Assessing the Implications of Inheritance Tax Reforms in the UK

A report for Onward by Oxford Economics. To discuss the report further please contact Alex Stewart: astewart@oxfordeconomics.com.

Foreword

It is a good thing to work hard, to save responsibly and to want to pass on the fruits of hard work to your loved ones when you die. That is how you incentivise risk-taking and wealth creation in an economy, and a more dynamic and successful economy benefits the whole of society.

Yet today, the scope of Inheritance Tax (IHT) in the United Kingdom has increased enormously – striking directly at the heart of those values. IHT liabilities hit a record £7 billion in the latest HMRC data. The standard nil-rate band threshold has been frozen at £325,000 since 2009, and fiscal drag means more and more families are being caught in its net. By 2029, around one in twelve estates will be paying IHT, which is more than triple the proportion of estates than in 2009. If this threshold had risen in line with inflation, it would now stand at £538,000, and the number of estates paying IHT would be much lower.

There has already been considerable controversy around IHT in the last two years, following the botched reforms to agricultural and business property reliefs announced by Rachel Reeves at Budget 2024.

Things will become worse in 2027. From April 2027, the Government is bringing pension funds and death benefits into the value of an estate for IHT purposes, reducing the attractiveness of pension saving. These changes mean unspent pension funds in the UK may face a double tax trap, being subject to both 40% IHT and income tax of up to 45% when withdrawn by beneficiaries at their personal marginal rate. More importantly still, they will also lead to a sharp increase in the number of estates liable to pay IHT.

This autumn, the political debate about the fairness and future of Inheritance Tax is therefore increasingly intense. Onward has decided to provide a crucial new dataset to ensure that this debate is grounded in the strongest possible evidence base. Over the summer, we commissioned Oxford Economics to model a number of possible scenarios for reform of the IHT threshold and the residence exemption. 

This is particularly important given the fact the Government is right at the limit of what it can tax, borrow and spend. We need to make sure that any reductions are costed properly and do not undermine our position in the gilt markets, particularly given ongoing geopolitical and economic tensions.

We are very grateful to Oxford for the work they have undertaken for us, and we hope it will inspire action as part of a balanced package to reduce the burden of taxation across society and the economy.

Sir Simon Clarke

Director, Onward

Executive Summary

Inheritance tax (IHT) liabilities in the UK have risen steadily in recent years, reaching a record £7.0 billion in the latest HMRC data, while the proportion of estates paying IHT has increased to its highest level since the mid-2000s. Looking ahead, a combination of frozen tax-free thresholds, rising asset values, and the planned inclusion of unspent pension wealth within IHT from 2027 is expected to drive a further expansion of IHT revenues and an increase in the number of estates caught by the tax.

Against this backdrop, Onward has commissioned Oxford Economics to model the implications of six potential reforms to the UK’s inheritance tax system. In these scenarios, changes are made to the standard nil-rate band—which has sat at £325,000 since 2009—and the residence nil-rate band—which has sat at £175,000 since 2020—all other elements of IHT policy remain unchanged (including planned changes to extend IHT to unspent pension pots at death in 2027/28).

The modelling explores the impact of each scenario on IHT liabilities and the number of estates in scope of IHT relative to both a no-change baseline and a full abolition of IHT which would reduce liabilities and estates in scope to zero. While uncertainty driven by assumptions around human behaviour and market forecasts are inherent in this type of analysis, the model aligns closely with past data and OBR forecasts. In the baseline, IHT liabilities are estimated to reach £12.6 billion by 2029/30 and 52,100 estates are expected to be liable for IHT, The key findings from these scenarios are as follows.

• Increasing the nil-rate band to £500,000 is estimated to reduce IHT revenues by around £3.0 billion in 2029/30 to £9.6 billion, while lowering the number of liable estates by approximately 40% to 31,000.
• A more substantial increase to a nil-rate band of £1 million is estimated to reduce IHT liabilities by more than half to £5.8 billion in 2029/30, while lowering the number of liable estates to around 12,000.
• A further increase in the nil-rate band to £2 million is projected to reduce expected liabilities in 2029/30 to £3.2 billion and cut the proportion of estates subject to IHT to around 4,000.
• A full exemption for primary residences is estimated to cut liabilities by just over 30% to £8.6 billion and lower the number of estates liable for IHT to 36,000.
• Combining a full exemption for primary residence with an increase in the nil-rate band to £500,000 sees liabilities fall to £6.6 billion in 2029/30 with around 22,000 estates facing IHT.
• While a full exemption for primary residence with an increase in the nil-rate band to £1 million reduces revenues to £4.0 billion in 2029/30 and lowers the number of estates liable to 8,000.

The scenario analysis shows that these reforms can return the IHT closer to levels seen following past reforms IHT policy. In the no-change baseline, IHT liabilities are projected to rise to 0.37% of GDP by 2029/30, with 8.2% of deaths resulting in an IHT liability. For example, these levels are substantially higher than those seen following the introduction of the £325,000 nil-rate band in 2009/10, when liabilities were just 0.15% of GDP and only 2.7% of estates were subject to IHT. If policymakers are aiming to return to these levels, raising the nil-rate band to £1 million would reduce IHT liabilities to 0.17% of GDP and lower the proportion of estates liable to IHT to 1.9%, bringing both measures back to levels broadly consistent with those seen when the £325,000 band was first established. 

Overall, the research highlights the trade-off facing policymakers. Higher thresholds would substantially reduce the number of families affected by IHT and move the tax closer to its historic scope, but would come at the cost of lower revenues for the Exchequer. As IHT liabilities continue to grow under current policy settings, so too will the debate around how to balance revenue generation against the increasing burden IHT imposes on households. As this analysis clearly shows, leaving thresholds unchanged is a policy decision of consequence.

Introduction

In 2023/24, inheritance tax (IHT) liabilities stood at £7.0 billion, their highest-ever level, while the proportion of deaths resulting in an IHT charge stood at 4.7%, the highest rate since the all-time peak of 6.0% seen in 2006/07.[1] These trends have been driven by a combination of continued rises in asset values and the government’s decisions to maintain the standard nil-rate band at its 2009 level and the residence nil-rate band at its 2020 level.

At the same time, the government has recently introduced restrictions to agricultural and business property reliefs, and plan to extend IHT to unspent pension pots at death from April 2027. This broadening of the IHT tax base combined with frozen nil-rate thresholds until at least 2030/31 has led the OBR to forecast receipts to rise to nearly £15 billion by 2030/31.[2]

In the context of increasing IHT revenues and more estates becoming subject to IHT, Onward have commissioned Oxford Economics to assess the implications of the following reforms to the UK’s current IHT policy relative to both a no-change baseline and full abolition of IHT.

• Nil-rate band reforms—increasing the nil-rate band from its current level of £325,000 to £500,000, £1 million, and £2 million.
• Residence nil-rate band reform—a full exemption on an individual’s primary residence.
• Combined scenario A—full exemption on an individual’s primary residence, plus increasing the nil-rate band to £500,000.
• Combined scenario B—full exemption on an individual’s primary residence, plus increasing the nil-rate band to £1 million.

In each scenario, all other existing aspects of the current IHT policy remain as they are at present or reflect planned changes in the policy (e.g., extending IHT to unspent pension pots at death in 2027). Deceased spouses can transfer any unused allowances to their surviving spouse at death (e.g., if the threshold is raised to £1 million per individual, then the combined threshold for a married couple would be £2 million) in all scenarios. The residence exemption is only available when the property is passed on to a direct descendant (a child or grandchild).

This research explores the impact of the scenario on the public finances and the proportion of deaths resulting in an IHT charge at the national level. Throughout the report, estimates are compared against historic levels to contextualise the findings. The scenarios are modelled to come into effect in 2027/28, and the findings are forecast out to the end of the current parliament.

While the baseline forecast for IHT revenues aligns closely with forecasts from the Office for Budget Responsibility (OBR), there is significant uncertainty surrounding any IHT forecast. This uncertainty is driven by both the underlying modelling assumptions and the Office for National Statistics (ONS) data that underpin this research. However, the key insights lie not in the absolute values but in the relative differences in revenues and the proportion of estates that are liable across the different policy scenarios.

Modelling approach

Model overview

The core dataset that underpins the analysis is the Wealth and Asset Survey (WAS) from the ONS. The WAS is a nationally representative survey that collects detailed information on the financial position of over 17,000 households across Great Britain.[3] The analytical framework used in this research closely aligns with the approach used by the Institute for Fiscal Studies (IFS) in their Reforming Inheritance Tax report.[4]  

There are three core elements to the model:

• Net wealth at death—wealth held by individuals is projected forward using a range of asset-specific assumptions to estimate the change in the value of different asset and liability categories over time. Assumptions regarding asset drawdown and saving are applied and vary according to the age of each individual.
• IHT liability at death—the IHT liability at death in each year for a household is estimated based a range of policy relevant characteristics such as relationship status, presence of children, as well as the composition of their net wealth.[5]^,[6],[7]
• Probability of death—ONS data on the probability of death are used to assign a likelihood that each individual in the WAS dies without a surviving partner based on their age and sex, and the age and sex of their spouse (if they have one).[8] The survey weight of each respondent is adjusted each year according to their probability of survival for the previous year.

For every individual in the dataset, their tax liability, were they to die in a given year, probability of death, and survey weight is combined to estimate overall tax receipts by year and the number of estates subject to IHT.

The model incorporates an assumed behavioural response by individuals to the proposed policy reforms. In response to these proposed policy changes, which would decrease an individual’s tax liability, there is less incentive for individuals to actively reduce the size of their estate subject to IHT (e.g., through tax-planning). As such, we assume that the average size of estates would be slightly higher than without the proposed policy changes.[9]

The WAS underestimates wealth at the very top of the wealth distribution. In 2019, the wealthiest individual in the WAS has a net wealth of around £30 million, but the Sunday Times Rich List indicates that there are 1,000 individuals with wealth of at least £120 million.[10]^{, [11]} Using insights from the Sunday Times Rich List, a pareto adjustment—a common technique used for this type of analysis—has been applied to better capture the very top of the wealth distribution. Doing so increases aggregate net wealth by 5.3%, a similar magnitude to that estimated by the IFS.[12]

Model testing

The model has been back-tested against outturn data from HMRC. Figure 1 shows the model’s baseline estimates for IHT liabilities and the proportion of estates subject to IHT compared to HMRC data. These figures illustrate that over the 2019/20 to 2023/24 period, the model closely tracks outturn data from HMRC. [13] This provides confidence in the robustness of the modelling framework and the scenarios that are analysed in the following sections.[14]

Figure 1: Back-testing model results against HMRC

The model’s no policy change baseline forecast for IHT liabilities has also been compared against the OBR IHT revenue forecast (Figure 2). By 2029/30, the model forecasts IHT revenues of £12.6 billion compared with OBR’s forecast of £13.7 billion, a difference of 8%. This difference reflects both differing geographic scope – OBR estimates relate to the UK, while our model covers Great Britain only – and minor differences in macroeconomic and financial market forecasts used in our analysis 

Given the uncertainties surrounding IHT revenue forecasting and differences in geographic scope, these estimates can be considered broadly aligned. Importantly, our revenue forecasts are slightly lower than the OBR’s and therefore the revenue estimates in the scenario analysis in the following section would be marginally higher if the model was fully aligned with the OBR’s estimates.

Figure 2: Comparing the model’s baseline forecast with IHT revenue forecast by the OBR

The interested reader can find further information on the assumptions that underpin this analysis in the appendix.

Scenario analysis

This section explores the impact of changes to IHT policy on IHT liabilities and the number of estates liable to IHT up to 2029/30. This section first covers the impact of changes to the nil-rate band only. It then proceeds to assess the impact of changes to the residence nil-rate band in isolation, and finally combinations of the two policies.

Changes to standard nil-rate band

Increases to the nil-rate band from its current level of £325,000 would reduce the revenue raised through IHT and the number of estates in scope of the tax.

In the baseline, IHT liabilities are expected to reach £12.6 billion by 2029/30 (Figure 3). An increase to a £500,000 nil-rate band is estimated to result in IHT liabilities of £9.6 billion in 2029/30, representing a decrease in IHT liabilities of £3.0 billion relative to the baseline. An increase to a £1 million threshold is estimated to result in a IHT liability of £5.8 billion in the same year, while a further increase to a £2 million threshold leads to a liability of £3.2 billion. 

Figure 3: IHT liabilities across different levels of nil-rate band (£ billion)

The standard nil-rate band has remained fixed at £325,000 since 2009/10, when IHT revenues were equivalent to 0.15% of GDP. In the baseline scenario, in which there are no changes to the UK’s IHT policy, IHT as a percentage of GDP is expected to reach 0.37% in 2029/30, or 0.22 percentage points above the level seen in 2009/10 (Figure 4). An increase in the standard nil-rate band to £1 million is estimated to result in IHT liabilities as a share of GDP of 0.17% by 2029/30, which would bring IHT policy more in line with where it was the last time the standard nil-rate band was changed.

Figure 4: IHT liabilities across different levels of nil-rate band (% of GDP)

In the baseline, the number of estates liable to IHT is predicted to reach 52,100 in 2029/30 (Figure 5). Implementation of a £500,000 nil-rate band is expected to decrease by the number of estates liable for IHT by 40% to 31,000. Increases in the nil-rate band to £1 million and £2 million would further decrease the number of liable estates to 11,800 and 3,900, respectively.

Figure 5: Number of estates liable to IHT across different levels of nil-rate band

When the standard nil-rate band was last changed in 2009/10, only 2.7% of deaths were liable to IHT. In the no policy change baseline, that figure is expected to hit 8.2% in 2029/30. Raising the threshold to £1 million would see 1.9% of estates subject to IHT in 2029/30, bringing the share of estates liable to a similar level that seen in 2009/10. Alternatively, a £500,000 threshold would bring the share of deaths liable to IHT in 2029/30 back to around their historic average seen this century of around 4%.

Figure 6: Share of deaths liable to IHT across different levels of nil-rate band

Full exemption of primary residence and wider reforms

A full exemption of primary residences from IHT would reduce revenue raised through IHT and the number of estates in scope of the tax. A full exemption of primary residence combined with increasing the standard nil-rate threshold will lead to further reductions, while abolition of IHT would reduce revenues and estates in scope to zero.

Full exemption of the primary residence from IHT is predicted to reduce IHT liabilities to £8.6 billion in 2029/30, marking a fall of just over 30% relative to the no policy change baseline (Figure 7). The addition of a £500,000 or £1 million nil-rate band on top of the full residence exemption would further reduce IHT liabilities in 2029/30 to £6.6 billion and £4.0 billion, respectively.

Figure 7: IHT liabilities across scenarios with full exemption of primary residence (£ billion)

A full exemption of the primary residence from IHT is estimated to bring IHT liabilities down to 0.25% of GDP (Figure 8). This would bring it in line with the level of IHT revenues as a share of GDP seen in 2017/18 (0.24%), when the residence nil-rate band was first introduced. A combination of full primary residence exemption and an increase in the standard nil-rate band to £500,000 or £1 million would further reduce IHT liabilities in 2029/30 to 0.19% or 0.12% of GDP, respectively. This would move the IHT revenue share of GDP closer to the level seen when the nil-rate was last changed in 2009/10.

Figure 8: IHT liabilities across scenarios with full exemption of primary residence (% of GDP)

With full primary residence exemption, the number of estates liable to IHT in 2029/30 is estimated at 36,100, down 31% on the no policy change baseline (Figure 9). The number of estates subject to IHT falls to 21,500 and 7,900 in 2029/30 under a combination of full primary residence exemption and an increase in the standard nil-rate band to £500,000 or £1 million, respectively.

Figure 9: Number of estates liable to IHT across scenarios with full exemption of primary residence

In the baseline, 8.2% of estates are expected to be liable for IHT in 2029/30. This figure falls to 5.7% of deaths with a full residence exemption in 2029/30 (Figure 10). Augmenting the policy to also include a £500,000 or £1 million nil-rate band would reduce the share of deaths liable to IHT in 2029/30 to 3.4% or 1.3%, respectively. By comparison, when the residence nil-rate band was introduced in 2017/18, 3.9% of deaths in the UK were liable to IHT.

Figure 10: Share of deaths liable to IHT across scenarios with full exemption of primary residence

Conclusion

The latest data from HMRC show that IHT liabilities stood at their highest-ever level in 2023/24, and that the proportion of estates in scope of IHT was at the highest level since the mid-2000s. Government policy and growth in household wealth is set to see IHT revenues and number of estates affected by the IHT continue to increase over the remainder of this parliament.

This report sets out the implications of a range of IHT reforms using a robust modelling framework. It shows that reforms to the standard and residence nil-rate bands can return the IHT policy closer to levels seen following past reforms. While there is substantial uncertainty about the exact level of revenues raised and estates affected, these findings highlight the trade-offs between reducing revenues and the number of estates subject to IHT.

As IHT liabilities continue to grow under current policy settings, so too will the debate around how to balance revenue generation against the increasing burden IHT imposes on households. In this context, it is important to highlight that leaving thresholds unchanged is a policy decision in and of itself.

Appendix

Probability of death

An individual’s probability of death is based on age- and sex-specific estimates from the ONS. This does not reflect that wealthier individuals are likely to have a lower probability of death than the overall population at any given age, but it is consistent with the approach taken in other research, including the IFS. For each year in the model, the survey weight assigned to each respondent is adjusted based on the survival rate from the previous year. The model captures two dynamics to estimate the number of deaths in each year:

• As individuals get older in the model their probability of death increases.
• As individuals get older in the model the remaining weight for each respondent is smaller.

Net wealth

The model separately estimates the personal and business wealth of each respondent in the WAS to take account of business property relief in the calculations of IHT liabilities. The personal wealth of an individual is calculated as the sum of:

• Financial assets, including current accounts, savings accounts, investment accounts, ISAs, and bonds.
• Property equity, including non-primary residence.
• Non-annuitised pension.
• Physical assets, including household contents, collectables, and vehicles.

Non-mortgage liabilities, excluding student loans, are removed to produce an estimate of net wealth.

Business wealth is estimated as the sum of:

• The individual’s share in the value of the business after deducting outstanding debts.
• Value of unlisted shares.

To extrapolate and then forecast the value of net wealth for each individual, the model accounts for changes in asset value as well as drawdown (or accumulation) of assets over time. For each individual asset and liability, a separate assumption is made regarding the growth in value of that asset/liability.

Assumptions regarding asset drawdown (or accumulation) vary depending on age. Prior to retirement, individuals are assumed to continue to accumulate assets. Once an individual reaches retirement, they are assumed to withdraw the maximum tax-free lump sum from the pension, and a proportion of this is assumed to be withdrawn from the value of their estate (for example, to spend on holidays).[15]^{, [16]} Individuals are then expected to draw down their assets until their death. We assume a 5% real drawdown of non-annuitised pension wealth per year, a 3% real drawdown of financial assets per year and a 1% real drawdown of property wealth per year.[17]

IHT liability at death

The IHT liability at death in each year for a household is estimated based a range of policy relevant characteristics such as relationship status, presence of children, and the composition of their net wealth. It is assumed that individuals will make full use of allowances to reduce their IHT liability. For example, deceased spouses are presumed to leave all their wealth to their spouse and transfer their unused allowances.[18]

In the case of unmarried individuals, the probability of their estate being assessed for IHT liability in each year is equal to their probability of death. For married couples, it is the probability that both partners would be deceased by the end of that year.

The current IHT policy includes a £175,000 residence nil-rate band which allows individuals to pass £175,000 of primary residence wealth to children or grandchildren exempt of IHT. The WAS identifies if a household has children living within the same house, but it does not provide insight on whether an individual has a child living outside of their household. We therefore use insights on the childless rate to estimate the probability that a given individual has children.[19]^{, [20]} IHT liability is estimated as the weighted average of the IHT liability that would be due if they had a child and the IHT liability that would be due if they did not have a child. The number of estates liable to IHT is forecast using the probability the individual has a child along with an assessment of whether they would be liable to pay IHT in the scenario where they have children and in a scenario where they do not.

Uncertainties

As with any modelling framework, there are a range of uncertainties which are detailed below.

• The modelling is based on a series of assumptions about asset and liability growth rates and saving and wealth drawdowns.
• The WAS does not allow for the identification of agricultural assets, although some of these may be part of the business assets in the dataset. The extent to which agricultural assets are taxed differently to wider business assets is therefore not reflected in the model. The model takes the same approach to this issue as the IFS.
• The model does not capture gifts that were given in the last seven years, which would be in scope of IHT in line with the approach taken by the IFS.
• The mortality rates applied in the model may over-estimate the probability of death for the wealthiest individuals.[21] As these individuals are more likely to pay IHT and be in scope for the tax, higher death rates would overstate their contribution.
• The model incorporates an assumption on the behavioural response to the policy changes, drawn from the literature. However, there is uncertainty regarding how individuals would respond to such policies, given they have a wide range of tax planning options that could be used.
• The WAS only covers households in Great Britain while IHT policy applies across the UK. However, HMRC data for 2023/24 show that Northern Ireland accounts for less than 1% of IHT liability in the UK, limiting the impact of this data gap on the report’s findings.

---

[1] HMRC, “Inheritance Tax liability statistics”, accessed August 2026

[2] OBR, “Economic and Fiscal Outlook: March 2026”, accessed September 2026

[3] Round 7 of the WAS is used as the latest round of the WAS (Round 8) does not have person level weights (which are required for the modelling) and does not meet the ONS’s accredited statistic status (due to issues relating to survey response rates over the pandemic period). 

[4] The Institute for Fiscal Studies, “Reforming inheritance tax”, accessed August 2026

[5] It is assumed that individuals will make full use of allowances available to them. For example, a spouse is assumed to leave all wealth to their remaining spouse

[6] It is assumed that the wealthiest 0.4% of those to die give 10% of their estate to charity. This is the same assumption as made by the IFS. Such donations are 100% exempt from IHT and reduce the IHT rate applied to the remaining estate to 36%

[7] The WAS does not allow for the identification of agricultural assets, although some of these may be part of the business assets in the dataset. The extent to which agricultural assets are taxed differently to wider business assets is therefore not reflected in the model. The model takes the same approach to this issue as the IFS.

[8] Nomis, “Mortality statistics – underlying cause, sex and age”, accessed August 2026

[9] Our assumption regarding the behavioural response uses estimates of the elasticity of taxable wealth with respect to the net-of-tax rate for IHT sourced from Advani et al (2021) “Revenue and distributional modelling for a UK wealth tax”, accessed August 2026

[10] Graham Scrambler, “The Sunday Times Rich List, 2019”, accessed August 2026

[11] The Sunday Times Rich List captures the value of global wealth of the wealthiest individuals resident in the UK. As such, not all wealth quoted in the list would be subject to UK IHT.

[12] The Institute for Fiscal Studies, “Reforming inheritance tax”, accessed August 2026

[13] The latest available data point.

[14] HMRC data relate to the UK, while our model relates to Great Britain. However, HMRC data for 2023/24 show that Northern Ireland accounts for less than 1% of IHT liability in the UK.

[15] Individuals can take 25% of their pension pot as tax-free cash, limited to £268,275.

[16] We derive estimates of the share of value of lump sums withdrawn from pension that are removed from the estate using questions asked in the WAS.

[17] These assumptions are consistent with those used by the IFS.

[18] Married, but separated individuals are not assumed to leave their wealth to their separated spouse.

[19] ONS, “Fertility for those born in different years, England and Wales”, accessed August 2026

[20] BBC, “The real reason for the rise in male childlessness”, accessed August 2026

[21] ONS, “Health state life expectancies by national deprivation deciles, England: 2018 to 2020”, accessed September 2026

---