The Greens want a 45% top rate. Te Pāti Māori wants 48%. Both pitches lean on a tidy assumption, that a higher rate on income over $180,000 turns straight into revenue. New modelling from economists Peer Ebbesen Skov and Norman Gemmell shows why that assumption falls apart the moment real people get involved.
Raising the top rate from 39% to 45% would raise about $1.1 billion a year if nobody changed behaviour. Factor in people earning or reporting less, and that drops to around $560 million. Add the hit to GST as high earners spend less, and the net gain to the Crown is roughly $300 million. That is about 27 cents of every dollar on the campaign leaflet.
The authors are careful to say these are illustrative numbers built on 2024 IRD data, not official costings of either party’s policy. Fair enough. But the direction is not in doubt, and voters deserve the realistic figure, not the mechanical one.
Small behavioural shifts, big revenue holes
The model uses an elasticity of taxable income of 0.25, drawn from earlier New Zealand research. At 45%, a high earner keeps 55 cents of each extra dollar instead of 61 cents, a fall of about 10%. Apply the elasticity and reported taxable income falls about 2.5%. That sounds trivial. Spread across the highest-earning slice of the country, it halves the take.
The researchers put the revenue-maximising top rate at around 56%, so technically neither party is on the wrong side of the Laffer curve. But they warn that stronger responses among top earners, which some international evidence supports, could drag that peak much closer to today’s 39%. Treasury’s own assumptions show how slippery this is. A 2022 OIA release showed officials revising the elasticity for non-wage income up from 2.5 to 3, while cutting the wage-income figure from 0.2 to 0.1. Salaried staff barely move. Business owners, contractors and investors move a lot.
We have already run this experiment
This is not a hypothetical. When Labour introduced the 39% bracket from April 2021, IRD went looking for the money. In 2022, RNZ reported that high earners had declared nearly $3 billion less income than expected. IRD estimated the average affected self-employed person might report 13% less, dropping from about $191,000 to $166,000, and that structuring could cost around $150 million. Trusts and companies had jumped 28% since the rate rise was first promised.
John Cuthbertson, then tax lead at Chartered Accountants Australia and New Zealand, pointed to the obvious culprit in 2022, the 11-point gap between the 28% company rate and the 39% personal rate. Deloitte’s Robyn Walker cautioned at the time that legitimate business reasons explained some of the shift. Both can be true. Either way, the money moved.
A 2023 IRD briefing put hard numbers on it. The share of taxable income treated as trustee income rose from 58-62% before the 39% rate was announced to 71-73% after it was enacted.
Trusts soften the blow, companies do not
The trust question cuts both ways. If half of the vanishing personal income simply reappears in family trusts taxed at 39%, the Crown still collects it, just at a lower rate. Under that scenario the modelled revenue peak rises to 66%. That is the optimistic reading.
The less comfortable one is history. In 2025, Treasury’s analytical note traced trustee income surging through 2001-2010, when the top personal rate sat above the trustee rate, falling when the rates were aligned in 2011, then rising again from 2021. It put the revenue impact of this kind of sheltering at roughly 1% of GDP, and flagged timing games such as dividends pulled forward ahead of rate rises. Every gap in the system gets used. A 45% or 48% personal rate sitting beside a 28% company rate is a gap a good accountant will find by Christmas.
The cost nobody puts on the leaflet
Revenue is only half the bill. At 39%, each extra net dollar raised already costs the economy around $1.20 in efficiency losses. At 45% that climbs to roughly $2. That is extra hours not worked, deals not done, and expansion decisions shelved by exactly the owner-operators and senior staff who drive growth.
The authors also note that any inequality benefit depends entirely on how the money is spent, which is a more honest framing than most campaign material offers.
What happens next
None of this proves a higher top rate raises nothing. It proves the gap between the headline and the bank deposit is enormous, and that the gap widens as the rate climbs. Any party serious about the policy should publish costings that account for behaviour, trusts and the company rate, and explain how they would close the 28% loophole without hammering genuine small businesses. Until then, business owners should treat $1.1 billion claims as what they are, the number before anyone picks up a phone to their accountant.
Sources
- 1News: How much could NZ raise from a higher top tax rate? (2026-11-02)
- RNZ: How much could NZ raise from a higher top tax rate? We crunched the numbers (2026-11-03)
- The Conversation: How much could NZ raise from a higher top tax rate? We crunched the numbers (2026-11-02)
- RNZ: IRD says high income earners may be trying to avoid new tax bracket (2022-04-07)
- Treasury: OIA response 20210239 – Information on increase in income tax rates (2022-04)
- Inland Revenue Briefing Note BN2023/159: Behavioural response to the 39% personal tax rate (2023-05-16)
- Treasury Analytical Note 25/07: How ageing and other economic factors have impacted New Zealand’s tax system (2025-10)
Join the discussion
Add useful context, ask a good question, or challenge an idea — keep it specific and respectful.