Which taxes are best and worst for growth?

Tax affects economic growth by reducing consumer spending and lowering incentives to invest. But different fiscal policies have variable overall economic effects, with taxes on income better than those levied on corporate profits in terms of their wider impact on GDP.

Taxes generally have a negative effect on economic growth. Theoretically, they act as a disincentive on whatever is taxed – corporate taxes reduce business investment; and indirect taxes like value added tax (VAT) reduce consumption. Essentially, if your incentive to do an activity is reduced because some of that incentive is taken away in tax (that is, it is made more expensive), you wouldn’t do as much of the activity. This is the direct, negative effect on growth that is present in most taxes.

Taxes also take money out of the economy, reducing private sector demand and lowering GDP. For example, as income taxes reduce people’s take-home pay, they have less to spend. If the government doesn’t spend those tax revenues (via public services, social security payments, etc.) and instead uses them to pay down public debt, that is a direct reduction in GDP.

So, can taxes support growth at all? Yes, but the effect is indirect.

First, taxes enable governments to spend more. They can spend it in growth-enhancing areas – although this does depend on whether governments invest in areas that contribute more to economic growth than the areas from which the tax was taken.

Second, taxes can create a better business environment if improved public finances lead to lower economic risk and lower expected future interest rates. Then again, a raft of factors can influence this: it is entirely plausible that improved public finances under a government without a long-term plan can create a worse business environment than worse public finances but with a long-term plan to improve productivity.

Nuance matters. What’s important is where the tax comes from and how we leverage the advantages while minimising the disadvantages.

Which taxes have the biggest impact on growth?

Simulations can help economists to analyse possible changes in tax on the wider economy. Using NiGEM – the macroeconomic model designed by the National Institute of Economic and Social Research (NIESR) – it is possible to simulate the effect on GDP of a change in tax rates that raises an average of £3 billion of income per year. This exercise can help researchers to assess the impact and relative strength of taxes on the economy.

Raising the income tax rate has by far the least negative effect on GDP. In the long run, the simulation shows that the economy pretty much returns to baseline levels, with a slight increase in potential output.

The opposite is true for corporation taxes. A rise in the corporation tax rate leads to a severe and negative initial fall in GDP. Potential output also decreases. This leads to lower productivity, higher inflationary pressures and deteriorating economic circumstances in the long run.

A rise in indirect taxes (such as VAT) does not affect GDP quite as badly as a rise in corporation taxes, but it does affect GDP more substantially than a rise in income taxes. Indirect taxes operate largely through the price channel, increasing the prices of goods. By artificially raising prices, demand is curtailed.

Figure 1: The impact of an increase in different types of taxes on GDP and potential output (percentage difference from base)

Source: NIESR NiGEM model

So, purely from a growth perspective, the ‘worst’ tax to increase would be corporation tax (taxes on company profits). Companies won’t invest if they do not have the incentive to do so. This reduced investment is bad for the economy in the short run, but even worse in the long run. Without investment, productivity falters and economic growth further down the line is worsened. Any supposed benefits of improved public finances are more than offset by the long-term negative consequences.

Rises in income taxes, on the other hand, can be better for the economy (or at least, less negative). I say ‘can be’ because this statement also relies on a variety of factors.

The first factor is who gets taxed. Income taxes reduce personal income, which can lead to lower consumption and lower GDP. This depends crucially on the extent to which changes in income affect consumption.

Research shows that the consumption of those on lower incomes tends to be more sensitive to changes in their income than the consumption of those on higher incomes (Drescher et al, 2020; Jappelli and Pistaferri, 2014). This makes sense intuitively. If you barely make enough to eat each month, a boost to your income will let you relax a bit and perhaps treat yourself to something that you’ve been holding off from buying. But if you have far more than you need, an extra pound in your pocket won’t really change much.

The second factor is how much income tax increases will act as a disincentive for labour supply. In other words, will people work less if they are paid less?

Economic theory gives two conflicting answers to this question. With lower pay, you have less incentive to work (or to put it a different way, it costs you less to not work and enjoy some leisure), so you work less. Unfortunately, you would also need to work more to achieve the same income to maintain your level of consumption (for example, if your pay decreases, you might need to work more to afford to pay rent).

Although country-level data show that hours worked decrease as per capita income increases (Bick et al, 2018), evidence at the level of individuals suggests the opposite. This effect can vary dramatically, though, depending on people’s socio-economic status (Whalen and Reichling, 2017). Again, nuance matters.

The essential point that the economic effect of a rise in income taxes will vary greatly depending on exactly who is taxed. Income, gender, wealth, family status and other factors all have a part to play.

It is also worth noting that the preceding analysis is measured purely in terms of GDP growth. There are a whole host of other qualitative judgements that need to be made.

For example, income taxes have the worst immediate effect on real personal disposable income. The economy overall may grow, but people may not feel like this translates into better circumstances at an individual level – at least not until GDP growth translates into higher wages to make up for the higher taxes. A growing economy at the expense of household finances is a precarious situation politically.

What can be done to mitigate tax rises?

If governments plan to raise taxes, there are two main ways to mitigate the negative effects. The first way is to reinvest the tax revenue directly – for example, in green infrastructure, education or research. The second way is through an increase in business confidence (assumed to come about if better public finances lead to expectations of increased economic stability).

If the money is reinvested (that is, if the money raised in tax revenue is spent on public investment spending instead) the negative effects certainly diminish. In fact, NiGEM simulations suggest that for a rise in income taxes, there is a net positive effect as the distortionary impact of the tax is outweighed by the increase in long-run output, with only a very minor dip in GDP in the short term.

Figure 2: The impact of increasing taxes and reinvesting on real GDP (percentage difference from base)

Source: NIESR NiGEM model

It is also possible to explore the impact that increased credibility may have on the overall economic impact of increased taxes. If governments can convince financial markets that lower public debt leads to greater economic stability in the future, this can raise business confidence and increase investment.

Figure 3 provides an example, showing the range of outcomes under an increased corporation tax. If better public finances improve business conditions, the negative impact of the tax can be mitigated somewhat.

Figure 3: The differing impacts of a corporation tax increase on real GDP under different credibility scenarios (percentage difference from base)

Source: NIESR NiGEM model

Are there other options?

Optimal taxes should aim to avoid distorting economic activity – steering the economy away from the theoretically best distribution of goods under perfect competition and free trade. For example, a tax on something that I consume will lead to me buying less and the producer selling less – and we’re both made worse off as a result.

But if taxes have to be distortionary, they should ideally aim to be ‘corrective’ and spread the costs in a fairer way. The classic example of the value of a corrective tax is as a response to a factory that is dumping toxic waste into a stream. The pollution affects the health of the public using the stream. So, a fair corrective tax would place the cost of the pollution (the poor health of local people) onto the original polluter. By making the act of polluting more expensive, the polluter pays for the damage they cause. A distortion, but a fair one.

A carbon tax is a good example of this in practice. Considering the increasing severity of climate change, this tax is extremely important. There are several policies in place that act like a carbon tax in the UK.

There is an emission trading scheme that applies to energy-intensive industries, aviation, mining and power generation, which covers 28% of greenhouse gas emissions in the UK. There is also a carbon price support – a tax of £18 per tonne of carbon emissions. This applies to the power sector and covers about 24% of emissions. There are also excise duties on petrol and diesel, which indirectly tax carbon.

These taxes are good as they raise revenue, they act as disincentives to the production of harmful emissions and they place the costs of pollution on the polluters. Keeping these taxes in place and ensuring that they are adjusted to reduce carbon in line with net-zero targets is of paramount importance.

Another important development is the European Union’s carbon border adjustment mechanism (CBAM), set for introduction by 2027. Charging an additional price for the emissions embodied in relevant imports, this tax works to prevent 'carbon leakage', which occurs when carbon taxes in one country create incentives for the movement of production to a country without carbon taxes. This reform is a key part of the carbon tax framework.

Another alternative scheme worth considering is a land value tax. This is a tax levied directly on the unimproved value of land, not including anything built on it. Essentially, the value taxed is based entirely on location. This is a more efficient tax option that does not distort markets or restrict supply. This is because land is in fixed supply, so cannot be reduced in response to increased taxes.

Land value taxes also encourage development of land. Holding onto land without using it, purely for speculative purposes, is costly. But developing land as much as possible increases returns without increasing taxation.

In this way, it is also fair in the sense that landowners are taxed for public investment outside the landowners’ control. This enhances the value of the land (through transport links, for example) but not their investment on it (such as building improvements). Indeed, one study finds that shifting taxes from capital and labour to land can increase overall welfare in an economy by 5.2% and output by 26% (Kumhof et al, 2021). This is substantial.

In this example, the tax burden is shifted entirely onto producers (that is, the owners of land) and cannot be shifted onto consumers (that is, renters). Essentially, as landowners hold a monopoly on the land that they own, the rent that they charge is entirely demand-driven and already at the maximum value. This idea is supported empirically by studies finding that land value taxes are shifted entirely onto land capitalisation (Høj et al, 2018), meaning that the cost of the tax is absorbed entirely by changes in house prices – it is not passed on to tenants or future owners of the land.

In practice, the design of this policy should consider several factors. These include how the initial value of land is calculated and updated, whether there should be specific exemptions (for example, for places of cultural significance such as art galleries and cathedrals, which may own high-value land but have a low income stream to support it), and whether the introduction of this policy will increase financial risk, given that much land is mortgaged and so forms part of a bank’s balance sheet. Markets will adjust but may need to be given time to do so.

Conclusion

To summarise, the best way for governments to raise revenue would be through income taxes targeting the earners who are least likely to spend or work less as a result. If governments direct the revenue raised from this tax towards productive investment, while assuring financial markets of a clear, credible plan for future economic stability, the effect on economic growth can be maximised.

Beyond this, governments should also ensure that carbon taxes develop to meet net-zero targets and explore land value taxes as an alternative fiscal instrument. Taxation and growth are closely connected. A nuanced policy approach is vital.