Written evidence submitted by Helen Miller, Institute for Fiscal Studies

The OTS and tax simplification

I was on the advisory panel for the 2021 Review of the Office of Tax Simplification.

Broadly, my personal view is that the OTS, while far from perfect, was doing useful work and could have been made more impactful. On balance therefore, I would not have recommended scrapping it.

I think one of the key challenges for the OTS was that, more often than not, tax simplification cannot be separated from tax policy choices. The tax system is clearly far too complex. And in some cases, there are fairly narrow (but important) technical issues that can be addressed in isolation – i.e. where there is no disagreement about what a tax policy is trying to achieve but there are ways that it could be changed to make it simpler.  But in many cases complexity arises as a direct result of policy choices. For example, VAT is complex because policy makers have chosen to put in place many zero-rates and exemptions. Much complexity arises in how we tax incomes because policymakers have chosen to set very different rates of tax for different income sources (e.g. labour income vs capital gains). In some cases, better policy choices would also lead to a simpler tax system. In other cases, there are trade-offs between simplicity and other policy goals, and these need to be weighed up by policy makers. Therefore, I think complexity needs to considered at the same time as policy and should be (and I’m sure is) considered within the Treasury. But the OTS (even without a remit that includes making policy recommendations) could also address simplification related to policy choices by making conditional statements of the type: If policy wants to achieve X, a better way to do it would be Y. (This was reflected in the first recommendation of the review).

One of the benefits of the OTS was that it brought together tax experts and practioners to discuss an area of tax in detail. I think more could have be done to make sure that OTS was always engaging with the right sets of people, but that kind of engagement is important to get a full understanding of how taxes are working (or not working). It is also important to consider the current stock of policies and the complexity they create and not only complexity as it arises with new policies. In principle, the OTS was free to look at areas of tax that were most problematic, even if not a current political priority.

Another benefit of the OTS was that it put out public reports. Again, I think more could have been done to make the OTS outputs more accessible and impactful. But, I think there is benefit in having published documents (as opposed to internal civil service briefings) that set out issues of complexity. This facilitates discussion of issues within the wider tax community.



Which thresholds and cliff edges should be simplified? Three top picks from Helen Miller


  1. End of childcare entitlements at £100,000.

Discussion of this issue can be seen here: . This report includes:

Free childcare entitlements (or 15 or 30 hours) and access to the “tax free childcare” subsidy scheme, are limited to families where no parent earns £100,000 or more. Above that level, the entire value of that support is removed - creating a “cliff-edge” effect whereby families can easily be worse off overall even after a substantial pay rise.

By extending free entitlements to 1 and 2 year-olds, the government will significantly expand the scope of those cliff-edges, and their size - since many more families will now find free entitlements at stake for two or more pre-school children at the same time.

The distortions that this can create are among the most severe you will ever see within a tax and benefit system.  A  parent with two children under 3 whose childcare provider charges England’s average hourly rate for 40 hours per week would, after these reforms, find that their disposable income (i.e. earnings net of tax and childcare outgoings) falls by £14,500 if their pre-tax pay crosses £100,000. Disposable income would not recover its previous level until pre-tax pay reached £134,500, meaning a parent earning £130,000 would be worse off than one earning £99,000.


I think this cliff edge should be addressed given it is so large for those affected. If the childcare entitlements were made universal, the revenue effects could be offset with a higher top rate of income tax (for example, and if desired). 


  1. High Income Child Benefit Charge

Discussion of this issue (and a figure showing marginal tax rates) can be seen here:

This report includes:

The high-income child benefit tax charge, introduced in January 2013, means that child benefit is tapered away for families whose highest-income partner has an annual income above £50,000. For every £100 of that individual’s income above £50,000, child benefit entitlement falls by 1%. This means that if the higher-income partner’s income is below £60,000, then the family is eligible for a partial payment, whereas if income is at or above this amount, the family is not eligible for child benefit. These thresholds are frozen in nominal terms, meaning that every year more families lose eligibility for some or all of their child benefit. At the time the policy was announced, 13% of families with children (1 million) lost some or all of their child benefit. As a result of the freeze, we calculate the figure is now 26% (2 million families), and this is set to rise to 31% (2.5 million) in 2025–26 if the freeze continues.

One effect of tapering away child benefit is an oddly shaped marginal tax rate schedule, which jumps up in the range where child benefit is withdrawn, then jumps back down again. An employee with two children faces a marginal tax rate (including employee NICs) of 32% if they earn just under £50,000. This rises to 60.9% at a little over £50,000 when they begin to pay higher-rate tax and start to lose their child benefit, before falling to 42% when their earnings reach £60,000. For those with more children, the jump is even more stark, as they have more child benefit to lose.

It is difficult to think of a reason why the marginal tax rate schedule should be hump-shaped in this way, and even harder to imagine why the size of the hump should by default get higher and move down the real income distribution every year.


If child benefit is going to continue to be tapered away (and the benefit is continuing to rise in normal terms), the thresholds should be uprated. As an alternative to setting (and ideally uprating) the start and end point for tapering, policy could set the starting point for tapering and the taper rate, where the latter could be common across families (rather than increasing with the number of children). Of course, another alternative is to move back to a universal benefit.


  1. VAT registration threshold

Many people have noted that there is significant ‘bunching’ of businesses below the VAT threshold. This was noted, for example, in an OTS report on VAT (here:

The VAT threshold is a cliff edge. Usually, cliff edges should be avoided. However, in the case of VAT, removing the threshold by either (i) bringing all businesses into VAT or (ii) operating some kind of taper would add significant complexity.

More importantly, the bunching highlights not simply that the threshold is leading some businesses to change their behaviour, but that the VAT system in general will be leading firms to reduce their sales and / or face a large compliance burden. Said differently, it seems likely that large compliance costs are deterring some firms from wanting to grow and therefore have to register for VAT. But the compliance costs do not end once a firm is registered for VAT – they are there for all registered businesses.

Much of the compliance cost related to VAT stems from the large number of zero-rates and exemptions present in the UK system. (The amount the UK raises through VAT is comparable to most other developed economies (other than the US), despite the fact that the UK has a much narrower VAT base (because it applies zero rates more widely) than most other countries.) Therefore, I think a focus of policy should be reducing VAT compliance costs. In particular, I would recommend moving to a largely uniform VAT (i.e. removing most zero rates and exemptions). 


May 2023