The 76-Öre Problem: Wealth Taxes, Exit Taxes and the Cost of Raising Revenue

Vänsterpartiet's 2026 campaign has reopened a familiar Swedish argument about taxing mobile wealth. The party wants a billionaire tax and has separately backed exit taxation for individuals. The dispute is usually framed as fairness against flight: either the very wealthy contribute more, or higher taxes drive them abroad.

The more useful question begins after that argument. Once migration, changes in saving and investment, tax planning and other behavioural responses are included, how much of a mechanically expected tax increase actually reaches the state? And what does the economy give up to collect it?

Editorial image generated by the author.

The best Scandinavian evidence does not support the strongest version of either side's case. Wealth taxation does cause some wealthy people to move, but the resulting loss of investment and employment appears much smaller than the most dramatic warnings suggest. Migration is also only one part of the response. Combine it with changes in wealth accumulation, investment, avoidance and evasion, and much of the revenue that appears available on paper disappears before it reaches the treasury.

A tax can therefore raise revenue while being far more expensive than its headline yield suggests. Positive tax revenue and positive economic value are not the same thing.

Norway is a warning, but not quite the one usually presented

Norway has become the obvious European case study. After increases in wealth and dividend taxation, the country saw an unusual wave of departures among its wealthiest residents. Figures assembled for Norway's 2026 tax commission cover people who had net wealth above NOK 100 million or appeared on Kapital's list of the 400 richest Norwegians during 2021–2025. Sixteen people in this group emigrated in 2021. The number rose to 46 in 2022 and another 46 in 2023, before falling back to 16 in 2024 and 17 in 2025. The 2022 group had paid NOK 285 million in wealth tax the previous year; the 2023 group had paid NOK 130 million.

The commission cautions that migration has many causes, including work, study and family. It nevertheless reports that many of the wealthiest emigrants themselves identify Norwegian capital taxation as a central motivation, particularly the combination of wealth tax and dividend tax.

That does not prove the more dramatic claim that Norway lost more tax revenue than it gained. Popular calculations sometimes multiply emigrants' estimated fortunes by the headline wealth-tax rate, ignoring valuation discounts, exemptions and the difference between economic wealth and the taxable base.

The chronology is also often muddled. Norway did not introduce an exit tax and then watch its billionaires flee. A general exit tax on unrealised share gains had existed since 2007. What changed between 2022 and 2024 was that Norway tightened the regime substantially, including abolishing an old five-year limitation, as the government tried to ensure that gains accumulated while resident in Norway remained taxable there.

The stronger criticism begins after the revenue is counted. Even if a wealth tax raises money overall, how much of the potential revenue survives once taxpayers respond?

What happens to a SEK 100 tax increase?

A study by Katrine Jakobsen, Henrik Kleven, Jonas Kolsrud, Camille Landais and Mathilde Muñoz, forthcoming in the American Economic Review, offers unusually good evidence. Using Swedish and Danish administrative data and several major historical wealth-tax reforms, the researchers examine both migration among wealthy taxpayers and what happens to the businesses they own.

Their estimate is that a one-percentage-point increase in the top wealth-tax rate reduces the long-run stock of wealthy taxpayers by roughly 2%. Migration, however, is not the largest fiscal response.

Estimated fiscal response	Revenue effect
Mechanical wealth-tax increase	+100
Migration response estimated in the new study	−22
Intensive-margin response, calibrated from earlier Danish evidence	−54
Combined implied net revenue	+24

The provenance of those numbers matters. The new study estimates that migration, including the associated loss of revenue from other tax bases, removes about 22 öre for every krona mechanically raised by the wealth tax. The additional 54 öre comes from earlier Danish research on the intensive margin: changes in saving, investment, avoidance and evasion. Combining the two, the authors calculate that roughly 76 öre of every mechanically expected krona can disappear through behavioural responses. The full paper estimates the corresponding marginal cost of public funds at about 4.2.

The 76-öre figure is therefore not a single observed loss from one reform. It is a calibration built from two strands of Scandinavian evidence. Nor does it imply that the historical wealth taxes were beyond the point at which higher rates reduce total revenue. The researchers conclude that migration effects were too small for abolishing the taxes to pay for itself.

The difference between SEK 100 and SEK 24 is nevertheless hard to dismiss. A government could correctly say that a tax increase raised revenue. It would be equally correct to say that most of the revenue apparently available before behaviour changed never materialised.

There is another reason to be cautious about applying the estimate mechanically to a future Swedish billionaire tax. The reforms used to identify the migration response systematically reduced or abolished wealth taxes. The authors explicitly say that they cannot rule out asymmetric behaviour: people may react differently to a new tax increase than they did to a tax cut. Enforcement, international information exchange and the tax systems of competing countries have also changed since the historical Scandinavian reforms.

The billionaire abroad is only the visible part

A wealthy entrepreneur moving abroad makes an excellent newspaper story. There is a person to photograph, a fortune to quote and an apparent tax loss to calculate. Most behavioural responses leave much less visible evidence.

Most wealthy taxpayers do not move. They change what they do: saving differently, altering dividend policies and ownership structures, moving assets between categories, changing borrowing and investment decisions, or devoting more effort to tax planning. According to the Scandinavian calibration, these responses are fiscally more important than migration itself: about 54 öre of lost potential revenue compared with 22 öre associated with migration.

This also makes the familiar claim that an emigrating entrepreneur simply takes an equivalent slice of the economy along too crude. The Swedish evidence shows substantial losses inside firms controlled by wealthy emigrants, including lower employment, investment, value added and tax payments. Much of the effect is associated with firms closing.

But economies reallocate resources. Around 45% of firms that closed following wealthy-owner emigration were subsequently absorbed through mergers with other Swedish firms. Accounting for those acquisitions reduces the estimated employment effect by around 40%. Workers can find other employers, viable businesses can acquire new owners, and other investors can finance some projects abandoned by departing owners.

The resulting aggregate effects are much smaller than the firm-level losses. A one-percentage-point increase in the top wealth-tax rate is estimated, through the migration channel, to reduce aggregate employment by roughly 0.02%, investment by 0.07% and value added by 0.10% in the long run. Those effects are real, but they do not support the idea that every millionaire crossing the border takes an equivalent share of national production along.

There is, however, one kind of loss that is much harder for administrative data to capture.

The company Sweden never sees

Administrative records are excellent at observing a wealthy Swede who leaves after becoming successful. They are much worse at observing someone who leaves before becoming wealthy.

Consider two entrepreneurs. The first has spent 25 years building a Swedish industrial business and now owns shares worth SEK 5 billion. Sweden raises capital taxation and she considers moving abroad. Her departure will appear clearly in tax records. Researchers can observe the taxes she stops paying, follow her company and estimate what happens afterwards.

The second is 29 years old and owns shares worth SEK 5 million in a young company that might eventually become worth billions. He looks at the combination of future wealth taxation and an exit tax on unrealised gains and concludes that, if the company succeeds, moving later could become prohibitively expensive. So he moves while the company is still small. In a study of wealthy emigrants, Sweden has barely lost anything. In the counterfactual Swedish economy twenty years later, the difference could be substantial.

The same measurement problem applies to people who never arrive. A German engineer or founder choosing between Stockholm, Copenhagen, Amsterdam and London does not need to flee Sweden if Sweden is never selected in the first place. A venture-capital firm can locate its next fund elsewhere. An international specialist can accept another offer. These decisions do not appear in Swedish statistics as lost taxpayers because Sweden never acquired the tax base.

This is a hypothesis about a difficult counterfactual, not something the Scandinavian migration study proves. That limitation is precisely why it matters. Historical tax records are naturally much better at measuring what happens to an existing stock of wealthy residents than at measuring companies, founders and investment that might have existed under another policy.

When an exit tax changes the timing

Norway's tax commission has begun grappling with this problem directly. Representatives of startup and growth companies told the commission that they were considering moving earlier than they otherwise would because an eventual exit-tax liability could become unmanageable if their companies appreciated substantially. Three commission members went further, warning that the combination of Norway's wealth tax and strict exit taxation could encourage young entrepreneurs to leave before establishing a business.

The broader commission does not adopt that stronger conclusion as established fact. It does, however, acknowledge that exit taxation can affect inward mobility as well. For people expecting sufficiently large future gains, the possibility of an exit-tax claim may reduce the attraction of moving to Norway in the first place. The commission therefore proposes an exemption for people who live in Norway only temporarily.

It also recommends substantially lowering Norway's wealth-tax rate while making valuation more uniform. The commission could not agree on a single rate, but proposed a range between 0.25% and 0.75%, compared with today's top rate of 1.1%.

A narrow exit tax can still be defended. Suppose someone acquires shares for SEK 1 million while resident in Sweden, watches them rise to SEK 101 million and then deliberately moves to a low-tax jurisdiction immediately before selling. There is a coherent argument that Sweden should retain taxing rights over the SEK 100 million gain accumulated during Swedish residence.

The 76-öre estimate, however, is evidence about recurrent wealth taxation, not exit taxation. Exit taxes enter the argument differently. They alter the cost and timing of migration. If recurring taxation is already high, making departure after success expensive increases the value of deciding where to live and establish a company before success arrives. A rule intended to prevent late tax-motivated departure can therefore strengthen the incentive for an earlier one.

Taxing the stock without shrinking the flow

Governments can see today's stock of wealth. There are a known number of wealthy residents, companies, properties and investment portfolios. Their approximate values can be estimated and a potential tax yield calculated. Existing wealth therefore makes an unusually tempting tax base.

The future flow is harder to put into a budget spreadsheet. Who will start the next successful company? Where will it be incorporated? Where will venture capital cluster? Where will a founder reinvest the proceeds from a first successful business? Which country will a highly mobile specialist choose?

A country can extract more revenue from its existing stock of wealth while making itself less attractive as a place to create the next stock. Over a single fiscal year that effect may be invisible. Over several decades it could be much more important.

Small European economies cannot ignore this. Sweden is not a closed system. People can move relatively easily, financial capital crosses borders with little friction, and companies can choose among several broadly comparable European jurisdictions. That does not mean the country with the lowest tax rate automatically wins. Sweden offers skilled workers, infrastructure, institutions, education and social stability, all partly financed by taxation. Location decisions depend on the whole package.

Tax differences still have a price. The difficult empirical question is how large it is.

When SEK 1 of revenue costs more than SEK 1

The marginal cost of public funds is an attempt to capture this problem: how costly is it to obtain another unit of government revenue after behavioural distortions are included? Combining their migration estimate with the earlier Danish intensive-margin evidence, the researchers calculate an MCPF of roughly 4.2 for the Scandinavian wealth taxes they study.

That does not mean the government literally destroys SEK 4.20 whenever it collects SEK 1. Nor is 4.2 a universal constant that can simply be applied to any future billionaire tax. The paper stresses substantial uncertainty around the estimate, and the calculation depends on the tax systems studied, enforcement, available methods of avoidance and evasion, and assumptions about other economic effects associated with very large fortunes.

It does reinforce a point developed in an earlier discussion of where a tax burden ultimately settles: the statutory tax bill is only the beginning of the economic calculation. People change behaviour, and those changes determine both how much revenue survives and who eventually bears the wider cost.

Efficiency is not the whole welfare calculation. Redistribution has value. A krona transferred from someone with billions in assets to a child growing up in severe poverty can generate more social value than leaving that krona with the billionaire. Public investment, education and infrastructure can raise future productivity. The researchers themselves note that some programmes aimed at disadvantaged children have estimated social returns high enough to exceed their calculated cost of raising wealth-tax revenue.

The same standard must apply to the spending side. Saying that a billionaire tax will finance welfare does not complete the analysis any more than saying that a billionaire will move abroad does. If much of the mechanical revenue disappears through behavioural responses, the expenditure eventually financed by the tax needs to be valuable enough to justify the cost of raising it.

The wrong test of a tax

There is a defensible case for Swedish exit taxation aimed narrowly at gains accumulated during Swedish residence. Vänsterpartiet's current motion proposes exit taxation for individuals, while Sweden already has a ten-year rule that can preserve taxing rights over certain gains after emigration. As of 11 August 2026, Finance Minister Elisabeth Svantesson said possible reform of that rule was still being considered within the Government Offices.

Closing an obvious opportunity to move immediately before realising large gains is not economically irrational. What would be harder to justify is treating an exit tax as proof that Sweden no longer needs to worry about the mobility of capital and entrepreneurs. It closes one route of adjustment while leaving many others open, and it can change when people make their location decisions.

The best case against higher wealth taxation is therefore not that every wealthy person will leave. They will not. Nor does the Scandinavian evidence show that wealth taxation inevitably reduces revenue; it shows the opposite. And the measured effects do not support the idea that every departing entrepreneur takes all associated jobs and investment along.

The harder problem is that behaviour can turn a large mechanical tax base into a much smaller realised one, while some of the most consequential long-term responses are precisely those least visible in the data: founders who move before becoming wealthy, businesses established somewhere else and investors who never arrive.

None of this proves that Vänsterpartiet's proposed billionaire tax would fail. Sweden has not implemented such a tax, and its eventual design would determine much of its economic effect. What it changes is the standard by which success should be judged.

The relevant calculation is not simply how much wealth exists in Sweden today and what percentage of it the state could collect. It is how much revenue remains after people adjust, what happens to other tax bases and productive activity, and whether the public value created with the money is large enough to justify those costs.

A government can raise more tax revenue and still leave the country worse off. It can also impose genuine economic costs and still make society better off if the revenue is used well enough. For a country competing for mobile people, companies and capital, the demanding question is whether the tax system leaves it capable of creating the next generation of wealth worth taxing.