one buyer type as the reference, state why it is the right reference for this stake, and record how the answer changes under the others. That note is often the most useful page in the file, because it shows a board exactly where the disagreement with the owner sits.

The test is simple to state and easy to fail. Ask whether the number would move if a different person owned the same shares. If it would, the figure reflects the owner's circumstances rather than the value of the stake. An owner who needs cash for a personal reason may want a higher price, and a buyer will not pay for that need. Strip the owner's private circumstances out of the model and write down what was removed. A reader who sees the list can judge whether the adjustment was honest.

The IFRS 13 idea that the holder's intention does not change fair value is a useful discipline here. Whether the current owner plans to keep the stake for twenty years or sell it next spring should have no bearing on the figure. If the valuation quietly depends on the owner's plan, the valuer should say so and show the effect separately, so the board can see what it is paying for.

Chapter two: 2. Build the cash flow view

A discounted cash flow model turns a forecast into a present value. Its output is only as good as three inputs: the forecast, the rate used to bring future cash back to today, and the assumption about how the business grows after the forecast ends. None of the four sources gives a rate or a growth figure for a private company. Each of the three therefore has to be argued from evidence in the file and recorded with its source, even when the evidence is the company's own history.

Test the forecast against the company's own plan and that plan's track record. Ask who prepared it and for what purpose. A projection built for a lender, a budget or a tax estimate may be cautious or optimistic in ways that do not suit a buyer's view. The IRS says estimated business taxes are paid quarterly, based on income expected in the year. That shows expected income is used for a tax purpose. The IRS homepage says nothing about how such income should be valued, and a tax estimate should not be mistaken for a valuation forecast.

The long-run growth assumption deserves its own page. Small changes in growth or in the rate can move the answer more than revisions to the forecast years do. Run the model at several settings and show the range, not a single figure. The settings themselves must come from a valuation specialist who can justify them in writing. This article does not offer a rate, and any rate found in a supplier's brochure should be treated as a starting question, not an answer.

Decide where each risk sits. The standard's language on risk assumptions is the reason for this discipline. If the forecast already haircuts a risky contract or a slow market, adding a large premium to the rate counts the same risk twice. Ask the valuer to list each risk and say whether it is in the cash flows, in the rate or in neither. A risk that appears in no place is a gap, and a risk that appears in two places is an overstatement.

Chapter three: 3. Test the result against comparable multiples

A multiple asks what similar businesses sell for, measured against a figure such as earnings or revenue. It is a useful cross-check and a weak starting point. For a private company the set of comparable businesses is usually thin, and their prices are rarely published in a form that is easy to compare. None of the four sources gives a comparable-company multiple. Every peer figure in your file therefore needs a named dataset, a date and a note on how it was adjusted.

Choose peers on the facts that move value: size, growth, margin, customer concentration and the type of deal. A minority stake in a small company compares poorly with the sale of a listed group. Be careful with acquisition premia. A King's College London doctoral thesis by Susen, dated 1 August 2024, analyses 100 global acquisition announcements from 2010 to 2019 and links environmental performance to acquisition premia. Our source list does not treat that thesis as a primary source. Even taken at face value, a premium paid in a takeover is a different quantity from the price of a passive minority interest. Keep acquisition premia out of your multiple unless you can show, in writing, that they belong there.

Keep the peer set small enough to defend and wide enough to mean something. A list of twelve companies with no screen for size or deal type is worse than four you can explain. Write down why each peer was kept and why each was dropped. If the multiple moves sharply when one company is removed, the set is too fragile to carry the price, and the valuer should say so before the board relies on it.

Chapter four: 4. Apply discounts you can explain

Two discounts are common in private company work, and neither appears in any source we reviewed. The first, a discount for lack of control, reflects that a minority holder may have little say over dividends, the sale of the company or its management. The second, a discount for lack of marketability, reflects that a private stake may take a long time, and a discount on price, to sell. The sources give no size for either. Their size is a matter of evidence the valuer gathers and documents for this company, not a figure that can be copied from another deal.

Each discount must name the right or the exit it takes away, and the valuer should show the evidence for its size. The usual error is to stack discounts so that one risk is counted several times over: once in the rate, once in the cash flows and again in a marketability discount. Ask the valuer to write one sentence per discount. It should say what the discount removes, what evidence supports its size, and what would make it larger or smaller. A discount that cannot be written in one sentence is not yet ready for a board paper.

Reconcile the answers before choosing one. Present the cash flow range, the multiple range and the discounted result side by side. If they disagree widely, the gap is the finding. The forecast may be too generous for a buyer, the peers may be the wrong peers, or a discount may be doing work it should not. Record which method carried the weight and why. A board that can see that reasoning is better placed to defend the number than one handed a single figure with no workings behind it.

Eight questions before the number goes to the board

Each question should be answered yes, with the evidence for it in the file.

  • Would the value stay the same if a different person owned the same shares?
  • Are the forecast assumptions written in one place and tied to the company's own plan?
  • Is the source of the discount rate and of the long-run growth rate recorded?
  • Are the comparable companies matched on size, growth, margin and deal type?
  • Have acquisition premia been kept out of the multiple?
  • Does each discount name the right it removes and the evidence for its size?
  • Has any single risk been excluded from being counted twice across the rate and the discoun
  • Has a valuation specialist reviewed the method before the number is used?

Outgrace checklist

Where it breaks

The framework breaks in three places, and the sources show it through their silence. First, none of the material we reviewed gives a rate, a growth figure, a multiple or a discount for a private company, so every input is a judgement that must be evidenced in the file. Second, the tax basis is unsupported. The IRS homepage covers identifiers such as EINs, estimated taxes, employment taxes, transcripts and tax-exempt status, and it gives no private company valuation guidance. Third, the Federal Reserve homepage lists minutes, speeches, press releases and payment services, with no valuation or minority-stake guidance either.

Where the price feeds a tax filing, the tax position belongs to an accountant or tax adviser, not to this framework. Ask that adviser which valuation basis the relevant tax authority expects, and get the answer in writing before the figure is used. Where the stake carries legal rights, such as a shareholder agreement with exit terms, a lawyer should read those terms against the method, because the method cannot see them.

None of the four sources contradicts another, so there is no conflict to report. The gap is one of absence rather than disagreement. Where a reader finds a published discount range or a sector multiple, treat it as a claim to verify, not a benchmark to adopt. Check who published it, when, on what sample and for what kind of transaction. A figure you cannot trace to its dataset should stay out of the board paper.

the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date

International Accounting Standards Board, IFRS 13 Fair Value Measurement, ifrs.org

On Monday, do three things. Write the buyer's assumptions on one page and name the buyer type you have chosen as the reference. Ask the valuation specialist to set out, in writing, the source for the rate, the growth assumption, the peer set and each discount, and to show the range. Then ask the auditor or tax adviser which purpose the number serves and which standard or tax basis applies to it. Pricing a minority stake needs a qualified valuation specialist. Where the figure touches a tax position or a legal right, an accountant or lawyer should review it before anyone relies on it.

Newsletter

The Briefing

One email a week: the leaders we profiled, the ideas worth stealing, and nothing else.

Weekly, free, one click to leave. Read our privacy policy.