The question of how to design tax policy that both speeds recovery from the current economic crisis and contributes to long-run growth is weighing heavily on the minds of policymakers the world over. Tax reductions to increase demand in the short run may conflict with tax reforms aimed at increasing output and promoting long-term growth. This is important as short-term tax concessions can be hard to reverse, implying that policies to alleviate the crisis could compromise long-run growth.
Lucas (1990) provides a very clear exposition of the view that capital income should not be taxed, at least in the long run, because such taxation reduces investment and, therefore, growth. Empirical support for this view is provided by Kneller et al. (1999) and Bleaney et al. (2001) who provide evidence that income taxes are more harmful for growth than consumption taxes.
In recent research (Arnold et al. 2011), my ex-colleagues at the OECD and I have carried the analysis further. We have:
- Studied a panel of 21 OECD countries over 34 years to estimate the effect of tax structure on growth in more detail than earlier work;
- Complemented this macroeconomic approach with a closer look at the underlying micro mechanisms, by using both industry and individual firm data; and
- Analysed how to use the tax system to both speed economic recovery and promote growth.
New evidence from 21 OECD countries
Although it was not possible to distinguish growth increases that are transitional from those that are sustainable in the long run, we do find that some tax changes can increase innovation and entrepreneurship and so influence long-run growth. To be cautious, however, we present our results on the assumption that the growth is simply transitional.
In this spirit, we report the long-run effects of changes in the tax mix on the level of GDP. The tax changes are revenue neutral tax, and the estimation controls for the accumulation of physical and human capital and population growth as well as the overall tax burden1. We show that an increase in the use of (personal2 and corporate) income taxes that is balanced by a decrease in the use of consumption and property taxes will reduce long-run GDP. Within this, we find that an increase in corporate income taxes (financed by an increase in consumption and property taxes) has a stronger negative effect on GDP per capita than a similar increase in personal income taxation (see appendix for table of results).
If, on the other hand, we increase consumption and property taxes at the expense of lower income taxes, we would expect long-run growth to be higher – where the positive effect of a property tax increase is substantially larger than for a consumption tax. Our results therefore suggest that a revenue-neutral shift away from income taxes would increase GDP per capita by between 0.25% and 1% in the long run.
Moreover, our analysis of the micro mechanisms reveals that reducing both corporate and top personal income tax rates would increase the rate of total factor productivity growth, the main driver of growth in per capita incomes in OECD countries. This suggests that some of the effects we present are likely to have a sustained effect on growth.
Tax cuts and economic recovery
Turning to the effect of tax cuts on economic recovery, the question arises as to whether these cuts will increase short-term expenditure on either investment or consumption. For instance, while cutting corporate tax is good for long-run growth through its effect on productivity growth, it is unlikely to increase investment quickly during a recession.
Cuts in the top personal income tax rate are similarly good for longer-term growth but, as with corporate taxes, temporary cuts are unlikely to increase demand much, as those on higher incomes are not credit-constrained. The arguments here are:
- consumers have a long planning horizon and a temporary tax cut will have a small effect on their permanent income;
- households realise that the tax cuts will have to be balanced by future tax increases. This indicates that, as with corporate tax cuts, there is a conflict between the wish to increase expenditure in the short run and the wish to promote growth in the longer run.
In contrast, cutting personal income taxes for credit-constrained low-income households can be expected to be relatively effective in the short run as such households are likely to spend more of any tax cut than high-income families, even if the cut is only temporary. Such a tax break would also stimulate labour supply (Brewer et al., 2010) and thus growth (see appendix).
Although it may not be consistent with a pro-growth tax agenda, it is sometimes argued that cutting consumption taxes is the best way of increasing consumption expenditures in the short term. A problem here is that a cut in the main consumption tax rate gives greatest absolute benefit to those who spend most, i.e. those on medium and high incomes who are likely to save most of any tax cut. Nevertheless, a temporary cut might be most effective as it could induce people to purchase durable goods earlier than they had planned, provided they could afford it. Indeed, the Symposium on the Economics of VAT Cuts in the March 2009 edition of Fiscal Studies (with papers by Crossley et al. 2009, Barrell and Weale 2009, and Blundell 2009) suggests that it was the temporary nature of the recent VAT reduction in the UK that accounted for the bulk of the consumption increase. This suggests that an almost equal temporary increase in consumption could have been achieved, without any budgetary cost, by leaving the rate at 17.5% for a short period and announcing that it would rise to 20% after that.
As house-price falls have been a major symptom of the crisis in several countries, there is a temptation to provide special tax measures to limit the fall. However, as discussed above, taxation of immovable property is the least harmful tax for economic growth. In fact, well-designed taxes on immovable property can even increase growth by reallocating capital away from tax-subsidised housing towards activities that are more productive. The only way that temporary tax concessions on housing could be beneficial would be a reduction in taxes on housing transaction (which discourage the efficient reallocation of housing and reduce labour mobility) that is linked to a future increase in recurrent taxes on housing.
In summary, the best tax cut for increasing demand and promoting long-run growth appears to be a reduction in personal income taxes and social security contributions on low-income households. This is likely to be particularly effective in countries where the cut can increase monthly incomes immediately, rather than waiting for a tax assessment at the end of the year. Moreover, it has the added distributional benefit of improving the living standards of families on modest incomes, both by directly increasing their disposable income and by giving them a greater incentive to work.
As economies emerge from recession, the challenge remains how to raise taxes without knocking the economy off course. To be clear, the rise in government debt has been such that simply restoring taxes to pre-crisis levels will not suffice unless permanent cuts in expenditure are planned. On the upside, however, this presents a unique opportunity to change the structure of the tax system to promote economic growth. We therefore argue that the tax increases after the crisis should focus on taxes that have been shown to be least harmful to growth, i.e. particularly recurrent taxes on immovable property and general consumption taxes.
Countries vary widely in their use of property taxes. So, while it is unlikely that those countries with already high levels of such taxes will want to increase them, there is considerable scope for raising them in the other countries. This will require careful planning as these taxes are often unpopular and perceived as regressive. Nonetheless, increases could be introduced gradually and regressivity avoided by regular re-valuations and provisions for people who have difficulty paying the tax.
Increases in general consumption taxes are also relatively good for growth. However, many European countries already have high VAT rates and raising their standard rates further could increase fraud. Nevertheless, many countries make considerable use of exemptions and lower rates of VAT. Substantial revenues could be obtained by removing these provisions. Some of them are designed to reduce the apparent regressivity of the tax, but they are poorly targeted because richer people spend more on these goods.
From a distributional point of view, it is better to have a uniform VAT on a broad base and use some of the additional revenues to assist low-income households.
Arnold, JM, B Brys, C Heady, A Johansson, C Schwellnus, and L Vartia (2011), “Tax policy for economic recovery and growth”, Economic Journal, 121:F59-F80.
Barrell, R and M Weale (2009), “The economics of a reduction in VAT”, Fiscal Studies, 30:17-30.
Bleaney, MF, N Gemmell and R Kneller (2001), “Testing the endogenous growth model: public expenditure, taxation and growth over the long-run”, Canadian Journal of Economics, 34:36-57.
Blundell, R (2009), “Assessing the temporary VAT cut policy in the UK”, Fiscal Studies, 30:31-38.
Brewer, M, E Saez, and A Shephard (2010), “Means testing and tax rates on earnings”, in Institute for Fiscal Studies, Dimensions of Tax Design: The Mirrlees Review. Oxford University Press, 90-173.
Crossley, TF, H Low, and M Wakefield (2009), “The economics of a temporary VAT cut”, Fiscal Studies, 30:3-16.
Kneller, R, MF Bleaney, and N Gemmell (1999), “Fiscal policy and growth: evidence from OECD countries”, Journal of Public Economics, 74:171-190.
Lucas, RE (1990), “Supply-side economics: an analytical review”, Oxford Economic Papers, 42:293-316.
Table 1. Estimated cross-country effects of the tax mix on long-run GDP per capita3
1 It should be noted that the value of the coefficient of the overall tax burden does not necessarily represent the effect of an increased tax burden on economic growth because it takes no account of how the additional revenue is spent.
2 The definition of personal income tax used in this study includes social security contributions paid by both employees and employers.
3All equations include short-run dynamics, country-specific intercepts and country-specific time controls. Standard errors are in brackets. *: significant at 10 % level; ** at 5% level; *** at 1 % level