With Budget 2024 fast approaching, there’s a lot more focus on tax than before. The coalition government has vowed to deliver tax relief to most households, although the final design and cost of the tax package is yet to be revealed. But questions continue to be asked about if tax relief is inflationary, and why it’s even happening in the first place.

This analysis sets out the current level of taxation, why the proposed tax cuts aren’t inflationary (but why Liz Truss’s proposed cuts in the UK were), and how much it would cost to adjust tax brackets for either inflation or the same proportions of taxable earnings that previously prevailed – given that tax brackets haven’t been adjusted since 2011.

In the 2023 fiscal year, 50% of government tax revenue came from individuals, amounting to nearly $56b (see Chart 1). Most of this tax was from “source deductions” – tax taken “at source” during payroll. For most of us, that’s through PAYE the Pay As You Earn system, where employers send the income tax on employees’ wages straight to IRD. A quarter of the government tax take was from GST, the indirect tax of 15% on top of the price of almost all goods and services, amounting to $28b.

Corporate tax, the 28% tax on profits (not revenues) from companies, was the third-largest tax contributor. In 2023, it provided nearly $18b, or 16% of the tax pie. The remaining 9% came from other direct and indirect taxes, including resident withholding tax on interest and dividend revenue (for example the tax you pay on the interest earnings from your bank account) and alcohol, tobacco, petrol, and other excise taxes, levies, charges, and duties.

Tax makes up more than $9 in every $10 the government collects as revenue each year, and so trends in government revenue and taxation generally move together.

As Chart 2 shows, Core Crown Revenue as a proportion of GDP declined to 31.2% in 2023, just 0.2 percentage points above the long-term average of 31.0% over the past 30 years. But forecasts also showed Core Crown Revenue to be high relative to history over the next five years, averaging 32.7%.

Although Core Crown Expenses were set to drop back closer to their long-term average towards the end of the forecast period in 2028, Revenue was set to remain at levels not previously seen since 2006/07.

Over time, the higher level of revenue than expenses (the government earning more than it is spending) is set to pull the government accounts back into surplus.

We’ve previously discussed how tax relief could influence inflation, including a 2007 piece from Matt Nolan and a recent opinion piece from Chief Forecaster Gareth Kiernan. As always, the answer to “are tax cuts inflationary” is “it depends”. On what? The way they’re paid for.

Liz Truss’s ill-fated mini-Budget in the UK in 2022 provides a clear reference point. Ms Truss proposed £45b in unfunded tax cuts – essentially taking the same Budget starting point as before and then cutting tax by £45b. Her expectation was that the tax cuts would produce a substantial second-order effect, where lower tax would stimulate spending and investment, creating more economic activity, which could be taxed at the new lower rates. This model relies on unfunded tax cuts generating more GDP, and that GDP eventually being taxed. But higher GDP requires higher consumption and investment – both factors that, in a capacity-constrained economy, would generate inflation. Hence the political implosion that occurred.

This model isn’t what is planned for New Zealand. The coalition government is proposing “fiscally neutral” tax relief, effectively funding the tax relief from savings in government spending, so that there’s a switch between government spending and household spending.

Consider this example: The government and households spend $100m a year each, totalling $200m. If households were to spend $50m more, and nothing else changed, spending would total $250m – a clear expansion of demand. But if households spend $50m more and the government spends $50m less, then total spending remains the same, at $200m. The only change is who is doing the spending.

The Minister of Finance has recently highlighted that Treasury advice suggests that “fiscally neutral tax relief…reduces inflationary pressure and nominal interest rates”. That’s likely to be true, given that households on aggregate will save some proportion of their tax cuts. Households might spend 99% of the total funds, whereas government would’ve spent the entire 100%, thereby meaning that households will produce relatively less demand than government with the same funds.

So, tax cuts delivered by an equal reduction in government spending should be inflation-neutral, or possibly even slightly disinflationary.

New Zealand’s tax rates were last comprehensively adjusted in 2011, and aside from the introduction of a new 39% tax rate on income over $180,000 in 2021, these same tax rates have been in place for the last 13 years. Over that period, consumer price inflation of 36% has occurred, and average weekly earnings have increased 60%.

Higher earnings over time might only buy the same quantity of goods, as inflation erodes the value of money. For example, $100,000 today would buy roughly the same amount of stuff as $73,700 would have got you back in 2011. But even though you’re only getting the same amount of goods and services, you’ll be getting taxed more as your pay increases and more of your earnings are in a higher tax bracket.

As an aside, no, your entire income doesn’t get taxed all at the higher rate, but let’s talk about marginal tax rates next time.

Minimum wage earners also provide a useful comparison. In 2011, the minimum wage was $13.00/hr, meaning a full-time worker would earn $27,040 (gross) a year – far below the $48,000 level at which the next tax rate kicked in. That earner would have paid $3,752 in tax, 13.9% of their gross income. In 2024, the minimum wage is $23.15/hr, or $48,152 annually. That earner today would pay $7,465.60, or 15.5% of their income, with a minimum wage earner now falling into the third-highest tax bracket. Although only $152 of the 2024 minimum wage earner’s income working full-time is taxed at 30%, the fact that everyone working full-time now falls into the 30% tax backet is a concern.

Our analysis shows that inflation-adjusting the tax backets from 2011 to today would result in $4.6b less in personal income tax being paid, essentially meaning that around $4.6b of income tax collected this year is from inflation pushing people into higher income tax over the last 13 years.

Based on wage inflation (proxied by average weekly earnings), tax revenue in 2024 will be $6.9b higher than if income tax brackets were adjusted for wage inflation.

Inflation-adjusting tax brackets would see some substantial moves to income tax brackets, which are summarised below in Table 1.

If inflation-adjusted since 2011, tax brackets would look considerably different to the current reality. The previous $14,000 upper limit for the first bracket would now kick in close to $19,000, the $48k bracket would shift to $65k, and the $70k bracket to nearly $95k.

Given high inflation in recent years, even inflation-adjusting back to 2017 levels would cost $3.69b, with a nearly $4k increase to the lowest tax bracket, a nearly $13k increase to the second bracket, and a nearly $19k increase to the third bracket.

For inflation just over the last year, we estimate a ballpark cost of around $630m, which would see the brackets adjusted to kick in at around $14,600, $50,000, $73,000, and $187,000.

The other way to assess tax adjustments like this could be to look at the share of income earnt in each tax bracket. Using the proportions from 2011 and applying them to the 2022 taxable earnings distribution, 43% of taxable earnings fall into the $70-$180k tax bracket, compared to just 27% in 2011.

In 2011, earners exclusively in the bottom tax bracket earnt 4.6% of total taxable income, compared to 1.3% in 2022, and so on.

If the same proportions of taxable income were applied to current tax brackets today, the brackets would kick in much higher, with the $14k bracket shifting to $20,500, the $48k bracket shifting to $65k, and the $70k bracket shifting to $104k. Based on the 2011 proportion of taxable income, the $180k bracket would shift to $283,500. But given the $180 bracket came in for 2021, it’s reasonable to exclude any change to this threshold from this analysis.

If tax brackets were adjusted for proportionality as outlined above, and the $180k bracket remained as it is, the cost of forgone tax revenue would be $4.9b.

Treasury maintains a useful tool, the Aggregate Personal Income Tax Revenue Estimate Tool (APITRE). The public version of APITRE was last updated in September 2022 with tax year 2020 data, and Infometrics has updated the base data with tax year 2022 data – the latest available).

This tool allows us to broadly estimate income tax changes. The Treasury does note that “This tool is not designed to model large changes to the tax system because it assumes that the tax change has no impact on the taxable income distribution (i.e. it is exactly the same in both scenarios). In reality, large changes to the tax system are likely to alter the amount that people work.” Although true, inflation in just the last year has been high enough to trigger the tool’s warnings.

The tool allows for a 10% indirect and company tax offset, given that lower income tax paid would generally then see higher GST, company rax, and excise duties collected as people spent their tax cuts.