Norway raised the top rate of its wealth tax in 2022. Among the households the increase reached, out-migration rose from 0.2 per cent to more than 2 per cent inside a single year.
That is the sharpest departure response to a wealth tax anyone has measured, and it is the strongest published evidence against the case this site argues. Christine Blandhol's study Curbing Tax Flight? takes it from Norwegian tax filings, which record when a person ceases to be resident, and then follows the businesses those people owned.
The composition is worse than the headline. 40 per cent of those who left were firm owners, and their companies went on to record revenues about 12.6 per cent lower than comparable firms whose owners stayed. The departures took operating businesses with them.
That is the part of the result that should trouble anyone arguing for this policy. Wealth at the very top is distributed so unevenly that a headcount and a revenue figure can differ by an order of magnitude, and the productive assets attached to that wealth are distributed the same way. An owner leaving is not one taxpayer leaving. The firm's investment decisions and its supplier relationships travel with the owner, or the revenues do, which on this evidence amounts to the same thing. Read as a migration rate, the Norwegian number understates what happened.
The question is not whether anyone leaves. It is how many, and what it costs. Norway gives an unusually clear answer to the first and a partial one to the second.
What is counted and what is modelled
Blandhol also estimates that the reform reduces long-run aggregate output by 1.3 per cent. That figure does not belong in the same category as the migration numbers.
A rise from 0.2 per cent to above 2 per cent is a count taken from records of who filed where. The 1.3 per cent is what a model produces when those counts are projected forward under assumptions about investment, about how many of the departures are permanent, and about what the firms would otherwise have done. The estimate is disputed in the literature and it is presented here as contested.
Anyone citing Norway against a wealth tax should be able to say which of the two they mean. The migration counts came out of filings, so a different set of assumptions leaves them exactly where they are. The 1.3 per cent would change.
What the case does not settle
Two things limit what the Norwegian result can tell us about Britain, and neither of them makes the departures smaller.
The first is a confound. The 2022 package raised dividend tax at the same time as the wealth tax, so an owner weighing up a move faced both changes at once. The study cannot separate the two.
The second is the threshold. Norway's wealth tax begins at around 1.7 million kroner of net wealth, roughly eighty times lower than the £10 million floor in the model this site works from. At that level it reaches people running ordinary companies, whose money sits inside the firm instead of beside it. An annual charge on an illiquid holding can mean pulling cash out of a working business to pay it, which is a reason to leave that a tax starting at £10 million generates far less often.
That is a real difference and it is not a rebuttal. A higher threshold changes who is affected. It does not establish that the response among those people would be milder, because no country has run that experiment.
Where this sits in the wider evidence
Norway is the sharp end of the distribution. The general estimate for wealth taxes comes from Scandinavian administrative data analysed by Jakobsen, Kleven, Kolsrud, Landais and Muñoz in the American Economic Review: one percentage point on the top rate reduces the stock of wealthy taxpayers by about two per cent. The wider economic effects they measure are small, with value added down 0.10 per cent.
The two findings are compatible. An average taken across Scandinavian rate changes will not show what one large increase did to the very top of one country in one year, and the Norwegian result is a warning about the tail of the distribution rather than about the average.
The half of the paper that rarely gets quoted
Blandhol does not stop at measuring the exodus. The same study models an exit charge levied on the market value of the firm, and finds that it curbs tax flight, with the largest effect among the more productive entrepreneurs, while raising aggregate output.
That result is modelled, so it carries the same discount as the 1.3 per cent estimate. No country has been observed running the policy it describes.
Where the finding sits matters. The departure figures in this paper are the ones quoted against wealth taxes. The exit-charge result sits in that paper too, and it is quoted far less.
The point Norway actually settles
Every tax provokes a response, and the useful questions about any of them are how large the response is, which design choices change it, and what each of those choices costs.
Norway answers the first with an uncomfortable number, a rise from 0.2 per cent to above 2 per cent in twelve months. It says almost nothing about the other two. Using it as though it closes all three is the move to watch for, and it is made in both directions: by people who cite the departures as a settled verdict on wealth taxation, and by people who reach for the confound and the threshold before they have stated what happened.
What would settle it
The evidence that would move this argument does not exist yet.
A Norwegian revenue series covering the years after 2022 would show whether the wealth tax base left with the people who left, which is the question a headcount cannot reach. A study that separates the wealth tax from the dividend tax, using a group exposed to one and not the other, would show how much of the response the wealth tax itself caused. Either would change what this case proves.
The remaining gap may never be filled. No country levies a recurrent charge at a genuinely high threshold on a population whose wealth is mostly liquid, so whether illiquidity drives the response or merely accompanies it has nowhere to be tested.
Until then Norway remains the best-evidenced case there is, measured on a tax that starts about eighty times lower than the one under discussion here, alongside a second tax rise it cannot be separated from.
Blandhol counted who left and what became of the firms they owned. Whether the Norwegian wealth tax base left with those people is a separate question, and the receipts evidence that would answer it is not yet in.
Sources and evidence status
Blandhol, Curbing Tax Flight? — Princeton working paper, December 2025 (KB-055). Jakobsen, Kleven, Kolsrud, Landais & Muñoz, Taxing Top Wealth — American Economic Review, 2026 (KB-054). Threshold parameter for the reference model: Tax Justice UK (KB-002).
Migration figures are T1 Verified Fact, measured on Norwegian administrative data. The 1.3 per cent long-run output effect is T3 Modelled Estimate and is recorded as contested in Document 02. The exit-charge finding is also a model result from the same paper and is not an observed outcome. The 2022 reform raised dividend tax alongside the wealth tax, so the departure response cannot be attributed to the wealth tax alone.
No figure in this article is an estimate for a UK tax on wealth above £10 million. No such estimate exists.
0 comments