2 November 2016
Abolish twenty taxes and go for growth, says IEA
- Measuring taxation and government spending as a proportion of national income is beset with difficulties. However, it is clear that there has been a strong upward trend in taxation and government spending as a proportion of national income in the developed countries over the last 100 years. At the beginning of World War I, the UK government spent just over one eighth of national income; now it spends between 40 per cent and slightly over 45 per cent (depending on how national income is measured).
- Government spending as a proportion of national income in the UK is at a similar level to that in Germany but is considerably higher than in Switzerland, Australia and Ireland. Government spending is also somewhat higher in the UK than in Canada, the US and New Zealand. Thus, it is clear that a welfare state of significant size can be provided at levels of government spending far below those in the UK. It is also notable that, in some UK regions, government spending is between two thirds and three quarters of regional GDP, depending on the definition employed.
- It is government spending rather than taxation that ultimately determines the total burden of government activity on the private sector. Although government spending can be financed by borrowing or creating inflation, all government spending is normally a call on resources that have alternative uses or involves transfers from one group in society to another. If a government borrows money, this has to be financed at the time of borrowing and then serviced by future generations of taxpayers.
- Although there is a general perception that there has been a significant reduction in government spending since 2010, there has not been. Nominal spending increased by 4.6 per cent in total between 2010 and 2015 and real spending fell by just 0.5 per cent a year. Real spending per capita fell by a little more – about 1 per cent per year. By 2015, government spending remained at over 45 per cent of GDP (at factor cost). over the five years to 2020, nominal government spending is planned to continue to rise, real government spending to go up slightly and spending as a proportion of national income at factor cost fall to 41 per cent.
- Some forms of government spending can promote growth, especially capital spending. However, this does not mean that the resources concerned might not be better employed elsewhere, including for supply-side enhancing tax cuts. Spending on government consumption tends to harm growth. Spending on badly designed transfer payment systems can be especially damaging as they can reduce work incentives. Despite rising government spending in general, there has been a fall in spending on investment while there has been a huge rise in spending on government consumption and transfer payments. Government spending on investment is extremely volatile, but, by way of example, it has fallen from over 6 per cent of national income in the late 1960s to less than 3 per cent today. Welfare payments have almost doubled over the same period as a percentage of national income to their current levels of around 14 per cent (this excludes health and education spending). Even during the period 2010–15, there was a significant reorientation of spending. Public order and safety (comprising much of the ‘nightwatchman’ functions of the state) had reductions in real spending of about 15 per cent while transport saw cuts of 10 per cent. Meanwhile, spending on both health and social protection increased in real terms. These trends will not be reversed and may be further reinforced in the years to 2020.
- The taxes that have to be raised to finance government spending damage growth through a number of channels. These include reduced incentives for work and saving and reduced incentives for entrepreneurship. over the last three decades a considerable amount of evidence has been amassed that suggests that tax has a significant negative impact on growth. This evidence suggests that a 10 percentage point increase in the tax or government spending burden is associated with approximately a 1 per cent fall in the growth rate in the long term. The negative impact of tax on growth is also far greater than the benefits to growth that might arise from increased government spending, even if that spending is on investment or subsidising private sector research and development. There are, however, problems with the ‘growth regression’ approach that generates these results. For example, they cannot necessarily distinguish between cause and effect.
- In many ways, regulation and tax are substitutes for each other and have similar effects. For example, politicians can attempt to improve living standards by regulating wages and other working conditions or by providing in-work welfare benefits. Robust new modelling has been undertaken as part of the research for this project on the impact of tax and regulation on productivity. This modelling is not subject to the problems of the earlier tax and growth studies. This research finds that a 10 percentage point fall in a combined index of top marginal tax rates and regulation relative to its trend produces a rise in output over about thirty years of 24 per cent. This is equivalent to an increase in the growth rate over the thirty years following the cut of about 0.8 percentage points per annum. The model does not distinguish between regulation and tax. However, the model uses data from a period in which falls in marginal tax rates have been much more frequent than changes in regulation. As such, it is reasonable to infer that a 10 percentage point cut in the top marginal tax rate would bring about the improvement in growth indicated by the model.
- In theory at least, we can identify a growth-maximising, a welfare-maximising and a revenue-maximising share of taxation in national income. The last of these points is the ‘top’ of the Laffer curve – that is, the point at which, if the government tries to raise more in taxation, the negative effects of taxation on growth are so great that tax revenue is actually reduced. The maximum sustainable level of government spending is a little different from this as a result of the existence of non-tax-based government revenue and the complications of debt dynamics. The welfare-maximising share of taxation in national income is greater than the growth-maximising share because there may be some functions that increase welfare while reducing measured economic growth.
- The growth-maximising share of government spending in GDP is probably in the range 18.5–23.5 per cent of national income, using current (July 2016) UK definitions at market prices. The welfare-maximising share is probably in the range 26.5–32.5 per cent of national income. The maximum sustainable level of government spending is probably in the range 37–38 per cent of national income. Even on the projections set out in the March 2016 Budget and without any relaxation of the purse strings, government spending will only just come down to sustainable levels by 2020/21.
- All taxes do not affect growth in the same way. For example, taxes on mobile capital and high marginal rates of tax on income affect growth disproportionately. Taxes on land, consumption and on economic activities that lead to harmful ‘spillover’ effects reduce growth to a lesser extent and can even improve economic welfare.
- The overall design of the tax system also affects economic growth. Ideally, a tax system should have low negative effects on welfare and economic efficiency; low administration and compliance costs; fair and non-discriminatory procedures in the way companies and individuals are treated; and be transparent and easily understandable. The UK system is a long way from meeting these goals.
- More specifically, tax systems should be designed so that there is a broad understanding of the basic facts of who pays how much, with simple rules, thresholds, schedules and rates. Tax should be codified so that taxpayers know what is expected and can arrange their affairs accordingly with any ambiguity kept to a minimum. To facilitate this, a formal tax strategy should be adopted by the government, with changes announced in advance and with explanations of how each change helps bring the system closer to implementing the overall strategy.
- The UK has a very badly designed tax system with high marginal rates, huge complexity, taxes that discourage wealth-creating economic activity and wide-ranging exemptions. Current property taxes provide a good example of how badly our tax system is designed. These taxes raise nearly 4 per cent of national income. Council tax is based on a complex system of thresholds and is, in fact, regressive, at least in part. Stamp duty land tax artificially depresses property values, discourages investment and distorts the allocation of assets – for example, by discouraging older people from moving to smaller accommodation when they no longer need as much space. Current property taxes should be abolished and replaced with taxes that damage economic welfare to a much lesser degree.
- overall, the UK government should abolish twenty current taxes. Among other taxes, corporation tax, national insurance, capital gains tax, inheritance tax, council tax, business rates, the television licence fee, the apprenticeship levy, stamp duties, alcohol duties, tobacco duties, vehicle excise duty and air passenger duty should be abolished. A radically reformed tax system should be designed to raise around 20–25 per cent of national income. It would comprise the following main elements:
- A flat-rate income tax set at 15 per cent of income above a personal allowance of around £10,000. Distributed corporate profits would also be taxed at this rate.
- VAT set at 12.5 per cent, with most exemptions abolished.
- A new housing consumption tax on rents and imputed rents to mimic VAT at 12.5 per cent.
- A new location land-value tax.
- Fuel duty retained at around half the current rate.
- If this package were implemented, static modelling would suggest that the poorest decile would enjoy tax cuts worth 26 per cent of gross income, followed by 19 per cent, 17 per cent and then 13 per cent for the fourth poorest decile before further falling to 7 per cent for the fourth richest decile. The benefit of the proposed tax changes then rises slightly with income.
- Because lower taxes would lead to higher growth and because there would be a tax system that led to far fewer distortions of economic decisions, it is likely that employment, productivity and wage levels would rise considerably. These effects are likely to disproportionately benefit the poor as they are more likely to be at the margins of the labour market in insecure, low-paid employment or unemployed.
- Tax policy should be formulated to increase certainty, stability and transparency. The government should set out a formal tax strategy that expresses its medium-term and long-term intentions. Tax policy decision-making should be taken out of annual budgets and implemented separately from periodic statements involving economic and fiscal forecasting so that fuller attention can be devoted to the implications of both tax policy changes and changes in budgetary and economic forecasts, respectively.