Two companies earn the same operating profit in the same year. One keeps a third more of it. Nothing illegal happened, no clever loophole, no offshore letterbox, one of them simply made structural decisions, years earlier, that a tax code rewards, and the other didn't. That gap is the whole subject of tax strategy: not shaving the final bill, but designing the business so the bill is smaller by construction.

The quick version

  • Strategy beats avoidance. Real tax strategy is about how you finance, locate, and time activity, decided up front, not about clever moves after the year closes.
  • Debt has a tax edge, but only up to a point. Interest is deductible and dividends aren't, so borrowing creates a "tax shield", bounded by the rising cost of financial distress.
  • The structuring era is over for aggressive games. A 15% global minimum tax and the arm's-length rule have closed most of the profit-shifting that once worked.
  • Know your real rate. The tax line on your accounts (effective rate) is not the cash that leaves the building (cash rate). Manage both.

The idea in depth

Start with the single most cited result in this whole field, because it tells you where the only durable, legal tax advantage actually comes from.

Debt is cheaper because the taxman helps pay for it

In 1958, Franco Modigliani and Merton Miller published a theorem that, under perfect markets with no taxes, said something startling: it makes no difference whether a firm funds itself with debt or equity, value is value. Five years later they published a correction. In "Corporate Income Taxes and the Cost of Capital: A Correction" (American Economic Review, 1963, pp. 433–443), they added the one thing the first paper had left out: corporate tax. And tax changes everything, because of one asymmetry. Interest paid on debt is deductible from taxable profit; dividends paid to shareholders are not. So every dollar of interest quietly reduces the tax bill, while every dollar of dividend doesn't. That saving is the interest tax shield, and it makes debt structurally cheaper than its headline rate suggests.

What to do with that: when you compare financing options for a project or an acquisition, price debt on its after-tax cost, not its coupon. A loan at 8% interest in a country with a 25% corporate tax rate costs the business roughly 6% after the shield (8% × [1 − 0.25]). That is not a footnote, it is often the difference that flips a deal from marginal to worthwhile, and it is exactly why the after-tax cost of debt sits at the heart of any WACC calculation.

The honest limitation: taken literally, the 1963 result implies a firm should fund itself with almost all debt to maximise the shield, which is obviously wrong, and no serious person believes it. The shield is real but bounded. The dominant correction is the trade-off theory of capital structure, associated with Stewart Myers, which says firms balance the tax benefit of debt against the rising costs of financial distress, the risk, and the real expense, of not being able to pay. Load up too much and lenders charge more, customers get nervous, and a bad quarter becomes an existential one. The tax shield is a tailwind, not a licence.

flowchart LR
  A(["More debt in the mix"]) --> B(["Bigger interest tax shield
(value up)"])
  A --> C(["Higher risk of financial distress
(value down)"])
  B --> D(["Optimal capital structure
= where the two balance"])
  C --> D
The trade-off theory in one picture: debt's tax benefit and its distress cost pull in opposite directions. Leaders Loop

Where you put the work, and what the rules now allow

The second half of tax strategy is structural geography: where legal entities sit, where intellectual property is owned, and how the parts of a group price what they sell to each other. For two decades, multinationals used this to move reported profit into low-tax jurisdictions. That window has largely closed, and a leader who doesn't know it has closed is planning against rules that no longer exist.

Two changes did it. First, the arm's-length principle, enshrined in Article 9 of the OECD Model Tax Convention, requires that when one part of a group sells to another (the classic case being a subsidiary licensing a brand or patent from a parent), the price must be what independent companies would have agreed. You cannot make your Irish unit "buy" your logo from a Bermuda unit for an invented fortune just to strip profit out of Ireland. Second, and more decisively, the OECD's Pillar Two global minimum tax took effect from 1 January 2024 in most adopting jurisdictions. It sets a floor: large multinational groups, those with annual revenue above €750 million, must pay an effective rate of at least 15% in every country they operate in. Pay less somewhere, and a "top-up tax" claws the difference back. The economic point of routing profit to a 2% jurisdiction simply evaporates.

The era of structuring your way to a near-zero tax rate is over. What's left is genuine strategy: financing, timing, and incentives the code actually intends.

flowchart TD
  A(["A cross-border structuring idea"]) --> B{"Is there real
activity behind it?"}
  B -->|"No"| X(["Avoid, fragile under
the arm's-length rule"])
  B -->|"Yes"| C{"Effective rate
at least 15%?"}
  C -->|"No"| Y(["Pillar Two top-up tax
claws back the saving"])
  C -->|"Yes"| Z(["Legitimate, durable
tax strategy"])
A simple screen for any structuring idea after 2024: substance first, then the 15% floor. Leaders Loop

The shift this forces: stop thinking about offshore arbitrage and start thinking about real activity in places that offer legitimate incentives, R&D credits, capital-allowance regimes, patent boxes that require the patent to actually be developed where it's booked. These reward substance, not paperwork, and they survive the new floor. If your group is anywhere near the €750m threshold, the move is also defensive: model your Pillar Two exposure now, because the top-up tax can land in a jurisdiction you weren't watching.

Where to tread carefully: this is the part of the toolkit where a leader should be most humble. Cross-border tax is a regulated, fast-moving, jurisdiction-specific field, and the consequences of getting it wrong are not academic. Apple learned this the expensive way: in September 2024 the EU Court of Justice ordered it to pay Ireland €13 billion in back taxes, ruling that the arrangements amounted to illegal state aid. Use this article to ask sharper questions; use a qualified tax adviser in the relevant jurisdiction to answer them.

Your tax rate is two numbers, not one

A practical trap catches leaders who only read the income statement. The tax line in your accounts gives you the effective tax rate, the accounting expense as a share of pre-tax profit. But that's not the cash that actually left the business this year. The cash tax rate is. The two diverge because of timing: accelerated depreciation, losses carried forward, and other temporary differences mean you often book a tax expense in one year but pay the cash in another. (For a clean walk-through of the effective, statutory and cash distinctions, AnalystPrep's CFA Level 1 notes set them side by side.)

The practical rule: when you forecast cash, for a budget, a covenant, a runway, model the cash tax rate. When you forecast reported earnings, use the effective rate. Confusing the two is how a profitable-on-paper company gets surprised by a tax payment it can't comfortably fund.

A worked example

The figures below are illustrative, chosen for clean arithmetic, not a recommendation or a forecast.

A mid-sized company is buying a competitor for £20m and weighing two ways to fund it: all equity, or £12m of debt and £8m of equity. The corporate tax rate is 25%; the debt carries 7% interest.

  • All-equity: no interest, so no shield. The full cost of capital is the (higher) cost of equity. Clean balance sheet, but the most expensive money.
  • With £12m of debt: annual interest is £840,000 (£12m × 7%). Because interest is deductible, that £840k reduces taxable profit, saving £210,000 in tax each year (£840k × 25%). The effective cost of that debt isn't 7%, it's about 5.25% after the shield.

That £210,000 a year is real, recurring, and legal, it exists purely because the code treats interest and dividends differently. But the trade-off theory is now in the room: £12m of debt also adds fixed repayments the business must make in good quarters and bad. If the acquired company underperforms, that obligation is the thing that turns a disappointment into a crisis. The right answer is rarely "maximum debt", it's the level where the shield is captured and the distress risk is still one you can sleep through. That judgement, not the arithmetic, is the leadership part.

Frequently asked questions

Is tax strategy the same as tax avoidance?

No, and the distinction matters. Tax strategy means making genuine business decisions, how to finance, where to locate real activity, when to invest, in a way that's mindful of their tax consequences, using incentives the law deliberately offers. Aggressive avoidance means engineering artificial arrangements whose main purpose is to dodge tax. The first is good management. The second is increasingly fragile, expensive when it fails, and reputationally toxic.

Should we just take on as much debt as possible for the tax shield?

No. The shield is real but it's only one side of the ledger. The trade-off theory of capital structure says the optimum is where the tax benefit of additional debt is offset by the rising cost of financial distress. Past that point, more debt destroys value rather than creating it, and it's your fixed repayments, not your tax bill, that sink a business in a downturn.

Does the global minimum tax affect my company?

Directly, only if you're part of a multinational group with revenue above €750 million, that's the Pillar Two threshold. Below it, the rules don't bind you, but the principle still should: structuring profit into very low-tax jurisdictions is a strategy with a shrinking shelf life, and customers and regulators are less tolerant of it than they were.

Why does our reported tax rate differ from what we actually paid?

Because of timing. The effective rate on your income statement reflects the accounting tax expense; the cash rate reflects what physically left the bank. Temporary differences, like depreciating an asset faster for tax than for accounts, push the cash into a different year. Both numbers are correct; they answer different questions.

Whose job is tax strategy, finance's or mine?

Finance owns the mechanics; you own the decisions tax attaches to. Financing structure, acquisitions, where to base a new operation, build-versus-buy, these are leadership calls with large tax consequences. The job isn't to become a tax expert; it's to know enough to ask the right question before the decision is locked, not after.

Related in the Toolkit

Where to go next