Needing to raise a substantial amount of revenue, the Treasury took a broad approach, choosing NICs over VAT and income tax. Increasing NICs is less unpopular with voters than increasing other taxes and public misunderstanding concerning the nature of the tax additionally allows some room for political manoeuvre. The end result, though, is three taxes on income – income tax, NICs and the HSC levy – with three different bases and a UK tax system that is both unnecessarily complex and further tilted to favour the self-employed.

All policies involve trade-offs. This is never more true than when it comes to tax policy. The perfect tax policy will raise lots of revenue, not create any economic distortions, be popular with MPs and the public, be administratively easy to pay and collect and not add to the complexity of the tax system.

No tax policy will ever meet these requirements. There is bound to be a tension between some of these objectives. In the unlikely event that there ever was a tax policy that satisfactorily met these objectives, a previous government would have already implemented the policy and subsequent governments would have sought to raise more and more revenue that we would have reached the point that any further attempt to exploit the policy would run the risk of losing revenue or provoking unpopularity.

I make these points to set out the context in which ministers operate. Raising revenue is hard and there is no perfect way to do so. It involves compromises and trade-offs that usually results in decisions being tactical rather than strategic. The consequence is a tax system that is ill-designed and unnecessarily complex.

The announcement of the new health and social care levy (HSC levy) is a very good case study in how these trade-offs come together and how, as a consequence, our tax system becomes less coherent.

The politics of raising serious revenue

This is not the place to consider the merits of additional expenditure for the NHS or social care or, to the extent that additional resources are to be spent on social care, what is the best way for this to be done.

There are arguments that can be made for protecting the assets of recipients of social care through some kind of insurance model but, for one reason or another, the government has decided not to pursue such an approach. Again, the merits of this decision are outside the scope of this article.

The task given to Treasury ministers and officials was to raise a substantial amount of revenue: £12bn a year. The scale of this will have determined the broad approach.

Attempting to raise this amount of money from a series of relatively small tax increases would be a very high-risk strategy. Two lessons from the infamous "omnishambles" Budget of 2012 – in which a long list of revenue raising measures were set out – are of relevance here (I was the tax minister at the time, so these lessons are seared into my memory). The first is that the greater the number of tax increases, the greater the chance that MPs and the media will identify something that is objectionable. The second lesson was that it is easier for government MPs to oppose measures that raise a small amount of revenue than a large amount of revenue. Attempting to block a measure that raises £100m is unlikely to be seen as dissenting from the broader thrust of the government’s economic and fiscal policy. Dissenting on a £12bn tax policy is a much more serious matter.

Some will argue in favour of a big tax focused on the wealthy such as capital gains tax, inheritance tax or even a wealth tax. Without getting into the details of the merits of these proposals, a combination of lack of revenue, political unpopularity and administrative challenges would have meant that the Treasury would have spent little time looking at such options. And the fundamental point is that if you want to raise a lot of revenue you have to tax a lot of people.

Our tax system is already heavily dependent upon the wealthiest which is not to say that governments in future will not attempt to raise revenue from them. But such measures are more than likely to be political – to gain wider public consent for higher taxes on a wider section of society – than as a means of raising serious sums of additional revenue.

The big three – and why it had to be NICs

This leaves the three big taxes that make up two-thirds of all government revenue – VAT, income tax and national insurance contributions.

VAT is to where Conservative chancellors of the exchequer traditionally turn when in need of a serious boost in revenue. Sir Geoffrey Howe in 1979, Norman Lamont in 1991 and George Osborne in 2010 announced increases to the main rates of VAT on the basis that consumption taxes were less likely to damage an enterprise economy than taxes on income. There appears to have been little consideration of doing that on this occasion, probably because of concerns about its regressive nature (it is not as regressive as is often thought, but it is certainly not as progressive as the alternatives). The Conservative electoral base depends on lower income voters more than it once did and a VAT increase might land badly with such voters.

Income tax is the largest tax. It is also the highest profile one and an increase in income tax rates would be totemic (the basic rate has not increased since 1975 when it went up from 33% to 35%).

The third big tax is national insurance, and it was to this tax that the government turned to raise revenue. The reason for that is very simple: increasing NICs is less unpopular with the public than increasing other taxes.

National insurance contributions is a tax that provokes sharply different reactions between the informed and the uninformed. For many of the general public who have better things to do than focus on tax policy, it is not even a tax but, as the name suggests, an insurance contribution. You pay money in and, as a consequence, your entitlement to payments out in particular circumstances increases. People might not think that there is a specific pot of money with their name on it to which they are entitled but there is a sense that paying NICs involves creating contractual entitlements from the state.

To the extent that this was once true, it really is not now. Successive governments have concluded that benefits should either be universal or paid to those with the lowest means not those who have made the greater contributions. Our benefits system has become less contributory but not every member of the public has noticed.

Related to the contributory principle is the issue of hypothecation. There are some items of government expenditure that the public are more enthusiastic about supporting than others, usually ones which they expect to benefit from themselves in due course – like healthcare and pensions. Hypothecated expenditure on warships or prison maintenance might prove less popular.

The point remains that the public is more content to contribute to the exchequer if it thinks it understands and likes what the money is spent on. We saw this in 2002 when Gordon Brown increased NICs to "fund the NHS".

Linking revenue with expenditure appeals to the electorate but is usually something of a confidence trick. If expenditure was really dependent on a specific revenue stream, a fall in tax receipts would result in cuts in spending. In a recession, for example, tax receipts can fall substantially whereas it can be difficult and undesirable for spending to follow suit.

The Treasury is also institutionally sceptical of linking expenditure with specific revenue streams. Taxes should be raised in the best way possible; expenditure should be spent in the best way possible; linking spending with revenues merely makes those tasks harder. The Treasury is also aware that, in reality, hypothecation is a one way bet: expenditure would rise when revenues are buoyant, but when revenues are depressed expenditure will be paid for by other taxes or borrowing.

In addition to the contributory and hypothecated nature of NICs, employers’ NICs popularity benefits from a widespread failure to understand tax incidence. The entity that has the legal liability to pay a tax is not necessarily the entity that bears the economic cost. This is always the case with taxes on businesses because all taxes are paid by people in the end – whether they be employees, shareholders, customers or suppliers. In the case of employers’ NICs, the economic consensus is that employees pick up the bill in that their pay will be lower than would otherwise be the case (at least over time).

This might be the view of economists, but it is not clear that this is appreciated by voters. When explaining why he had increased NICs rather than income tax, the prime minister argued that "income tax is not paid by businesses, so the whole burden would fall on individuals, roughly doubling the amount that a basic rate taxpayer could expect to pay". This provoked some gnashing of teeth from those who understand tax incidence, but it is likely to have had many viewers (including quite a few MPs) nodding along.

Broader than NICs but narrower than income tax If the popularity of NICs (or to put it more accurately, the relative lack of unpopularity) is based on public misunderstandings of who pays for it and where the money goes, does this make it a bad tax?

Much of the criticism of the NICs increase is that it is regressive. To be fair to the government, this is not the case. A person earning £20,000 a year will pay 0.65% of that in additional NICs (or HSC levy) whereas for a person earning £100,000 that figure will be 1.13%.

It is, however, the case that NICs is less progressive than income tax, in part because its base is more limited. It is a tax on earnings but, unlike income tax, is not levied on dividends, rental income or pensions. It is also not paid by those over the state pension age.

Many of these points were made when the health and social care package was briefed to the Daily Telegraph in advance of last week’s announcement. As a consequence, it appears that the additional NICs charge/HSC levy will apply to dividend income and earnings for those above the state pension age, making the levy broader based than NICs but narrower based than income tax.

What has not been addressed is a longstanding concern of the Treasury that our tax system treats self-employment much more generously than it treats employment. Given the true incidence of employers’ NICs, employees lose out from the 1.25% increase in employees’ and the 1.25% on employers’ NICs, whereas the self-employed face an increase of 1.25% on class 4 contributions only. Our tax system is further tilted to favour the self-employed (much, I suspect, to the frustration of Treasury officials).

Time for closer alignment of income tax and NICs?

There was a time when the government looked at more closely integrating income tax and NICs. The OTS undertook a review of the matter and as tax minister I was supportive of going as far as we could. There were significant challenges although it was possible to do this without making the bases of the two taxes identical (however desirable that might be). One group that would have lost out would have been those with multiple jobs who benefited from a separate NICs allowance for each job. There is no moral justification for a person with multiple jobs paying a lower amount of tax than a person with the same income from one job, but moving from the existing position would have been politically challenging.

In the end, we did not make progress, but close alignment between our two taxes on income would have been desirable from the perspective of tax simplification. Now, however, we have three taxes on income – income tax, NICs and the HSC levy – with three different bases.

One can see how it happened. The need to raise revenue; clear polling evidence showing what the public would tolerate; and some last minute criticism that resulted in an extension of the base to dividends. The money gets raised and the damage to the government (given that revenue raising is generally unpopular) is relatively limited. And the coherence and simplicity of our tax system is further diminished. 

This article was first published by Tax Journal.