Why the LBO Tax Shield Is Capped: The 30%-of-EBITDA Interest Limit That Denies a Shield to Half the Interest Bill in a Highly Levered Deal

Every LBO Model Assumes a Shield the Tax Code No Longer Fully Grants

Open any LBO model and find the tax line: it multiplies pre-tax profit by the tax rate, and pre-tax profit is struck after the full interest expense. The model is quietly asserting that every pound of interest reduces taxable profit by a pound — the tax shield that makes debt cheaper than equity, and the reason a sponsor levers a deal in the first place. In a lightly geared company that assertion holds. In a leveraged buyout, where the whole point is to load the balance sheet with debt, it stops being true well before the interest line ends.

Since 2017 the major jurisdictions have capped how much interest a company may deduct, regardless of how much it actually pays. The UK introduced the Corporate Interest Restriction (CIR) from 1 April 2017; the US enacted §163(j) in the same Tax Cuts and Jobs Act; the EU rolled out the same limit through the Anti-Tax Avoidance Directive (ATAD) from 2019. All three descend from the same source — BEPS Action 4, the OECD project aimed squarely at companies stripping taxable profit out of high-tax countries with intra-group debt. The target was aggressive multinationals; the collateral damage is every leveraged buyout, because an LBO is, by construction, exactly the high-leverage profile the rule was written to catch.

The Cap Is 30% of EBITDA — and an LBO Runs at Twice That

The fixed-ratio rule in all three regimes is the same headline number: net interest deductions are limited to 30% of tax-EBITDA. Interest above that line is disallowed in the current year. The arithmetic of why this bites an LBO so hard is trivial, which is exactly why it is so often missed.

Take a business doing £100M of EBITDA, bought at 6.0x leverage — £600M of debt, a standard sponsor structure — at a 10% all-in cost once you blend the margin and the OID. That is £60M of interest a year: 60% of EBITDA. The cap sits at 30% of EBITDA — £30M. Half the interest bill is above the line before the model has done anything wrong.

The effect does not stop at the cash line. The whole case for debt rests on its after-tax cost — Kd × (1 − t), the formula that makes 10% debt cost 7.5% after a 25% shield. When half the interest earns no shield, that formula is describing a company that does not exist.

Disallowed Interest Is Deferred, Not Destroyed — Which Is the Whole Argument

The instinct is to treat the lost shield as gone. It is not. Interest disallowed under the cap is carried forward — indefinitely under both UK CIR and US §163(j) — and can be deducted in a later year when the company has spare capacity beneath the 30% line. And spare capacity is exactly what an LBO manufactures over its life, through the cash sweep: every year the debt is paid down, interest falls, EBITDA grows, and the gap between the cap and the actual interest bill widens. By Year 4 or 5 the same company levered down toward 3x may sit comfortably under the cap, at which point the carried-forward interest reactivates and shields profit that would otherwise be taxed.

So the honest framing is not “the shield is lost” but “the shield is a timing problem.” And in private equity, timing is not a footnote — it is the return. Two things make the deferral bite:

The cap, then, converts a chunk of the tax shield from a Year-1 cash benefit into a deferred tax asset of uncertain value. That is a materially worse asset than the model’s clean full-rate shield — and the models that get this right carry the disallowed interest as a separate schedule rather than netting it into one tax line.

The US Made It Worse in 2022 by Switching From EBITDA to EBIT

The 30% number hides a second lever: 30% of what. From 2018 to 2021, US §163(j) measured the cap against adjusted taxable income computed on an EBITDA basis — depreciation and amortisation were added back before applying the 30%. From 2022 that add-back disappeared, and the cap is now struck against EBIT. For a capital-intensive borrower the difference is large.

The same deal, the same interest, the same statutory 30% — and nearly a third less deductible interest, purely because the base dropped from EBITDA to EBIT. A business carrying £30M of D&A loses £9M of additional deductible capacity to a definitional change most models never picked up. It is the clearest reminder that “30% of EBITDA” is shorthand, not a rule — the base matters as much as the percentage.

Two Escape Hatches: The De Minimis and the Group Ratio

The cap is not universal. UK CIR carries a £2M de minimis: a group with net interest below £2M a year is unaffected entirely, and the EU equivalent sets a €3M safe harbour. That threshold is why the cap is a mid-market-and-up problem — a small bolt-on financed with a few hundred thousand of interest never touches it, while a £600M term loan blows through it on day one. It also means the constraint scales with deal size in a way that flatters small sponsors and punishes large ones.

The second relief is the group ratio rule, which lets a highly geared group deduct interest up to its worldwide net-interest-to-EBITDA ratio instead of the flat 30%, where that group is genuinely leveraged against third-party lenders rather than intra-group debt. For a standalone LBO whose entire capital structure is external bank and bond debt, the group ratio can lift the cap above 30% — but it is a calculation, not a default, and it rarely rescues the most aggressive structures. Neither hatch changes the core lesson: above the threshold, the shield is capped, and the model has to say so.

The Verdict: The Shield Is Real, but It Is Rationed

“Interest is tax-deductible” is the first thing a candidate learns about leverage, and in a highly levered deal it is the first thing that is only half true. Since 2017 the deduction has been rationed to roughly 30% of EBITDA across the UK, US and EU — and an LBO, running interest at 50–60% of EBITDA, sails straight past the line. Half the shield the model books arrives late, if it arrives at all: deferred to years when deleveraging has restored capacity, at risk of dying on exit if the hold ends first, and tightened further in the US by the 2022 switch to an EBIT base.

The candidate who says “the debt is cheaper because interest is tax-deductible” has the textbook. The one who says “the deduction is capped at 30% of EBITDA, so in a 6x deal half the interest earns no shield until the company deleverages, the disallowed interest sits as a carryforward that a five-year hold may never fully use, and the real after-tax cost of the debt is closer to 8.75% than 7.5%” is describing the deal a tax structurer actually signs off. The shield is real. It is also rationed — and the ration is where the model and the deal part company.

Take Your Preparation Further

The cap only makes sense once the debt it sits on does. See how the interest bill is built in how a leveraged loan is priced; how deleveraging restores capacity beneath the cap in the LBO cash sweep and debt schedule; why PIK interest is caught by the same limit even though no cash leaves the business; and how the shield feeds the after-tax cost of debt in the WACC calculation. For the broader case for leverage, start with debt versus equity financing.

To build the debt schedule, the interest line and the tax shield inside a live deal model, download the LBO Model Template — PE Ready, and for the financing and valuation mechanics on one page, see the free Valuation Methods Cheat Sheet.

Ready for personalised feedback? Book a 1-on-1 mentoring session with an experienced IB/PE professional.