The 50p tax rate and inequality

17th March 2014

Posted on March 7, 2014 by David Eiser at The Economics of Constitutional Change

Alex Salmond has twice this week refused to commit to re-introducing the 50p rate of income tax on incomes over £150,000 – first following his New Statesman lecture in London, and then at First Minister’s Questions on Thursday.

In making the case for independence, the Scottish Government has frequently argued that having access to tax and benefit levers would enable it tackle inequality more effectively than successive Westminster governments have tended to do. Against this backdrop, the First Minister’s refusal to support the re-introduction of the 50p rate seems contradictory – but is it?[i]

In Scotland in 2010 (the latest year for which we have detailed HMRC statistics), only around 14,500 individuals paid income tax at the additional rate. These individuals represent just 0.5% of all Scottish income taxpayers (and less than 0.3% of Scotland’s population). Yet these 14,500 individuals accounted for 12% of total Scottish income tax receipts, implying an average tax bill of £85,000 per person (although the median tax bill for these individuals is £65,000).

These additional rate taxpayers are clearly a key part of the tax base of an independent Scotland. Yet this base is fragile; the taxable income of additional rate taxpayers tends to respond fairly fluidly to changes in the top rate of tax, which means that tax rises can lead to very marginal or even negative increases in tax receipts. The extent to which these individuals’ incomes are responsive to changes in the tax rate is difficult to estimate with certainty. Evidence produced by HMRC indicates that, once reasonable assumptions about the response of top incomes to the tax rate are factored in, varying the additional rate between 45 and 50p could lead to either a small rise or a small fall in total tax revenues at UK level; but in either case the effect would be small.

So reinstating the 50p additional rate seems unlikely to raise additional revenue for an independent Scottish government. But could the policy be justified on the basis of its effect on inequality?

An interesting perspective on this question has been raised in a paper by Emmanuel Saez and Thomas Picketty. Looking at 18 OECD countries between the 1970s and early 2000s, they observe a correlation between reductions in the top rate of tax and increases in top pre-tax income shares.

Saez and Picketty suggest two potential underlying reasons for this correlation, with each explanation lending itself to quite different policy responses. On the one hand, it may be that lower top tax rates encourage work effort and business creation among the most talented[ii]. If this were the case, lower top tax rates would lead to increased economic activity by the rich and hence increased economic growth (and potentially therefore incomes and employment). On the other hand, lower top tax rates might increase the incentives for high-paid individuals to bargain for higher pay. This might be particularly likely to happen in complex organisations, where the performance of individual managers is difficult to measure, and where top earners are able to partly set their own pay by bargaining harder or influencing compensation committees. If this were the case, cuts in top tax rates might lead to an increase in top income shares, but these increases in top incomes would come at the expense of incomes among the remaining 99% of the population, unambiguously increasing inequality.

To test which of these two scenarios best explains the observed correlation between top tax rates and top income shares, Saez and Picketty compare economic growth rates and top tax rates across OECD countries. If the first explanation is correct, then there would be a relationship between top tax rate cuts and economic growth. In fact, Saez and Picketty find no evidence that lower top tax rates lead to higher economic growth. They conclude that ‘a substantial fraction of the response of pre-tax top incomes to top tax rates may be due to increased rent-seeking at the top, rather than increased productive effort’.

This finding is important because of what it implies for the role of tax policy in achieving a balance between encouraging economic growth on the one hand, and mitigating inequality increases on the other. If the response of top incomes to tax rates were due solely to a reduction in productive effort among top earners, then Saez and Picketty calculate that top tax rates could be set at around 55% without deleterious effects on growth; this is not much higher than the current effective marginal tax rate on high earners in the UK (considering the combined effect of income tax and national insurance contributions). However, if the response of top income to tax rate was due solely to a rent-seeking effect, then top tax rates could be set much higher, perhaps as high as 80%.

A caveat here is of course the fact Scotland is closely integrated with the rest of the UK. This means that raising the top rate of income tax significantly above the rUK rate might have more significant negative effects on Scottish tax revenues than would be the case if rates were raised simultaneously in both Scotland and rUK. Equally however, it would also mean that cutting top tax rates in Scotland below the UK rate could increase Scottish revenues. There is an argument that, as incomes are lower in Scotland than rUK, the ‘optimal’ top tax rate (i.e. the one that maximises revenue) will also be lower in Scotland than rUK[iii]. The degree of interaction between the Scottish and rUK economies means that Alex Salmond’s refusal to commit to reintroducing the 50p rate can be seen as rational, even if it does seem inconsistent with the SNPs social democratic positioning.

Appetite for top tax rate increases will depend to a large extent on beliefs about the extent to which the salaries of top earners reflect productivity or rent seeking. Lack of support for a 50p top tax rate is not necessarily contradictory of a desire to reduce inequality, if one believes that these salaries are ‘fair’ in the sense that they reflect the skills, productivity and entrepreurial flair of those who earn them. And if top salaries do reflect an element of rent-seeking, then it might be possible to address this through other policy interventions – reforms to salary setting arrangements for example – rather than through tax policy. However, if the Scottish Government is as serious as it says it is about reducing income inequality, then committing to reintroduce the 50p rate would demonstrate that its criticism of Westminster’s efforts to address inequality is more than just rhetoric.


[i] The 50p rate has been in place since 2010, but was reduced to 45p by the coalition government in last year’s Budget. Labour has pledged to restore the 50p rate if they are elected at the next election, with Ed Balls arguing that ‘those with the broadest shoulders should bear a fairer share of the burden’.

[ii] Economic theory suggests that an income tax cut could also lead to a fall in pre-tax incomes, as people might decide to work less and enjoy the same level of net income. However, Saez and Picketty find evidence of the reverse: reductions in top tax rates lead to increases in pre-tax incomes.

[iii] Assuming that the elasticity of incomes with respect to the tax rate is the same in Scotland and rUK, then given the existing additional rate threshold and median incomes of additional rate payers, we calculate that the optimal level of additional rate taxation is about 2% lower in Scotland than rUK.

Article originally published at http://esrcscotecon.com/2014/03/07/the-50p-tax-rate-and-inequality/