Is stock-based compensation a real cost? A practitioner's guide to the SBC add-back debate, IFRS 2 basics, dilution and how it flows through the QoE bridge.
Few line items provoke a sharper disagreement across a deal table than stock-based compensation. The seller's adviser presents a glossy "adjusted EBITDA" with the whole charge added back, arguing it never touched the bank account. The buyer's diligence team looks at the same number and sees a genuine, recurring cost of retaining the very people who run the business. Both are looking at the same figure in the accounts. They reach opposite conclusions. As the analyst holding the pen on the Quality of Earnings analysis, you are the one who has to decide whose logic survives - and defend it in the report.
Stock-based compensation (SBC) is where accounting theory, cash economics and negotiation collide. Get the treatment wrong and you either overstate maintainable earnings by a full margin point or hand the seller a valuation windfall they have not earned. This article walks through why the debate exists, what IFRS 2 actually requires, how SBC threads through the EBITDA bridge, and what a defensible treatment looks like when a partner challenges it.
Stock-based compensation is remuneration settled in the company's own equity rather than in cash. The common forms are share options (the right to buy shares at a fixed strike price), restricted stock units or RSUs (a promise to deliver shares once vesting conditions are met), and growth shares or "sweet equity" in private-equity-backed businesses. The employee earns these instruments by delivering service over a vesting period, typically three to four years.
The economic logic is simple: the company hands over value to employees, and that value has a cost. Under IFRS 2 Share-based Payment (and its US counterpart ASC 718), the business must recognise an expense in the income statement equal to the grant-date fair value of the award, spread over the vesting period. For options, that fair value is usually estimated with a Black-Scholes or binomial model; for RSUs it is broadly the share price at grant.
The crux of the debate in one sentence: SBC is a non-cash expense but not, for most businesses, a non-recurring one. Those two ideas get conflated constantly, and conflating them is how bad adjustments get made.
That distinction between non-cash and non-recurring is the single most important idea in this entire article. Hold onto it.
The seller's adviser will make three arguments, and they are not stupid arguments.
First, no cash leaves the business. EBITDA is meant to be a rough proxy for cash operating performance, and SBC never hits the cash flow statement as an outflow - it is added back to profit in the operating cash flow reconciliation, right alongside depreciation. If we add back depreciation, the argument runs, why not SBC?
Second, it is dilutive, not operational. The cost of SBC is ultimately borne by existing shareholders through dilution of their ownership stake, not by the operating business. So it belongs in the equity story, not the earnings story.
Third, in many private deals, the existing scheme dies at completion. Founder growth shares crystallise and are paid out; the old option pool is cancelled and replaced by a fresh incentive plan designed by the new PE owner. If the historical charge relates to instruments that will not exist post-deal, why should it depress maintainable earnings?
Each of these has a kernel of truth. The problem is that the first two prove far less than the seller wants, and the third is the only one that sometimes genuinely works.
The buyer's diligence team pushes back hard, and here is the reasoning you will usually end up siding with.
SBC is real remuneration. If the company stopped granting equity, it would have to pay more cash to retain the same talent - particularly engineers, salespeople and senior management in high-growth businesses. The equity is a substitute for cash salary, not a gift. Treating it as free understates the true cost of running the business. This is precisely the logic that underpins sound revenue and cost-quality analysis: maintainable earnings must reflect what it actually costs to keep the engine running.
"Non-cash" is a weak defence. Plenty of accrual-based costs are non-cash in a given period - think of a bonus accrued in December and paid in March, or a pension charge. We do not add those back simply because the cash timing differs. The relevant test for an EBITDA adjustment is recurrence and normality, not cash timing. Warren Buffett's line on this is worth remembering: if compensation isn't an expense, what is it, and if it doesn't belong in earnings, where in the world does it belong?
Dilution is a genuine cost to the buyer. The buyer is acquiring equity value. If that equity is being steadily handed to employees, the buyer's future ownership is being eroded - which is exactly why it should be captured somewhere in the analysis, whether in earnings or explicitly in the equity bridge.
The mainstream diligence view, and the one you should default to, is that recurring SBC is an operating cost and should not be added back to EBITDA. Deviations from that default need a specific, evidenced reason.
Numbers make the debate concrete. Consider a software business with reported EBITDA of £10.0m, an SBC charge of £1.5m, and a deal struck at an 11.0x EV/EBITDA multiple. Watch what the SBC decision does to enterprise value.
| Line | Add SBC back (seller view) | Treat SBC as a cost (buyer view) |
|---|---|---|
| Reported EBITDA | £10.0m | £10.0m |
| SBC adjustment | +£1.5m | £0.0m |
| Adjusted EBITDA | £11.5m | £10.0m |
| Multiple | 11.0x | 11.0x |
| Implied enterprise value | £126.5m | £110.0m |
| Price swing | - | £16.5m |
A single line item, argued one way rather than the other, moves headline value by £16.5m - roughly 15%. That is why nobody at the table treats SBC as a rounding item, and why your write-up has to be watertight. The multiple magnifies a £1.5m annual charge into an eight-figure valuation difference.
Takeaway: never let an SBC adjustment pass through on the "it's non-cash" reflex. At a double-digit multiple, the reflex is worth millions to whichever side you fail to challenge.
In the Quality of Earnings analysis, SBC has to be located, quantified and categorised - not just waved through. Your job is to split the total charge into buckets that each get their own treatment.
A defensible bridge separates the charge into what recurs and what does not. Suppose the £1.5m total SBC charge in the year breaks down as follows.
| SBC component | Amount | Nature | QoE treatment |
|---|---|---|---|
| Annual RSU programme (ongoing) | £0.9m | Recurring remuneration | No add-back; keep in EBITDA |
| One-off IPO-readiness option grant | £0.4m | Non-recurring, event-driven | Add back as non-recurring |
| Founder growth shares crystallising at deal | £0.2m | Deal-contingent, scheme dies at completion | Add back, flag in bridge and SPA workstream |
| Total | £1.5m | £0.6m add-back, £0.9m retained |
This is the analysis a partner wants to see. Not a binary "add it all back / add none of it back," but a component-level view where the recurring core stays in earnings and only the genuinely one-off or deal-contingent slices are stripped out. It mirrors the discipline you apply to any EBITDA adjustment: recurrence is the test, and you evidence each bucket from grant agreements, board minutes and the option-scheme register.
Two further mechanics matter. First, if you do retain SBC as a cost, be consistent about the run-rate: a fast-growing company may have granted far more equity late in the year than early, so the exit run-rate charge can exceed the reported annual figure. Model the forward cost, not just the historical one - the same logic you would apply to any run-rate EBITDA exercise. Second, watch the interaction with net debt: crystallising founder equity at completion may be settled partly in cash, in which case that cash payment is a debt-like item, not an EBITDA matter at all. Do not double-count it in both places.
The right answer is not identical across every business.
In a high-growth technology or SaaS company, SBC is often a material, structural part of the cost base - sometimes 10–20% of payroll. Adding it all back produces an "adjusted EBITDA" that bears little relationship to the cash cost of running the company, and sophisticated buyers will reject it out of hand. Here, the default of retain as a cost is strongest.
In a mature industrial business with a modest, stable long-term incentive plan for a handful of executives, the charge is smaller and steadier. It still recurs, so it still generally stays in earnings, but the stakes are lower and the argument less heated.
In a private-equity carve-out or founder exit, the historical scheme frequently does die at completion. This is the one scenario where a full or partial add-back can be genuinely defensible - but only if you can evidence that the new owner's replacement incentive plan is materially different, and only if you separately assess what that new plan will cost going forward. Cross-reference this with the carve-out diligence workstream, because incentive schemes are exactly the kind of item that behaves differently on a standalone basis.
Bolded takeaway: the sector and the deal structure decide how hard the seller's "scheme dies at completion" argument can push. In tech, almost never. In a clean founder exit, sometimes. Always evidence it.
Expect this to come up as a scenario or a "how would you treat…" question. Interviewers use SBC precisely because the lazy answer and the right answer sound similar for the first five words. Here is how a strong candidate handles it.
"My starting point is that stock-based compensation is a non-cash cost, but non-cash does not mean it should be added back - the test for an EBITDA adjustment is whether the item is non-recurring or abnormal, not whether cash moved. For most businesses, and especially for tech and high-growth companies, SBC is genuine, recurring remuneration; if the company stopped granting equity it would have to pay more cash to keep its people, so I'd default to leaving it in EBITDA as an operating cost. What I'd actually do is get the option-scheme detail and split the annual charge into components: the ongoing annual programme stays in earnings, while any one-off, event-driven grant - say an IPO-readiness award - or founder equity that crystallises and is cancelled at completion can be added back, because it won't recur under new ownership. I'd also check the forward run-rate, since a growing grant pool means the historical charge understates the go-forward cost, and I'd make sure any deal-contingent equity settled in cash is captured in net debt rather than double-counted in EBITDA. So my answer isn't 'add it back' or 'don't' - it's that recurring SBC stays in, and only the genuinely one-off or deal-contingent slices come out, with each one evidenced from the grant agreements."
That answer works because it leads with the right principle, rejects the common trap out loud, and then shows the component-level mechanics rather than a slogan. It signals you have actually done this on a live deal.
Stock-based compensation is the classic trap for the analyst who mistakes non-cash for free. The disciplined view is straightforward once you hold the distinction: recurring equity remuneration is a real cost of running the business and stays in EBITDA; only the one-off, event-driven or deal-contingent slices earn an add-back, and each one has to be evidenced and reconciled. Do that, model the forward run-rate, and keep deal-contingent cash out of the earnings line and inside net debt.
Remember the £16.5m in our worked example. At a double-digit multiple, "it never touched the bank account" is not an argument - it is a price. Make sure you are the one deciding which way it moves.
The Transaction Services Interview Programme (€119.99, one-time) includes a dedicated module on non-cash and equity-settled adjustments - SBC, IFRS 2 basics, the recurrence-versus-cash test, and how to defend a component-level add-back under partner challenge. Enrol today.
Hundreds of candidates prepared their interviews with this programme. Those who landed the role have one thing in common: they worked the cases before walking into the room.
