Support for taxing wealth rather than income has surged in American political discourse. Progressive lawmakers have floated federal wealth tax proposals for years, and the idea has now migrated to the state level in dramatic fashion. In California, the 2026 Billionaire Tax Act would impose a one-time five percent levy on the worldwide net worth of anyone with a fortune exceeding one billion dollars, a measure its backers claim will raise as much as one hundred billion dollars for state coffers. The political appeal is obvious. Wealth taxes promise to extract revenue from those least likely to feel the pinch while sidestepping the political toxicity of raising taxes on ordinary earners.
The trouble is that wealth taxes have already been tried, at length and in earnest, and the record is not encouraging. Sweden ran a wealth tax for nearly a century before abandoning it. On the other hand, Colombia has cycled its wealth tax on and off for decades. Both cases offer the same lesson: wealth taxes are administratively porous, chronically underperform their revenue projections, and drive the capital they are meant to tax out of reach. California’s proposal is poised to repeat that pattern, and the evidence suggests it may do so faster and more severely than either historical precedent.
Sweden introduced its modern wealth tax in 1911 and did not repeal it until 2007. Across that span, the tax underwent constant revision: new brackets, new reduction rules, new valuation schemes for corporate equity, each one grafted on to patch a problem the last reform had created. Despite nearly a hundred years of tinkering, revenue from the tax never rose to anything resembling significance. Aggregate wealth tax revenue never exceeded 0.4 percent of Swedish GDP in any year, and for most of the tax’s final three decades it hovered between 0.5 and 1 percent of total tax revenue. A tax that consumed enormous administrative energy for a century delivered a rounding error’s worth of public financing.
The reason was not a lack of political will to tax the wealthy. Sweden raised its rates repeatedly, and marginal wealth tax rates on large firms peaked near 2.5 percent in the early 1970s. But every rate increase was matched by a corresponding erosion of the base. Politicians themselves built in the mechanisms of escape: reduction rules that capped wealth tax liability relative to income, valuation discounts for unlisted business equity, and exemptions for land, forestry holdings, art, and controlling stakes in listed firms. When Sweden lifted its foreign exchange controls in 1989, capital flight accelerated sharply.
Swedish tax authorities eventually estimated that more than 500 billion kronor in assets had been illicitly moved offshore, with an equivalent sum held by wealthy Swedish expatriates living abroad entirely. Between 2001 and 2006 alone, aggregate Swedish household wealth grew by 60 percent while wealth tax revenue fell from 8.4 billion kronor to 4.8 billion kronor, a decline that finally persuaded the government that reform was hopeless and abolition was the only remaining option.
Colombia’s experience, drawn from decades of administrative tax data, demonstrates that the Swedish pattern was not a fluke of one Nordic welfare state but a structural feature of wealth taxation itself. Colombian taxpayers approaching a wealth tax threshold responded almost immediately by reporting less wealth, and taxpayers benefiting from rate cuts just as quickly reported more, a real-time demonstration of how sensitive declared wealth is to the tax rate applied to it. Across the various rate regimes Colombia has cycled through since 2003, researchers found that up to one-fifth of expected revenue was lost simply because taxpayers reduced what they reported.
More strikingly, the damage did not disappear when the tax did. Taxpayers who had learned to underreport their wealth during a taxed period kept reporting lower wealth for years after the tax expired, apparently out of a rational fear that the tax might return. Assets that third parties could not independently verify, such as stakes in closely held businesses, were manipulated far more freely than fixed assets like real estate, and the wealthiest Colombians increasingly moved their holdings into hard-to-trace offshore entities. Cross-referencing Colombian tax records against the Panama Papers showed that the introduction and escalation of the wealth tax coincided with a surge in Colombians establishing offshore structures, and that those who did so subsequently reported markedly less wealth to domestic authorities. A tax intended to capture wealth instead taught the wealthy how to make it disappear.
California’s proposed Billionaire Tax Act was projected by its proponents to raise 100 billion dollars, built on the assumption that only 10 percent of the taxable base would be lost to avoidance. Even before the measure has reached a ballot, that assumption looks untenable. Publicly-confirmed departures, including Larry Page, Sergey Brin, and Peter Thiel, had already removed $536.4 billion—almost 30 percent of the state’s billionaire wealth base—before the residency snapshot date used to calculate the tax. Expanding that count to include unconfirmed but publicly reported departures, such as Mark Zuckerberg’s relocation to Florida, pushes the lost share of the tax base above 41 percent. Applying migration elasticities drawn from the peer-reviewed literature on Swiss wealth taxation implies an even larger response, with more than half the tax base potentially exiting the state.
The revised revenue estimate—roughly $40 billion rather than $100 billion—is only the first half of the problem. California’s billionaires currently contribute an estimated $3.3 to $5.8 billion a year in state income taxes. If a meaningful share of them leave permanently, the state loses that income stream indefinitely, and the present value of that permanent loss can exceed the one-time wealth tax revenue collected. Modeling 100,000 combinations of plausible revenue, departure, and discount rate assumptions found that 71 percent of scenarios produce a net fiscal loss for the state, averaging roughly $24.7 billion. Sweden took the better part of a century to arrive at a wealth tax whose costs plainly outweighed its benefits. California’s version threatens to reach a similar verdict before a single dollar has been collected.
There is also a structural dimension worth stressing. Unlike Sweden’s tax, which was legislated and could in principle be repealed by ordinary statute, California’s measure is a constitutional amendment that permanently lifts the state’s existing cap on taxes on intangible property, with no sunset clause. Once that cap is gone, nothing prevents future ballot initiatives from imposing recurring wealth taxes at higher rates, transforming what is marketed as a one-time emergency measure into permanent constitutional infrastructure for wealth taxation.
Supporters of wealth taxation are not without a rebuttal. They point out that Sweden’s failures were partly self-inflicted, the product of politically-negotiated exemptions and reduction rules rather than any inherent flaw in taxing wealth, and that a tax with a broader base and fewer carve-outs might avoid the same fate. They also note that other researchers argue wealth taxes have not historically generated the scale of migration response critics predict, and that revenue lost to reported avoidance in Colombia still left a functioning, if diminished, tax.
These are serious arguments, and any honest accounting of wealth taxation should weigh them. But the pattern across two very different countries and nearly a century of combined experience, low revenue relative to political cost, aggressive base erosion, and capital flight that persists even after the tax is gone, is difficult to dismiss as coincidence. California is now poised to test that pattern once more, this time in real time and in public view.
