Inequality and social justice are playing a central role in the debate around Scottish independence. In “Scotland’s Economy: the Case for Independence”, the Scottish Government asserts that the UK Government has done too little to address rising inequality in the UK, and argues that high levels of inequality are likely to be having adverse consequences on Scottish economic performance and individual well-being. But, as argued previously by the Scottish Government’s Fiscal Commission Working Group, the report stresses the limitations faced by the Scottish Government in addressing inequality within the context of the existing constitutional settlement. Independence, it is argued, would give the Scottish Government access to the levers (notably welfare spending and personal taxation) necessary to achieve lower inequality.
But which of these levers are likely to be most effective in reducing inequality, and by how much? In this blog, we examine how specific policy changes that an independent Scotland could make – to welfare spending, personal taxation, and council tax – would influence the level of inequality in Scotland. The findings reported are based on a paper to be presented at the International Conference on the Economics of Constitutional Change on 20th September.
The results are derived from a model of household level income and benefits developed at the University of Stirling. The model takes data from the DWP’s ‘Family Resources Survey’, an annual survey of over 4,000 households in Scotland, and uses this to calculate the level of personal and indirect taxation paid by each household, as well as the amount of each major welfare benefit received. The model can be used to examine the impact of changes to tax rates and thresholds, or benefit eligibility and levels, on both aggregate government revenues and spending, and the distribution of income across households.
Marginal impacts of policy levers
First, we examine the affect of marginal (i.e. very small) changes in specific tax and benefit levers on overall income inequality. We measure income at the household level, after all benefits received and taxes paid, and after housing costs. Household income is ‘equivalised’, to take account of the size and composition of households (equivalisation takes account of the fact that, to enjoy a comparable standard of living, a household with two children will need a higher income than a household with no children for example). The measure of inequality we use is the GINI coefficient, a standard measure of inequality which measures how evenly income is distributed across all households in Scotland. This choice implicitly makes a judgement about how much we value changes in different parts of the income distribution relative to one another.
Figure 1 shows the affect of various tax and spend levers on net household income inequality. Each of these levers is expressed in terms of a £1m fiscal tightening. In other words, the effect represents what would happen to inequality if the Scottish Government increased its revenue from each tax, or decreased spending on each benefit, by £1m. We refer to this as a ‘marginal effect’ and would argue that this is too small to have a significant impact on how individuals respond to changes in these levers.
Figure 1: Marginal impact of various tax & spend levers on Gini Coefficient
The first two levers in Figure 1 relate to council tax, which is already wholly devolved to the Scottish Government. The marginal effect of raising revenue by increasing council tax rates on inequality is that inequality would slightly rise. This is interesting since a tax rise typically reduces inequality: the fact that raising Council Taxes actually raises inequality can provide an explanation for the popularity of the Council Tax freeze and the commitment that the SNP have shown to their policy on this in the face of very tight financial settlements. In contrast, an inequality reducing revaluation of the council tax bands (so that lower bands are reduced and higher bands are increased) could achieve a relatively large reduction in inequality whilst raising £1m in additional revenue. Why has this not been done? Jack McConnell ruled out a Council Tax revaluation at the time of the Burt Report on local taxation in Scotland in 2005. Perhaps losers would shout much more loudly than winners (and the fact that Council Tax liabilities are not that highly correlated with household income means that some poor losers will be found) and this means that politically a revaluation is difficult.
Figure 2 shows the percentage changes to average household net income driven by a Council Tax rise or a Council Tax revaluation that raises £1m overall, split by income decile. As can be seen, a Council Tax rise lowers household income across the piece, but proportionally more for middle income households: lower income households are shielded from council tax rises by council tax benefit; and higher income households pay only a very low percentage of their income in Council Tax. This gives rise to the inequality widening effects of a Council Tax rise: the gap between the middle and the top grows. On the other hand, a Council Tax revaluation hits the top of the income distribution and benefits the middle. But there exist a small number of low income households with exposure to the top Council Tax bands: the asset rich but income poor households that are living off their savings. The rise in Council Tax on these expensive properties under a Council Tax revaluation drives a very large negative effect on the average income of the lowest income households.
Figure 2: Household income % change by decile, under CT Rise & Revaluation
The middle four levers in Figure 1 relate to income tax. The first of these shows the effect on inequality of raising £1m in revenue by increasing each of the three income tax rates simultaneously. Varying tax rates simultaneously in this way is equivalent to the Scottish Rate of Income Tax (SRIT) that will become available to the Scottish Government as a result of the Scotland Act 2012. Figure 1 shows that increasing the SRIT would reduce inequality, although the effect, for a £1m rise in total revenue, is relatively small.
The remaining three income tax levers in Figure 1 are not available to the Scottish Government through the Scotland Act, but would become available following independence (or possibly through future constitutional change short of independence). Raising the upper rate of tax only would lead to a greater reduction in inequality, for a £1m rise in revenue, than an equivalent rise in the SRIT. Reducing the tax free allowance has virtually no effect on inequality, largely because such a move would benefit all taxpayers to a similar extent. The impact on inequality of raising the upper rate of tax is similar to the impact of lowering the upper rate tax threshold (i.e. the income at which individuals become liable for the upper rate of tax). Raising the tax free allowance was a LibDem manifesto commitment and is current UK Government policy. These results show that, if the objective is to reduce the overall tax burden, then raising the tax free allowance is the way to do this that has the smallest impact on inequality – almost all other tax reductions would raise inequality. Raising the tax free allowance is often mooted as a mechanism for benefiting those on low-incomes, but this is not the same as reducing inequality.
The last four levers in Figure 1 relate to welfare spending. These show the effect on inequality of a £1m reduction in spending on the state pension, tax credits, JSA/IS (out of work and very low income benefits) and IB/ESA (disability benefits). In each case, the reduction in benefit spending increases inequality, but the marginal effect is much higher for JSA/IS than it is for state pensions and tax credits, given that recipients of state pensions and tax credits are distributed more evenly throughout the income distribution.
The results in Figure 1 show that different policy levers can have very different effects on inequality at the margin (i.e. for a £1m increase in tax revenue or reduction in spending levels). But what can we say about the effects of specific policy scenarios on inequality? Can we use the information in this chart to construct policy scenarios that will reduce inequality in Scotland? In constructing such scenarios, as well as the marginal impact on inequality, we need to also consider the incentive effects of the change, and the possible size of potential changes. For example, increasing JSA & IS may have adverse effects on work incentives; and a revenue raising Council Tax revaluation, despite having strong effects on inequality per £1m of tax raised, may only have the potential to raise a small amount of tax.
We now illustrate how inequality would change under four possible tax and spend scenarios. Two of these scenarios would be possible for the Scottish Government under its existing powers and those made available through the Scotland Act 2012; we label these the ‘status quo’ scenarios. The remaining two scenarios relate to broader changes to the tax and welfare system that would only become available through independence (or significant devolution of the tax and welfare system to the Scottish Parliament); we label these the ‘independence’ scenarios.
For comparability, all scenarios represent a ‘no change’ on the overall balance of government revenues and spending in the Scottish public finances – broadly defined to include non-devolved areas under the status-quo – in the 2010-11 data.
Specifically, the scenarios we consider are:
- Status Quo A: Council tax abolition funded by a rise in the SRIT by 4.4%, (equivalent to a basic rate of income tax of 24.4%). The rationale for this scenario is that Figure 1 suggests raising revenue through the SRIT will reduce inequality whilst raising revenue through the Council Tax increases inequality. It is also a proxy for the Local Income Tax, which was SNP (and possibly LibDem) policy.
- Status Quo B: A radical reform of local taxation through reducing council tax bands A-C by 75%, 50%, and 25% respectively, maintain band D at its current level, and increasing bands E-H by 50%, 100%, 200%, and 300% respectively. This is revenue raising overall and this extra revenue is given back to taxpayers via a reduction in the SRIT (of 0.8%; this is equivalent to a basic rate of income tax of 19.2%). The rationale for this scenario is that theories of fiscal federalism and tax competition suggest that sub-national governments like the Scottish Government should seek to fund themselves using taxes on fixed factors like land and buildings, whilst reducing taxes on mobile factors like labour. This scenario may be expected to have supply side benefits, and radical revaluations make it inequality reducing overall, though not as much as scenario A.
- Independence A: Radical local tax reform as per Status Quo scenario B. Additionally, very low paid benefit rates (JSA & IS) are increased by 25%, whilst the state pension and tax credit rates are increased by 10%. This is paid for, in equal measure, by lowering the upper rate income tax threshold by £6,900 to £36,975, and raising the upper rate tax from 40% to 46%. The rationale for this scenario is to use Figure 1 to its full extent: Council Tax revaluation is inequality reducing, raising upper rate tax and lowering the upper rate tax thresholds are both inequality reducing, and increase benefit spending is inequality reducing. However, this scenario may be expected to have negative supply side effects through reductions in incentives: JSA and IS are unavoidably financially means-tested benefits.
- Independence B: Complete overhaul of the tax and benefits system in favour of a universal basic income of £7,800p.a. (£150 per week). This is funded by a flat tax of 56.75%, NB the top marginal rate of tax in all the other scenarios was 51% (via a 50% Additional rate of income tax since the calculation year is 2010-11, plus a top rate of National Insurance of 1%). In this scenario the top marginal rate of tax is the flat tax rate of 56.75% since Employees National insurance is abolished. Very low paid benefits, tax credits, and the basic state pension are all abolished. The radical local tax reform as per Status quo B is implemented. The rationale for this scenario is taking the universal versus means-tested tax & benefit system debate to one extreme. A basic income avoids poverty trap effective marginal taxes on the low paid as their benefits are withdrawn as they move into work. As such, this scenario may have supply side benefits – though this can be disputed. This analysis shows that such a universal benefit can be funded via the removal of all other benefits (other than disability), and through the removal of all tax thresholds and lower rates, at a tax rate only slightly higher than the top rate of tax in the base scenario.
In the following analysis, we present the results of each scenario relative to the 2010/11 fiscal position. Table 1 shows the impact of each scenario on several headline measures of inequality: the GINI coefficient, the 90-10 ratio (ratio of the 90th to the 10th income percentile), the 90-50 ratio and the 50-10 ratio. As previously, the income measure is the net equivalised income after taxes and benefits. Figure 3 shows the distributional effect of these scenarios, by examining the change in average weekly household income for each quintile of the income distribution.
Figure 3: Distributional effect of scenarios on average household incomes
- Status Quo A (abolition of council tax funded by an increase of the basic, upper and additional rates of income tax) has a small effect on reducing inequality, as measured by the GINI. As shown in Figure 3, this results from the reduction in average income for those in the top 20% of the income distribution, combined with small increases in income for those in the lower four quintiles.
- Status Quo B (revaluation of council tax bands and reduction of the basic, upper and additional rates of income tax) results in a smaller reduction in inequality as measured by the GINI, and by the 90-10 and 90-50 ratios. It actually sees a very slight increase in the 50-10 ratio though. This is because the widening Council Tax bands and reduced income tax helps the middle of the income distribution by more than it helps the bottom of the income distribution. This scenario does have an impact, but the inequality reducing effects of the council tax revaluation are largely cancelled out by the reduction in the SRIT.
The two independence scenarios result in more substantial reductions in inequality:
- Independence A (increase in benefit rates paid for by changes to the upper rate income tax threshold and rate) results in the poorest 20% of households having an increase in their weekly net income of about £19, while the richest 20% experience a decrease in weekly income of about £35.
- Under Independence B, the incomes of the poorest 20% of households remain the same as under the base case, since the basic income simply replaces the benefits received by the bottom quintile. Households in the second and third quintiles experience substantial increases as the basic income more than compensates for the extra tax paid. The richest 20% experiences a substantial decrease in their net incomes since they lose the benefits of the tax free band and the basic rate tax band on the first £44,000 of their earnings.
The fact that the Scottish Rate of Income Tax (SRIT) is a flat-rate tax effectively implies that income redistribution remains largely reserved to Westminster. Nonetheless, the analysis here shows that, even within the context of a fixed budget, the Scottish Government could achieve some reduction in inequality through its status quo policy levers. Council tax revaluation is a particularly effective tool for achieving change in the distribution of after housing cost income across households, but in order for the full effects of a Council Tax revaluation on inequality to be felt, it must be the case that any extra revenue raised is not redistributed using the SRIT.
Full control of the tax and benefit system would provide the Scottish Government with a more powerful set of levers with which to alter the income distribution. But it is also clear that achieving significant reduction in inequality through the tax and benefit system will require radical policy change. Independence scenario A for example involves substantial increases to out of work and low-paid benefit rates, combined with bringing higher numbers of individuals into the additional rate income tax. Despite the focus of the Scottish Government on universal rather than means-tested benefits, for example in the areas of tuition fees and prescription charges, a basic income is not on the political agenda. Radical changes are politically difficult, and being radical, the behavioural effects (e.g. work incentives even migration incentives) are less certain.