Re-establishing fiscal sustainability in the UK will depend on the government’s ability to raise tax revenues via increasing tax rates and/or widening the tax base. Redistributive implications of potential policy changes should be a key consideration.
The economic impact of the pandemic has been enormous: the global economy contracted by 4.3 %, the biggest drop in output since the Second World War. Such a sharp downturn forced governments in many western democracies to deploy massive fiscal support measures to soften the blow on households and to increase social safety nets.
Combined with significantly lower tax revenues due to slower economic activity, the sharp rise in spending led to swelling ratios of government debt to GDP, previously unseen except during wartime. In advanced economies, the average ratio of government debt to GDP exceeded 120.1% in 2020; and it is projected to remain high in 2021 (reaching 121.5%, according to the IMF Fiscal Monitor, April 2021).
Current and future tax revenues
In principle, policy-makers can limit the surge in public debt by either reducing public spending or raising tax revenues. But in the UK at least, the deep public spending cuts during the 2010s, together with the continuing cost of fighting the virus, mean that there is little room for further reductions in government expenditure. Put differently, establishing fiscal sustainability in the near future will depend critically on the ability to raise tax revenues, as well as on the costs of servicing the debt (that is, interest rates).
Broadly speaking, there are two main sources of greater tax revenues: higher tax rates; and a wider tax base – the share of incomes, profits and consumption that are subject to taxation. First, if tax rates are increased, more is collected per activity. Second, greater economic activity raises incomes, profits and consumption, expanding the pool from which to extract tax revenue. In addition and importantly, governments can look to close loopholes and amend rules to increase the tax base.
The challenge that all governments face when seeking to increase revenues is that tax rates and tax bases sometimes work in opposite directions. Taxes by their nature are ‘distortionary’, meaning that they change people’s behaviour: higher taxes reduce the incentive to earn income, potentially lowering economic activity and hence the tax base.
To evaluate the effectiveness of raising tax rates for raising revenues, it is important to consider the effects of the behavioural response of households and firms to the change in tax rates.
The macroeconomic impact of changes in tax rates
There is a growing body of evidence to suggest that tax changes lead to large responses of the aggregate economy (and thus in individuals’ behaviour). For example, it is estimated that for every $1 rise in tax revenues in the United States, gross domestic product (GDP, the aggregate measure of income) falls by $3; that is, the size of the ‘economic pie’ responds strongly and gets significantly smaller as a result of tax rises (Romer and Romer, 2010).
Indeed, aggregate economic activity is extremely sensitive to tax changes. A rise in tax rates leads to an economically significant short-run reduction in the size of the economy. This applies equally to changes in personal income taxes and corporate taxes (Mertens and Ravn, 2013).
Another study distinguishes between marginal income tax changes (the tax paid on the next unit of income) and average income tax changes (total tax as a proportion of total income), and finds substantial impacts on the aggregate economy when tax rates change (Mertens and Montiel Olea, 2018). In the UK, changes in income taxes have a higher impact on output than changes in consumption taxes, and the UK changes consumption tax rates as often as income tax rates.
Increases in tax rates also often result in higher government debt (Mertens and Ravn, 2013; Mertens and Montiel Olea, 2018), even in the face of lower public spending. That is, government debt increases despite a reduction in spending and an increase in tax rates, pointing to significant falls in tax revenues due to lower economic activity.
There is evidence that initiatives to reduce government debt by cutting public spending (and particularly transfer payments to households) are more successful than those relying on raising tax rates. That tax changes lead to greater movements in economic activity than spending changes also fits with recent empirical findings pointing to tax multipliers (the change in output as a result of a unit change in tax revenues) being larger than spending multipliers (Ramey, 2019).
Our recent research suggests that over the medium run (two years and longer), tax revenues do not increase in response to tax rate rises, and they often fall. These results come from a variety of different methodologies and are based on data from several different countries.
Importantly, our analysis distinguishes between changes in average taxes (total tax paid as a share of the total tax base) and changes in marginal taxes (the tax paid on the next unit of income). This distinction is important: whereas it is the average tax rate that determines how much revenue is raised overall, it is the marginal rate that drives individual incentives and behaviour.
With progressive taxes, where a higher rate of tax is paid on higher incomes, marginal tax rates are higher than average taxes and this difference is greater the greater the degree of progressivity. Importantly, we show that tax hikes that lead to an increase in the degree of progressivity are unlikely to generate higher tax revenues due to the behavioural responses of households and firms.
Simply put, there is a trade-off between revenue collection and the redistributive nature of the tax system; the more progressive and redistributive a tax system is, the less revenue is raised (for a given level of overall taxes).
How progressive are existing taxes?
A progressive tax system is one in which the rate at which individuals pay taxes increases with their income. It is important to note that currently tax rates are not progressive, or at least not as progressive as we might believe. While income taxes are typically progressive, they do not make up the majority of total tax revenues, and other taxes tend to be less progressive (and, in fact, are often regressive).
For example, consumption taxes take up a disproportionately higher level of poor households’ income. Indeed, one study shows that US tax rates are nearly flat with the poorest 10% of households paying rates of 25.6% tax on their incomes, and those households between the 80th and 90th percentile being taxed at 29.4% (Saez and Zucman, 2019). As the very rich have more flexibility over how and when they receive their income, the average total tax rate on the richest 400 US individuals is calculated to be at 23%.
In the UK each year, the Office for National Statistics (ONS) conducts the ‘Livings Costs and Food Survey’, reporting information on household income, taxes and benefits (as well as components, most extensively food expenditure). Using this data set, we calculate average tax rates per income ‘decile’ (tenth of the working population), as presented in Figure 1.
Figure 1: Direct tax rates by income decile
Source: Authors' calculations based on ONS data
As Figure 1 shows, while direct taxes of income and national insurance are broadly progressive, the remaining tax burden is regressive. This leads overall taxes to be close to a flat tax rate – households in different income groups pay approximately the same percentage of their income in taxes.
Those in the third decile pay an estimated 29% of their income in tax, whereas all other groups (outside the first decile) pay between 32% and 35%. Households in the lowest decile are seen to pay the highest average tax rate, although this is frequently thought to be due to temporarily lower incomes and/or an underreporting of incomes on the corresponding survey.
Overall, the conclusion from the UK is similar to that from the United States: the portfolio of different taxes leads to a broadly flat tax system. Those on low incomes and high incomes pay approximately equal proportions in tax.
What are existing marginal taxes?
It is a similar story for marginal taxes. The complication here is that many poorer households receive benefits from the government that are removed the higher their income becomes. This is an ‘effective marginal tax’, as net take-home earnings fall with greater income, both from higher taxes paid on this greater income and through the retraction of benefits.
A recent analysis of taxes and benefits in the United States finds that ‘one in four low-wage workers face marginal net tax rates above 70%, effectively locking them into poverty’, and that more than half of individuals have a lifetime marginal tax rate above 45%.
Similarly, in the UK, the introduction of universal credit means that for those on lowest income, for every £1 of post-tax income, benefits are cut by £0.55. This ‘tapering rate’ was reduced from £0.63 in the recent budget, announced on 27 October 2021, and as stated by the Chancellor of the Exchequer; ‘is a tax on working people’ (Hansard, 2021).
But even this lower taper rate can lead to marginal income taxes on low-income individuals going above 80%. This is because those on universal credit lose £0.55 for every £1 earned, but before this £1 is earned, taxes have to be paid. Looking across the whole income distribution and at all income related taxes, most workers pay marginal taxes above 45%. This means that most individuals (those on low and high incomes) face high marginal taxes on their income, which act as a disincentive.
Are there alternative solutions?
The surge in government debt levels in the wake of the pandemic is forcing policy-makers to find ways to increase tax revenues. As is set out above, the existing evidence suggests that this in unlikely to be achieved by increasing tax progressivity. What is worse, taxes in many countries are not especially progressive at present and have been becoming less progressive over time.
Is there a viable solution? Perhaps. First, it is important to note the differential effects of changing average and marginal taxes (as discussed above); increases in the latter do not increase revenue but those in the former do. Importantly, in many cases marginal tax rates are high across the income distribution and are often highest for the poorest groups.
Reforming the tax structure by bringing average and marginal tax rates more in line such that they provide fewer behavioural disincentives could lead to higher revenue, without increasing inequality or decreasing output. Such reforms could be particularly effective if they can broaden tax bases, reduce loopholes and distortions, and remove the incomes, consumption and profits that currently get taxed at 0%.
These prescriptions echo the findings from the 2011 Mirrlees Review, a comprehensive analysis of the existing UK tax structure, which concluded that the system was ‘inefficient, overly complex and frequently unfair’.
A clear alternative would be the use of unconditional benefits that do not get retracted and thus do not lead to high marginal tax rates on the poor. If this were paid by a flatter (less progressive) set of taxes, the economy would redistribute income in a less distortionary way, reducing the disincentives from high marginal tax rates.
Such unconditional benefits have become increasingly popular over the past decade in discussions around a ‘universal basic income’. Our results reveal that reforming the tax and benefit structure such that taxes are simpler and flatter, and benefits are unconditional and larger, could both increase incomes and tax revenues and reduce inequality.
In short, reducing the disincentives to work can increase tax revenues and thus improve fiscal sustainability, while unconditional benefits can maintain a redistributive system that supports those on lower incomes. Such a system would therefore address both debt sustainability and income inequality, the two key policy challenges of our time.
Where can I find out more?
- A future with high public debt: low-for-long is not low forever: IMF blog discussing the fiscal policy needed to anchor expectations for a riskier future
- Post-pandemic debt sustainability in the EU/euro area: this time may (and should) be different: Lorenzo Codogno and Giancarlo Corsetti consider the European Union’s Recovery Plan and the boost it should offer to the economy and fiscal revenues
- The art of assessing public debt sustainability: Discussion of principles to guide the design and implementation of debt sustainability frameworks
- What the pandemic means for government debt, in five charts: World Bank blog
Who are experts on this question?
- Richard McManus, Canterbury Christ Church University
- F. Gulcin Ozkan, King’s College London
- Dawid Trzeciakiewicz, Loughborough University
- James Cloyne, UC Davis
- Giancarlo Corsetti, University of Cambridge
- Jonathan Ostry, IMF
- Ricardo Reis, LSE