Should income tax and National Insurance be merged?

The UK has two taxes on income. Though different in origin, income tax and National Insurance contributions (NICs) are now very similar. The link between NICs and benefit entitlements is now vanishingly weak. Maintaining two separate taxes yields little benefit, but makes their combined effects less transparent and imposes extra administrative burdens. 

The case for integration on grounds of administrative simplicity has been somewhat weakened (though not eliminated) by the greater use of automated systems to handle payroll and taxes. But the case for integration on grounds of transparency remains very strong.

Current rate schedule lacks transparency

The tax schedule that people actually face on their earnings reflects the combination of income tax and NICs, and is shown below. 

Transparency and effective decision-making require that it be this combined schedule that is described and debated. Yet public debate is often restricted to headline income tax rates, which (outside Scotland) rise from 20% to 40% and then 45% at higher incomes. It is rare in popular discourse to hear discussion of the 12%, 32%, 42% and 47% rates that (in 2021–22) apply once employee NICs are taken into account, let alone consideration of the effect of employer NICs (which take the highest combined rate to 53.4%, as shown in the chart) or the different rate schedule that applies to the self-employed. Because attention is usually restricted to headline income tax rates, ignoring NIC rates (which fall from 12% to 2% for employees), the progressivity of the system is easily overstated.

Regrettably, politicians may sometimes find the lack of transparency useful. One example of how the separation of the two taxes can lead to confusion (to put it generously) is the Labour Party’s 2001 general election manifesto, which pledged not to increase income tax rates. In its first Budget after the election, Labour increased NIC rates. If Labour’s intention was only to rule out tax increases on forms of income that were subject to income tax and not NICs (such as investment and pensions income), that was certainly not made clear before the election. Whether the shortcoming was duplicity, incoherence or merely poor communication, we should strive to make such episodes unrepeatable. 

Separate taxes hinder simplification of rates 

The separation of income tax and NICs is an obstacle to making the combined rate schedule sensible and straightforward. For example, a simple combined rate schedule would have the thresholds for income tax and NICs aligned. Over many years, we have seen governments sometimes align the thresholds, but then break the alignment again when it becomes politically expedient to do so. (At present, the NICs upper earnings limit (UEL) and upper profits limit (UPL) are aligned with the income tax higher-rate threshold, except in Scotland; but the thresholds at which income tax, employee NICs and employer NICs become payable are all different.) The ease with which alignment has been abandoned on more than one occasion demonstrates that it is a poor substitute for genuine integration. And since Scotland has devolved powers over income tax but not NICs, it cannot exercise its right to vary the higher-rate threshold (as it has) while also ensuring it is aligned with the UEL and UPL.

Aligning thresholds within the current system is a poor substitute for genuine integration in a different sense too. Even where the thresholds do appear to be aligned, the fact that income tax and NICs are assessed over different periods (annual for income tax, pay period for employee and employer NICs) and on different measures of income means that they can effectively diverge depending on individual circumstances. Even outside Scotland, therefore, it is quite possible for someone to be above the income tax higher-rate threshold but below the UEL, or vice versa. Achieving coherence between the two taxes requires addressing the differences between them.

Dealing with perceived barriers to integration

In so far as income tax and NICs are fundamentally the same, transparency and administrative simplicity would clearly be well served by merging them. But there are still differences between the two taxes that would need to be dealt with.

In most cases, the differences actually make the case for integration even stronger. It is patently absurd, for example, to have one tax (NICs) assessed on earnings in each individual week or month in which an employee is paid and another (income tax) assessed on income over the whole year. Integration would force us to rationalise the system by moving to a single (preferably annual) assessment period. 

Some forms of income are subject to income tax but not NICs, principally:

• income from savings, investments and property;
• certain state benefits;
• private and state pension income;
• the earnings of individuals over state pension age (which are subject to employer, but not employee or self-employed, NICs).

(Similarly, some outgoings, such as employee pension contributions and certain expenses, are deductible for income tax but not for NICs.) It is sometimes suggested that these differences would be a barrier to integration. But on the contrary, they strengthen the case for integration. 

These categories of income are in effect taxed at an intermediate rate, above zero but below the full (combined) rate charged on ordinary earnings. If income tax and NICs were merged, the combined tax could, if desired, be charged at a lower rate on the categories of income listed above, corresponding to the de facto intermediate rate that now applies. Different tax rates for different income sources are perfectly feasible: we already apply reduced income tax rates to dividends and reduced NIC rates to self-employment income, and there have been many more such examples in the past.  

That would be an improvement. At present, the level of the intermediate rate is an accidental by-product of decisions made for other reasons, driven largely by misperceptions that make NICs politically easier to increase than income tax. And it is not only an unintended feature of the system; it is also a particularly opaque one. If some forms of income are to be taxed at an intermediate rate, this treatment should be made explicit. A single integrated tax would achieve this, and thereby prompt open debate about whether each form of income should be taxed in full, not at all, or at an intermediate rate. The level and coverage of the intermediate rate are little discussed at present because they are largely invisible.

The same applies to the link between National Insurance contributions and benefits. That link is now extremely weak. But if the government wanted to preserve the link that does remain, there seem to be no serious obstacles to devising a set of contribution conditions based on a merged tax that closely mimic the current rules. If a merged tax applied to savings income as well as earnings, for example, it would still be possible to base contribution records on the tax paid on earnings alone. And once again, integration of income tax and NICs would provide a good opportunity to rethink whether the current rules are what we would ideally like to achieve, and to reform them if not (including creating stronger links between contributions and benefits or removing the links altogether).

The biggest obstacle to integration is the fact that NICs, unlike income tax, is levied on employers as well as employees. One option would be to levy a merged tax just on the individual: in other words, to shift (perhaps gradually) the tax on the employer to the employee. Economic theory suggests that, in the long run, earnings would rise to offset this change in the formal liability for the tax so that the shift would not change individuals’ after-tax incomes or the cost to firms of employing people. But while that might be true in the long run, the challenges in transition would be formidable and the political difficulties are evident. An alternative would be to keep a separate (preferably simplified) employer tax while integrating employee NICs with income tax; but that would forgo some of the potential gains from simplification (in terms of both administration and transparency). This issue should be considered in conjunction with the question of whether to continue levying a tax just on income from employment, particularly given the disparity that creates between the tax burdens on employment and on working through one’s own business (we discuss that in detail here). If we wish to level out the tax treatment of different legal forms of work and business, some fundamental reform of employer NICs is needed anyway: either to phase it out, or to apply it (or an equivalent tax) to other forms of income as well.

Devolution 

A newer consideration in integration of income tax and NICs is devolution. Income tax rates and thresholds are mostly (though not entirely) devolved to Scotland and (to a lesser extent) Wales; NICs are not. There is no obvious obstacle to devolving rate-setting powers over a merged tax; if desired, there could be limits on the devolved governments’ powers (as there are now and have been in the past). There is a technical question as to how the remaining links between contributions and benefits, and the formal mechanism of the National Insurance Fund, would be dealt with; but there are several viable solutions to this, even if we assume that the government would not want to devolve powers over contributory benefits or the rules relating contributions to benefits. These would involve legislative and administrative adjustments, but no insurmountable obstacles.

Stalling progress 

Calls for a merger of income tax and NICs have been widespread for many years, but successive governments have rejected them. What we have seen instead is gradual moves towards greater alignment between the two taxes, making them ever more similar. But progress towards further alignment has slowed. In 2016, the Office of Tax Simplification produced two reports proposing (relatively modest) reforms to bring NICs closer to income tax; but the government rejected the main suggestions (for involving too much upheaval) and U-turned on its initial acceptance of two others, ultimately making only very minor technical adjustments.