UK Tech and the IHT Reform — Plain English Detailed

The full plain English version, with international comparators and four scenarios. About 3,000 words.

What this is about

In April 2026 the UK government changed how inheritance tax works for people who own shares in unlisted private trading companies. The change has been argued about loudly, and the arguments often miss what the actual disagreement is. This piece tries to explain it clearly, focusing on the UK tech cohort the reform most distinctively affects — founders, angels, VCs, and the funding stack that supports UK tech companies.

Three serious positions have been advanced by knowledgeable people in good faith. They agree on more than they disagree on. The disagreement is narrower than the public debate makes it sound. This piece sets out what each position is, what they share, what they actually disagree about, and what would help us choose between them.

It does not tell you which position is right. The author is personally affected by the reform and is open about that. The piece tries to give you the materials to form your own view.

What changed in April 2026

Before April 2026, if you died owning shares in a private trading company, those shares were generally not subject to inheritance tax. The relief that exempted them was called Business Property Relief. It had been part of the UK tax system since 1976 and applied without limit. Founder equity in a private company, family business shares, and certain other unlisted shareholdings could pass to the next generation tax-free, no matter how valuable.

From 6 April 2026, the relief is capped. A new £2.5 million allowance applies to the combined value of qualifying business and agricultural assets in an estate that get 100% relief; everything above that gets only 50 per cent relief, taxed at the standard inheritance tax rate of 40 per cent — an effective 20 per cent on the excess. The allowance is transferable between spouses and civil partners, so a couple can pass on up to £5 million of qualifying assets at 100% relief. It refreshes every seven years for individuals (ten for trusts), and is set to be CPI-indexed from April 2031 (subject to a future government implementing it via statutory instrument). It is in addition to the standard nil-rate bands a couple can already use, so the total they can pass on tax-free can reach about £5.65 million.

Tax due can be paid in ten annual instalments, in equal instalments over 10 years, interest-free.

HMRC's December 2025 estimate is that approximately 1,100 estates in total will pay more inheritance tax in 2026-27 as a result of the wider APR/BPR reforms. Of those, around 185 are estates claiming Agricultural Property Relief, and around 220 are "BPR-only" estates that hold qualifying unlisted trading-company shares of the kind this piece focuses on. (The remainder are mainly estates affected by the separate change to AIM-listed shares, which are now treated differently.) So the BPR-only cohort whose tax position this analysis addresses — founders, family-business owners, holders of unlisted private-company shares above the allowance — is around 220 estates a year, within a wider reform that affects around 1,100. The wider planning population whose lifetime decisions are affected (living owners of substantial unlisted shares, partners in venture capital and private equity funds, EIS investors) is several thousand households.

Why this is contested

The principle of the change is not really in dispute. Almost no other comparable country gave unlisted business assets unlimited inheritance tax relief. Most have either lower allowances and structural carve-outs for genuinely-operating businesses, or no inheritance tax but a different kind of capital tax instead. The pre-2026 UK position was an outlier, and bringing very large private holdings into the tax base above an allowance is a defensible direction of travel. All three positions discussed below accept this.

What is contested is the way the tax is collected. The mechanism the reform uses is to value the shares at the date of death and charge tax on that valuation. For shares in a private company that has not been sold or floated, this raises an obvious problem: there is no observable market price. Whoever pays the tax has to negotiate the value with HMRC, which is often a multi-year dispute. The asset itself is illiquid, so the tax has to be paid out of borrowing, dividends, or eventually a forced sale of the shares. And the people most affected — usually founders or families who built the business — have a clear option, which is to leave the UK before they die.

So the question is not whether the principle of the reform is right. The question is whether the death-event valuation mechanism, applied to this particular kind of asset, produces a regime that holds up. Three answers have been seriously argued.

The three positions

Position A — keep the system as it is, fix the rough edges

Position A says the reform is correct in principle and the operational issues are manageable. It would keep the death-event mechanism but adopt four practical fixes that close avoidance routes and reduce the dispute caseload at HMRC. It would not change the underlying mechanism for one asset class because doing so would be a special carve-out for the wealthiest holders.

The argument for Position A is that the affected cohort is small (HMRC's December 2025 estimate is approximately 1,100 estates affected by the whole APR/BPR reform per year, of which around 185 include an APR claim and the BPR-only unlisted-trading-company-share subset on which this piece focuses is in the low hundreds within that), the principle of fairness across asset classes is important, and the four practical fixes — explained below — address most of the genuine operational problems. Mechanism change carries political risk that the practical fixes do not.

The strongest version of Position A goes further. It says: the operational-mismatch argument is the same argument the wealthy have always made about every tax that affects them. Every tax produces friction. Every tax produces valuation disputes. Every tax produces behavioural responses among those most affected. The argument that this particular tax is uniquely unfit for this particular asset class is what advocates of every tax targeting the wealthy have always argued — that their situation is special, their assets uniquely illiquid, their behavioural response uniquely large. The historical record of these arguments is that they are almost always overstated and almost always advanced by the people most affected by the tax. Position A's strongest form is: this is that argument again, and we should treat it the way we have always treated such arguments — by adopting the practical fixes, resourcing administration, and seeing what the actual data shows over time, not by capitulating to predictions of behavioural response made by the people whose behaviour is being predicted.

The argument against Position A is that it does not address the most important problem with the reform: people who can leave the UK before they die can avoid the tax entirely, and the people who are most able to do this are exactly the people the reform is meant to capture. The four practical fixes help but do not solve this. The counter-argument is that the same was said of every previous tax on mobile wealth, and the actual relocation rates have usually been smaller than the people most affected predicted. Which framing is correct in this case is an empirical question the analysis cannot fully resolve.

Position B — change the mechanism for this kind of asset

Position B says that for shares in unlisted UK trading companies, the death-event valuation mechanism is fundamentally mismatched to the asset. Instead of taxing at death on a contested valuation, Position B would charge tax only when the heir actually sells the asset, at the price they get for it. The tax would be capital gains tax, not inheritance tax, but it would collect on the same underlying value at the same effective rate.

The argument for Position B is that it removes the worst features of the current regime. There is no contested death-date valuation because the tax falls on actual sale proceeds. There is no liquidity problem because the tax falls when there is cash to pay it. There is much less incentive for the original owner to leave the UK because the tax falls on the heir, not on the estate. Australia has run a comparable regime for forty years.

The argument against Position B is that it is structurally unequal — it treats one asset class differently from others brought into the inheritance tax base by the same reform — and that the mechanism is harder to design well than its supporters acknowledge. If the heir holds the asset for fifteen years without selling, what then? A long-stop rule that crystallises the tax after a defined period exists in most proposed versions of Position B, but the long-stop has its own valuation and liquidity problems.

There is also a structural feature of Position B that the public debate has not engaged with and that is worth being honest about. If Position B's long-stop is deferrable until the heir actually sells the shares — which most workable versions of Position B require — the state ends up holding an unfunded contingent claim against private company equity for potentially decades. Functionally, the state has economic exposure to the company's outcomes (upside through eventual CGT collection if the company succeeds, total loss if it fails) without any governance rights. Across the affected cohort over time, the state ends up with that contingent exposure across hundreds, eventually thousands, of UK private companies — selected by which founders or significant shareholders happen to die holding them. This is structurally a different relationship between state and private business than tax collection. It may be acceptable on industrial-strategy grounds, or even desirable if the state's incentives align with the companies' success. It may be undesirable as a quiet expansion of state economic interest in private business that has not been deliberated as such. Either way, it is a feature of Position B that the proposal's supporters have not made explicit, and a serious decision on Position B requires this question to be asked in public.

Position C — adopt the fixes, defer the mechanism question

Position C says: do the four practical fixes everyone agrees on, commission proper modelling on what is actually happening, and decide on the mechanism question later when there is better evidence. This is a hybrid: it accepts the reform in principle, accepts the four fixes, and explicitly leaves open whether the mechanism will eventually need to change.

The strongest version of Position C goes further. It commits the government, in advance, to switch to Position B's mechanism if the evidence at a defined review point shows that the current regime is not working. The trigger conditions are published. The legislation for Position B is drafted in parallel so that the switch can happen quickly if the trigger is met.

The argument for Position C is that it lets the government act on evidence rather than on advocacy. If the evidence at the review point supports holding the current regime, the regime stands. If the evidence supports change, the change happens without another round of consultation. The framework binds the government to the right answer under either outcome.

The argument against Position C is that the trigger may never fire even when the evidence supports it, because governments find ways to argue that thresholds are not really breached, or that the data needs revisiting, or that the next opportunity is not the right moment. The framework is only as good as the political commitment to honour it.

What everyone agrees on

The four practical fixes are common ground across all three positions. They would be adopted under any of A, B, or C. They are:

Closing the borrowing loophole. Under the current regime, an heir who inherits unlisted shares can use them as collateral for a large loan, take the cash, and never pay the tax — the asset stays in the family, the loan repays from dividends, and the tax is deferred indefinitely. This is sometimes called "buy, borrow, die." A rule that triggers a tax charge when inherited shares are used as collateral above a threshold would close this route. Cost to the Treasury: nothing. The rule only bites where avoidance is being structured.

Publishing a valuation methodology. Most of the dispute caseload at HMRC's Shares and Assets Valuation team is about how to value private company shares — what discount to apply for lack of marketability, how to handle preference share stacks, how to aggregate related holdings. HMRC currently has no published methodology, so each estate negotiates from scratch. A statutory safe-harbour methodology — adopt the formula, get fast-tracked approval — would reduce dispute volume without changing the substance of any valuation.

Tightening the rules on temporary non-residence. A UK resident who leaves the country can return after five years without the assets they sold during their absence becoming taxable in the UK. This window should probably be longer for high-value cases — ten years rather than five — combined with stronger enforcement on UK-incorporated companies. This captures pre-death relocation that is currently a route around the regime.

Practical guidance on company buy-backs. Where a company has cash flow, it can buy back its own shares from the estate to fund the inheritance tax bill. The legal mechanism for this exists but the guidance is patchy. Clear guidance from HMRC and the Treasury would help estates that can use the route. It is honest to note that this route does not work for early-stage companies that don't generate cash, or for limited partner interests in venture capital funds.

None of these is contested. The political debate has often skipped over the agreement to focus on the disagreement. That is a mistake — the four practical fixes would significantly improve the regime regardless of which of the three positions ultimately prevails.

What the disagreement actually turns on

If you can answer the following five questions, you can probably answer the policy question. Most of the public debate has been conducted as if the questions were already answered. They are not.

How many people are leaving the UK because of the reform? If the number is small, Position A is broadly vindicated. If the number is large, Position B's case strengthens. The data we have is partial — see the next section — but it is not zero.

How quickly is the dispute caseload at HMRC growing? If it is growing in line with the number of estates, Position A's resourcing solution works. If it is growing faster — because each dispute is multi-year and consumes disproportionate capacity — the system may not scale. This is partly a question about HMRC, partly a question about how fast the affected cohort is growing.

How effective is the borrowing-loophole rule against sophisticated structuring? If it cleanly captures the buy-borrow-die route, that route is closed under any of the three positions. If lawyers can route around it, the rule is more limited than its supporters think.

Can the long-stop rule in Position B be made to work? If the heir holds the asset for ten or fifteen years without selling, does the long-stop produce a tax bill they can actually pay? The honest answer is: only if the long-stop is paired with a payment-on-realisation provision, and even then the design is harder than its supporters acknowledge.

What is the regime for? This is the question the data cannot answer. If the primary purpose is revenue, the position that collects more wins. If the primary purpose is fairness across asset classes, Position A is uniquely consistent. If the primary purpose is to support the businesses the country says it wants to grow, Position B is most consistent. The question is political, not empirical, and ministers have not stated which they think is primary.

Reasonable people, working in good faith with the same evidence, reach different conclusions on these questions. That is most of what produces the disagreement between the three positions. A reader who reaches a clear view on the questions has, by implication, reached a clear view on the policy.

What is already happening

The behavioural response to the reform did not start in April 2026. It started when the reform was announced in October 2024. By the time the regime took effect, eighteen months of pre-positioning had already occurred. This matters for any analysis that uses post-April-2026 data: the most mobile members of the affected cohort have already, at least in part, left the dataset.

The Office for Budget Responsibility — the official UK government independent fiscal watchdog — published its costing of the related non-dom reform in January 2025. That costing assumes 25 per cent of non-doms with excluded property trusts will leave the UK as a result of the reform, and 12 per cent of other non-doms. These are the official assumptions in the Treasury's own scoring document for the non-dom reform — not for the BPR reform. They are useful as a reference point but should not be read as a direct estimate for the BPR-affected cohort. Founder equity in a UK trading company has different mobility characteristics from offshore trust assets — the company, its customers, and often its team are in the UK, which is harder to relocate than an offshore trust structure. The honest reading is that the OBR's figure tells us the UK government considers double-digit departure rates plausible for highly mobile wealthy populations under tax reform, but it does not tell us what the rate will be for the specific cohort this analysis is about.

The Financial Times analysed UK Companies House records — the public register of UK companies and their directors — and found that 3,790 UK company directors changed their primary residence to abroad between October 2024 and July 2025. The same period a year earlier produced 2,712. April 2025 alone saw 691 director departures, 79 per cent above April 2024 and 104 per cent above April 2023. About 150 of the directors who left between April and June 2025 went specifically to the United Arab Emirates. This data needs to be read carefully. The 3,790 figure is much larger than the BPR-affected cohort (around 220 estates a year) — most of these directors are not BPR-affected at all. The period also coincides with multiple other tax changes (the non-dom reform's effective date, capital gains tax changes, carried-interest changes), so the data cannot isolate the BPR-driven component. What the data shows is that wealthy UK residents in director-class roles were relocating in significantly higher numbers in 2024-25 than in 2023; it does not show that the BPR reform specifically caused this. The data is suggestive of a broader behavioural environment, not direct evidence about the cohort this piece is concerned with.

Sifted, the European technology trade publication, reported in May 2025 that four UK tax-advisory firms — Wilson Partners, Evelyn Partners, Founders Law, and Capital Partners — confirmed a marked uptick in UK-based tech-founder enquiries about Dubai relocation. Founders Law specifically reported that UAE relocation now features in 15–20 per cent of all new business enquiries received by the firm.

Named individual departures or planned departures include the founder of Revolut, who has moved to the UAE; the co-founder of Improbable (a technology company last externally valued in 2022 at approximately $3.4bn), who has been reported as preparing to emigrate, citing a mooted exit-tax rather than the BPR reform specifically; and a number of named non-tech billionaires.

A widely-cited figure of 16,500 millionaire departures in 2025, published by the consultancy Henley & Partners, has been forensically critiqued by Tax Policy Associates and by Tax Justice Network. The critiques document methodology problems serious enough that the figure should not be cited as evidence by anyone serious. The narrower findings — the OBR's own assumption, the Companies House data, the named cases — are not subject to those objections.

What this evidence does and doesn't tell us

The evidence shows clearly that pre-positioning is happening. It does not show how much of the wider cohort will eventually leave, or how much of the leakage is permanent versus precautionary. It does not show whether the people leaving are the marginal cohort the reform was designed to capture or the central cohort it was designed to retain. Honest answers to those questions need modelling that has not been done, or has been done and not published.

What the evidence does tell us is that any framework — including Position C — that waits for post-April-2026 data to drive a subsequent decision is calibrated against the wrong baseline. The early movers have already, at least in part, exited the dataset. So measured behavioural response will systematically understate true behavioural response. This does not invalidate the framework, but it changes how confidently we can read what it shows.

How other countries do this

This section sets out how comparable economies tax intergenerational transfer of business assets. The principle for which countries are included: G7 economies plus other significant developed economies with both inheritance taxation regimes and substantial private-business sectors. Australia, Canada, the United States, Germany, France, Japan, South Korea. Switzerland and the UAE are mentioned as relocation comparators rather than policy comparators. The list is meant to give a representative range of how serious economies handle this question, not to make a particular case.

Australia has no inheritance tax. Capital gains tax is charged when the heir actually sells the inherited asset, at the heir's eventual sale price. The deceased's original cost basis carries through. This is the closest precedent for Position B and has been operating for forty years.

Canada has no inheritance tax in name but treats death as a deemed sale of all capital assets at fair market value, charging capital gains tax on the gain. The heir then inherits at the new stepped-up value. This is structurally the opposite of Position B — the tax falls earlier rather than later — and combines with strong spousal rollover provisions and life-insurance funding.

The United States changed its position in July 2025 under the One Big Beautiful Bill Act. The federal estate tax exemption is now $15 million per person and $30 million per couple, made permanent and indexed to inflation. The step-up in basis on death is unchanged. In practical terms, US federal estate tax exposure now begins at around £11–12 million equivalent per person — far above the UK's new £2.5 million threshold.

Germany has inheritance tax with conditional carve-outs for genuinely-operating businesses. Standard relief exempts 85 per cent of qualifying business value if the business is held for 5 years and payroll is maintained. Optional full relief exempts 100 per cent if held for 7 years with a stricter payroll test. This is the clearest example of combining a death-event tax with structural protection for operating businesses.

France has the Pacte Dutreil regime, which provides 75 per cent exemption from transfer taxes on qualifying business shares, conditional on retention commitments and a management role for one of the heirs. The regime was tightened in the 2026 Finance Act.

Japan has inheritance tax at rates up to 55 per cent, among the highest in the developed world. There is no general business-asset exemption above conditional thresholds; small and medium business owners can defer or reduce inheritance tax through specific reliefs but only under strict succession-planning and employment-retention conditions. Japan's regime is meaningfully harsher on business assets above the threshold than the UK's new regime.

South Korea has inheritance tax at rates up to 50 per cent, with a controlling-shareholder premium that pushes the effective rate higher for owners of substantial company shares. The combined effect can produce effective rates above 60 per cent on the largest holdings. Korean family businesses have, in fact, frequently restructured or sold to manage these liabilities — but the regime has not been softened in response. South Korea's example is a counterpoint to the assumption that mobile founder-owners will always force regimes to soften.

Switzerland has no federal inheritance tax. The UAE has no inheritance tax. Both are frequently named as destinations for UK relocators.

The honest reading. The UK's combination is unusual but not uniquely harsh. Australia, the United States, Switzerland, and the UAE treat business assets more leniently than the UK now does; Japan and South Korea treat them more harshly. Germany and France use conditional carve-outs that the UK could in principle adopt and has chosen not to. The regime is one option among several. The cases that "this is what every other country does" (sometimes used to defend Position B) and "this is uniquely harsh" (sometimes used to attack Position A) are both overstated; international comparison does not resolve the policy question, it just shows the range of choices serious economies have made.

Four things that could happen

How this question resolves depends on what happens over the next two to three years. Four scenarios are worth thinking through. They are not exhaustive — they are the four that the available data and the reasonable range of expert opinion suggest matter most.

The relocation is small and the system runs smoothly. Departures stay below around thirty a year, the dispute caseload at HMRC stays manageable, and receipts come in close to forecast. Position A is broadly vindicated. The four practical fixes plus adequate HMRC resourcing are enough. Mechanism change becomes unnecessary. The political cost of the reform turns out to have been front-loaded — the announced response was sharper than the realised response. Some founders of pre-revenue companies still face genuine difficulty but the cohort is too small to drive policy change.

The relocation is meaningful but bounded. Departures run in the range of thirty to eighty a year, dispute caseload grows somewhat faster than the number of estates, receipts come in five to fifteen per cent below forecast. The picture is mixed. Holding the regime delivers most of the revenue but loses some; switching the mechanism may collect more but at the political cost of being seen to capitulate. The choice is genuinely contested on the empirics. Whether this is acceptable depends on what the regime is for — the question ministers have not yet stated.

The relocation is substantial. Departures run above eighty a year, the dispute caseload grows faster than HMRC can resource, receipts come in more than twenty per cent below forecast, and adviser-survey evidence shows widespread relocation planning. The regime is in trouble. Position B's case strengthens substantially. Mechanism change recovers some receipts going forward, but the cohort that left does not return. The damage to the perception of the UK as a place to build a long-term company is significant and persistent. Notably, the existing Companies House evidence is consistent with this scenario being already partly the realised outcome.

The data is contested or never quite arrives. The likeliest real-world outcome, based on the historical record of UK tax policy reviews. The modelling produces a mixed picture, the trigger thresholds in any conditional framework are challenged on definition, the political environment by month twelve differs from the one in which the framework was set, and the formal review does not produce a clean answer. The regime drifts in its current form for several years before any further change. This is the worst outcome for policy quality — the regime persists not because it is right but because the political conditions for changing it never quite cohere.

What this analysis cannot do

Three honest limits.

The analysis underplays what sophisticated holders actually do. People who own substantial private company shares do not just respond to mechanism design — they restructure to avoid the question entirely. Trusts established years before death, holding-company restructures that move shares out of personal estate, life-insurance funding of any residual liability, residence planning that uses the rules strategically. The analysis above treats the cohort as more responsive to policy and less strategic than it actually is. The Companies House data and the adviser-survey reporting suggest the broader environment is one in which mobile wealthy individuals are relocating at higher rates than recent baselines, but neither piece of evidence isolates the BPR cohort specifically. The honest reading is that the most mobile members of the affected cohort are probably not waiting for the regime to bite before responding, but the direct evidence on this is weaker than the analysis sometimes implies.

The analysis is asymmetric in what it can measure. The friction side of the regime — disputes, delays, valuation discounts — gets specific numbers. The outcome side — relocation, restructuring, capital flight, long-term reinvestment effects — gets qualitative treatment. This is not a choice; it is a limit of the available data. The reader should weight the analysis accordingly.

The normative question — what the regime is for — cannot be settled by any analysis. It is a political choice. The four scenarios above describe consequences for revenue, for fairness across asset classes, and for industrial-strategy alignment separately. They cannot tell the reader which objective should take precedence when they conflict. That is for ministers, voters, and the people most directly affected. This piece is structured to inform that judgment, not to pre-empt it.

What you can do with this

If you have read this far, you have most of what you need to form a view. The five questions above are the questions to keep asking when you read commentary on this issue. The four scenarios are the futures to test any proposal against. The four practical fixes are the common ground that any responsible reform should adopt regardless of which of the three positions prevails.

If you want to go deeper, the article in its tax-policy register is at the URL printed below, with the full version covering the same ground in more technical depth. The policy options paper presents the same analysis in HMT format for readers who work in policy. The political piece sets out what this looks like as a political problem. The honesty piece sets out where the public debate is currently misleading.

If you have a personal stake in the question — you own affected shares, you advise people who do, you are weighing your own residence decisions — this piece is not advice. The analysis is general, the disclosures are open, and the author is personally affected in the same way you may be. A piece written by someone who is personally affected can still be useful, but it should not be a substitute for advice from someone whose only job is the quality of that advice.

The work continues. The publication will return to the question as the data accumulates, and will be honest about which of its current readings turn out to have been right.