2.1 Income Tax

In April 2009, the Labour government announced that it would introduce a new ‘Additional Rate’ of income tax on incomes over £150,000, raising the top marginal rate from 40% to 50%. This 10-percentage-point increase was the first time that income tax had been increased for top earners in more than 30 years and was billed as ensuring that the richest shared the burden of the recovery from the Global Financial Crisis. The rate increase took effect from April 2010, but less than two years later (in March 2012), the coalition government announced that the top rate would be reduced to 45% from April 2013. The 50% rate therefore lasted only three years, from 2010–2011 to 2012–2013.

Following the first full year of operation (returns filed for 2010–2011), HMRC published its assessment of the revenue impact of the 50% rate [9]. Although the reform had originally been forecast to bring in an additional £2.4 billion per year (after taking likely behavioural responses into account), the report concluded that in fact it had yielded less than £1 billion, and it was ‘quite possible’ that the yield had actually been negative. This finding led the chairman of the Office for Budget Responsibility to remark that the UK may be ‘strolling across the summit of the Laffer curve’ [10]. In the March 2012 budget, George Osborne supported his decision to reduce the top rate back to 45% by stating that ‘no Chancellor can justify a tax rate that damages our economy and raises next to nothing’ [11].

Why did the 50% rate raise so much less than expected? The main reason, consistently with Slemrod’s hierarchy, had to do with timing effects. HMRC found a staggering 78% year-on-year increase in dividend payments amongst those with total incomes over £150,000 in the year prior to the rate increase (2009–2010), followed by a 73% fall in the year that the 50% rate took effect (2010–2011). Overall, HMRC estimated that between £16 and 18 billion of income was brought forward to 2009–2010 to avoid the 50% rate [9]. The precise extent of this ‘forestalling’ has been debated in subsequent research [12], but by any analysis it was very large. The same behaviours occurred again in reverse when the rate was cut, as individuals deferred dividend payments and other income (where possible) to the year after the reform.

The impact of timing responses has hindered attempts to estimate the underlying overall responsiveness to the reform. Based on data up to 2011–2012 (i.e., prior to the unwinding of the reform), Browne and Phillips concluded that ‘the significant uncertainty around our estimates mean that one cannot be sure that the 50% tax rate did not raise or indeed cost substantial revenues’ [12 p4]. So far there have been no studies that quantify the extent of real responses such as migration, retirement, or hours worked – despite the fact that these margins loom largest in political debates over the impact of raising taxes on the top. The balance of the existing evidence indicates that the 50% rate did not raise much additional revenue overall, but without more research into specific margins of response, it is impossible to know how much more could be raised under different policy conditions.

The most obvious lesson to be learned from the 50% reform is not to announce major tax rises (or cuts) in advance. It is regrettable that this mistake was made not just once or twice but three times in quick succession: as well as pre-announcing the 50% rate and its subsequent reduction to 45%, in 2015 the government pre-announced that it would be increasing the top dividend rate by 7.5 percentage points from April 2016, leading to yet more forestalling. It appears that the Treasury has now gotten the message: in the March 2020 budget, a major increase in capital gains tax (resulting from a reduction in the lifetime allowance for Entrepreneurs Relief) was announced with immediate effect, together with anti-forestalling provisions targeted at those who had already taken (equivocal) steps in anticipation of the increase. Although some industry organisations complained at the lack of advance consultation [13], this approach was clearly merited and should be adopted in future where necessary to prevent well-advised taxpayers from taking pre-emptive steps.

2.2 Non-dom reforms

‘Non-dom’ tax status allows individuals who reside in the UK but who claim that their permanent home (‘domicile’) is abroad to obtain an exemption from UK tax on their foreign income and gains, provided that they do not bring in (‘remit’) these sums to the UK. Non-doms are also exempt from inheritance tax on their foreign assets. The existence of this special tax regime has generated significant public controversy, mostly focused on individual high-profile cases. Using tax data on the full population of non-doms, recent research by Advani et al. (2022) has documented that the regime primarily benefits the super-rich [14]. Around four in 10 individuals with reported income over £5 million have claimed non-dom status, compared with only three in 1,000 of those with incomes under £100,000.

Since 2008, there have been a series of reforms to the non-dom regime, including the introduction (and subsequent extension) of a fixed fee – known as the Remittance Basis Charge – for non-doms who have been resident in the UK for more than seven years. In the 2015 general election, Labour proposed abolishing non-dom status altogether. The proposal suffered a setback with the emergence of a media clip of the Shadow Chancellor Ed Balls dismissing abolition in case it ‘cost Britain money’ as a result of non-doms leaving [15]. Nevertheless, the policy was sufficiently popular that the Conservatives saw it as an area of political vulnerability. To deal with this concern, in the first budget after the election (July 2015), George Osborne announced:

It is not fair that people live in this country for very long periods of their lives, benefit from our public services, and yet operate under different tax rules from everyone else. Non-dom status was meant to be temporary, but it became permanent for some people. Not any longer. I am today abolishing permanent non-dom tax status. Anyone resident in the UK for more than 15 of the past 20 years will now pay full British taxes on all worldwide income and gains … British people should pay British taxes in Britain – and now they will [16].

In fact, the detail of the policy looked rather different. HMRC ‘Technical Guidance’ published simultaneously with the budget speech stated, ‘Non doms who have set up an offshore trust before they become deemed domiciled here under the 15 year rule will not be taxed on trust income and gains that are retained in the trust’ [17]. This provision became known – including in the legislation itself – as the ‘trust protections’. Its effect was that rather than abolishing special tax treatment for long-term non-doms altogether, the government had deliberately preserved a loophole for this group. In particular, by rolling up their investment income and gains within an offshore trust – rather than paying out these sums immediately – ‘deemed doms’ could continue to shelter the returns on their wealth tax-free.

It is currently unclear how many non-doms who were affected by George Osborne’s 2017 reforms actually made use of this loophole. Its use is not entirely costless, because – unlike the position prior to 2017 – non-doms who rely on the trust protections are no longer able to spend their income and gains abroad. For those with enough other sources of funding to cover their worldwide spending, this would be an appealing way to continue living in the UK whilst paying virtually no tax on their investments. To this extent, the reform clearly failed to abolish permanent non-dom status and will have raised less revenue than it could have done if it was not for the loophole.

The lesson from this episode is somewhat different from the 50% reform. It is not the case that the government simply overlooked the trusts response: it was deliberately built into the legislation. In addition, over a long period of time, both Labour and Conservative politicians have accepted the idea that making non-doms pay tax on the same basis as other residents could lead to a mass exodus that would ‘cost Britain money’ despite the total lack of evidence for this concern beyond anecdotes. The failure to secure more radical reform to the non-dom rules appears to stem at least partly from elite capture of the policymaking process. Following the distinction drawn by Fairfield, the persistence of the non-dom regime illustrates not only the ‘instrumental’ power of the super-rich and their advisors and lobbyists but also the ‘structural’ power that they derive from being perceived as both highly internationally mobile and indispensable to the economy [18].

3.1 Capital taxes

Capital gains, gifts, and inheritances are highly concentrated at the top of the income and wealth distributions, so their tax treatment matters a lot for the super-rich. For example, around 92% of all taxable gains, by value, go to the top 1% ranked on total remuneration; 88% go to individuals with total gains exceeding £100,000 [2]. Inheritances are also highly concentrated and tend to go to the already-rich [19]. There is a serious lack of evidence on the distribution of gifts, especially at the very top, because these are typically not subject to tax and are not systematically reported to HMRC.

Although capital gains are currently taxed at much lower rates than regular income, this has not always been the case. From 1988 to 1998, capital gains tax (CGT) rates in the UK were aligned with income tax, with an allowance for inflation. The Mirrlees Review recommended returning to alignment, albeit with an allowance for the normal rate of return, measured as the interest rate on medium-term government bonds, rather than inflation [20]. A move towards aligning rates with an allowance for either inflation or the normal rate of return would be very welcome, but it implies a very large increase in tax on some kinds of gain, particularly those where the asset was acquired at a low (or zero) ‘base cost’.

The size of the behavioural response to aligning rates on income and gains would depend crucially on other policy choices. The most important is that alignment must be accompanied by removal of the uplift at death. At present, the deceased and their heirs are not taxed on any accrued gains. This uplift incentivises individuals to hang on to assets with substantial gains until they die in order to avoid the tax. The incentive would be even stronger if CGT rates were increased and could be expected to significantly reduce revenues as well as be economically distortive.

An issue that is less often discussed concerns how to remove the uplift on death. There are broadly two options. One option, as recommended by the Office of Tax Simplification, is that the person inheriting the asset would be treated as acquiring it at its original ‘base cost’ (i.e., the amount that the deceased had paid for it). This approach means that the gain accruing prior to death would not be wiped out altogether because it would be included in the value of the gain taxed on the inheritor when – or if – they come to sell it. An alternative would be to treat the death as a ‘deemed disposal’ such that tax would be payable on the gain immediately. This option could result in liquidity difficulties for the inheritor, especially when combined with inheritance tax, but it has the advantage of preventing indefinite deferrals of tax through successive gifts.

Although the UK has an inheritance tax with a high headline rate of 40%, in practice this is easily avoided by the super-rich. Analysis by the Office of Tax Simplification shows how the average effective tax rate declines with increasing wealth above £2 million, to just 10% for estates above £10 million [21]. This result is mostly driven by agricultural property relief and business property relief, which broadly exempt farmland and private businesses from IHT. Because larger estates tend to comprise a larger share of these types of asset, this means that effective rates tend to decline with increasing wealth.

There are several other significant exemptions from IHT. However, the effects of these are not accounted for in the Office of Tax Simplification analysis because the exempted assets are not routinely reported to HMRC. As such, they are not included in the denominator used to compute effective tax rates. The most important of these is lifetime gifts made more than seven years prior to death. Such gifts are likely to be heavily skewed towards the wealthiest donors, who have more liquid assets outside their house and pension (which are hardest to give away) and can afford to give during their lifetime without affecting their current standard of living.

There have been recommendations for major reform to taxation of inheritances and gifts by switching to a tax based on the amount received by the donee, rather than (as currently) on the amount given away by the deceased [20, 22, 23, 24]. Such a system is in place in some other jurisdictions, such as the capital acquisitions tax in Ireland. But even without such radical reform, inheritance tax could be made significantly fairer and more effective by broadening the tax base through removal of existing reliefs and by bringing all lifetime gifts (above a set amount) into tax. Advani, Hughson, and Tarrant estimate that removing the reliefs for agricultural and business property could raise around £3 billion, equivalent to a 60% increase in the total tax yield from IHT [25].

3.2 International mobility

The tax system is complicated enough even when considering just a single jurisdiction and a static population. However, for the super-rich, there are often also international dimensions to contend with – or take advantage of. These result from personal connections with multiple countries, assets located in more than one jurisdiction, or use of holding entities such as companies or trusts located in other jurisdictions. Such arrangements can lead to problems of double or non-taxation, depending on how different countries’ tax rules interact. A glance at the Sunday Times Rich List reveals that the UK’s top-10 wealthiest individuals all have some kind of international tie, whether through residence, nationality, or business interests. Even though this global footprint recedes as one moves lower down the list, it is obvious that the international mobility of individuals and assets is a first-order issue in relation to taxation of the super-rich.

Like most countries, the UK broadly taxes its residents on their worldwide income, gains, and (for inheritance tax purposes) assets. Subject to some important exceptions that I discuss later, individuals must therefore usually move themselves abroad – and not merely their assets – if they wish to reduce their UK tax. This basic point is often misunderstood by commentators and politicians who worry about the risk of ‘capital flight’ if top taxes are raised. Nevertheless, although things are not quite so simple as just moving assets offshore, there are a number of ways in which the super-rich can currently leverage foreign ties to reduce or eliminate their exposure to UK tax.

First, the non-dom regime perversely creates a powerful incentive for UK-resident non-doms to invest their assets anywhere except the UK. Although this problem has been mitigated to some extent by the introduction of Business Investment Relief in 2012 (which applies to investments in some qualifying UK companies), the relief is complex and consequently not widely used. To comprehensively eliminate the current disincentive to invest in the UK, the remittance basis of taxation should be abolished altogether (and if necessary, replaced with a different relief for new arrivals that does not privilege foreign investments).

Second, the location of assets can matter for enforcement, because despite recent improvements in the international exchange of information and harsher penalties, it is still more difficult to detect non-compliance involving offshore elements than in respect of assets held in the UK. Although there is a perception amongst UK tax professionals that non-compliance by the super-rich is very low, this is at odds with international evidence that uses data from tax leaks and amnesty schemes, which finds both that non-compliance increases at the very top of the distribution and that it is less often detected by standard audits [6].

Third, individuals who are non-resident in the UK are (broadly) exempt from UK tax, apart from on their UK assets. There is an emerging international literature on the effect of taxes on international migration [26], although currently no studies that focus on the UK specifically. One gap in this literature is that it tends to assume an ‘all or nothing’ model of migration in which individuals chose a single location to live and pay taxes. The reality is somewhat more complex than this, especially in relation to the super-rich, who may have substantial footprints in multiple countries.

UK tax residence depends on the number of days that an individual spends in the UK, whether their only home is in the UK, and (in some circumstances) the number of other ties that they have to the UK such as accommodation, work, and family. A relatively small reduction in an individual’s UK footprint may therefore be enough to acquire non-resident status, rather than needing them to emigrate entirely. Furthermore, where the individual is tax resident in both the UK and another country simultaneously, they may still be able to escape UK tax if they can show a closer connection with the other country. A recent example is Jonathan Oppenheimer, heir of the De Beers fortune, who was tax resident under the domestic rules of both the UK and South Africa but successfully resisted UK tax on the basis that he was a treaty resident in South Africa [27].

The top priority for reform in this area should be the abolition of domicile as a connecting factor. The most straightforward alternative would be a residence-based test, perhaps using a set number of years of residence to create a ‘tail’ for liability to inheritance tax, whether as a new arrival or after ceasing to be resident. The test for residence itself could be made stickier, for example, by requiring individuals to relinquish their home in the UK if seeking to acquire non-resident status. Furthermore, although the UK’s membership of the EU previously made it difficult (though not impossible) to implement an ‘exit tax’ for emigrants, there is now no legal impediment to introducing such a charge. For example, it would be possible for the UK to introduce a deemed disposal of assets for capital gains tax on the accrued gains of emigrants, as many other countries already do.

3.3 Trusts

The tax treatment of trusts is highly complex and remains something of an enigma to non-lawyers.¹ Most people (including economists) are used to thinking of ownership in binary terms: either you own an asset, or you don’t. If you own an asset, you get to choose what happens to that asset (control), and you get to make use of it and any income from it (benefit). In short, a trust allows you to split these two key features of ownership – benefit and control – such that they are vested in different people. The trustee controls the asset whilst the beneficiary benefits from it. There are various legitimate reasons for setting up a trust,² but the division of control and benefit creates major difficulties for tax policy.

It is often suggested that the tax treatment of trusts should be ‘neutral’ [28]. But neutral with respect to what? From the perspective of the person who puts assets into a trust (the ‘settlor’), this action is different from both making an outright gift of the asset (where benefit and control are given to someone else) or retaining the asset outright (where benefit and control are fully retained). Consequently, neither of these benchmarks is entirely apt as a basis for taxation. Trust assets effectively have two owners (the settlor and the beneficiary) but in different respects; what is more, the precise terms of control and benefit are almost infinitely variable. This is what makes taxing trusts so difficult.

The UK is not unique in facing these problems, which also arise for foundations and other similar legal structures in other jurisdictions. However, despite many years of complex reforms, a satisfactory approach has not been found. The current approach is piecemeal – there are different rules for different taxes for no apparent reason, and no answer to the conceptual question of how to deal with circumstances where ownership is split. Tied up with this is also the issue of connecting factors once again: For example, if a settlor or beneficiary is resident in the UK, does this mean that the entire trust assets should be liable to UK tax?

Assessing tax on the beneficiary might seem fairest from a welfare perspective, because this party (by definition) is the one who benefits from the trust assets. However, this approach runs into difficulties in relation to discretionary trusts, where it is often not possible to determine the identity of the beneficiaries prior to a distribution from the trust being made. Another option would be to treat trusts as separate taxable persons (like companies), but this in turn opens the door to reducing tax by fragmenting wealth across multiple trusts. Chamberlain proposes a hybrid approach that assesses tax on the trust, but with reference to the circumstances of the settlor and beneficiaries and the situs of the assets [29].

In the short term, one immediate reform should be to ensure that the residence of the trustees (who manage the assets) is entirely irrelevant for tax purposes. At present, the UK perversely imposes less stringent reporting requirements on trusts that are managed abroad, even when they have a UK resident settlor or beneficiary. This should be changed to ensure that HMRC gains a fuller picture of trusts that could incur a UK tax liability. However, this is a relatively modest and technical reform. More radical and comprehensive reform such as that proposed by Chamberlain [29] is possible: the barriers are more political than technical.