Will California's Wealth Tax Raise $100B or Lose $25B?
Why experts disagree by $125 billion on California's proposed billionaire wealth tax
The California initiative has reportedly collected enough signatures to show up on the ballot in November. This is exciting news if you’re a tax nerd. We will learn a lot over the next six months.
Something we’ve learned already is that experts disagree quite dramatically about the expected revenue the proposal will raise. On the pro side, Galle, Gamage, Saez, and Shanske (GGSS) have estimated the tax will raise around $100B. On the con side, Rauh, Jaros, Kearney, Doran, and Cosso (RJKDC) have estimated the tax will raise only $40B in gross terms. If we account for revenue losses in future years, they calculate a net present value of tax collections that is negative $25B.
These are both serious teams of economists and legal scholars. How can they be so far apart?
What is a wealth tax?
The political narrative tends to treat income and wealth as separate concepts. But they are closely related. That link can help clarify the debate around wealth taxes.
Consider a wealthy person with $100M of financial wealth invested in a mix of bonds and stocks. Assume that wealth generates a rate of return of 5%. That means $5M of income coming in each year.
Consider a 2% annual wealth tax. Just two cents, right? Well, that amounts to $2M of wealth tax liability for this portfolio, or 40 percent of the income flow. In other words, that 2% wealth tax is equivalent to a 40% income tax, levied on top of the income tax already in place.
In California, the federal plus state income tax rate on an investment portfolio like this might be around 40-50% depending on the asset mix. So, a 2% wealth tax would yield the equivalent to an income tax rate of 80-90%. California’s one-time 5% rate could mean a one-time income tax rate of 150% for some!
This is one reason why rich people are so angry about wealth tax proposals.
But California’s wealthy are different
Proponents of the California wealth tax raise an important objection to the logic above. As we’ve written before, many of California’s billionaires don’t look like this example. That’s because most of their wealth comes in the form of stock in listed companies, many of which don’t pay dividends at all.
When those companies reinvest their profits or buy back shares (that the California billionaires haven’t sold), the wealth of these billionaires goes up. But they don’t owe extra tax.1
Unrealized capital gains aren’t taxed, pass-through losses and other deductions shelter other income, and several other tax benefits and avoidance strategies flourish. When realizations do happen, long-term capital gains are taxed at preferential rates or passed on tax-free via step-up of basis at death and other estate tax tricks. When taking advantage of these tax rules, billionaires can face much lower actual baseline tax rates.
In practice, there are many wealthy individuals who pay little personal income tax. Warren Buffet, for example, famously said he pays a lower tax rate than his secretary. Under several assumptions, Balkir, Saez, Yagan, and Zucman (2025) estimate that the Forbes 400 paid around 24% all-in effective tax rates (including corporate income taxes), which is lower than 30% for the full population and 45% for top labor income earners.2
Returning to the wealth vs income tax comparison, for some of these billionaires, the realized returns on their wealth have also been much greater than 5%. An average realized rate of return of 20% or 50% would mean that even a 5% wealth tax is just 10% or 25% of the change in their wealth. However, that doesn’t include the additional income tax the billionaire would have to pay when they paid the wealth tax from liquidated stock.
To pay $5M in wealth tax on $100M in wealth that had approximately zero initial price, the billionaire would owe another 40% or $2M in income tax, making the wealth tax effectively a 7% one-time tax.
Nevertheless, it seems clear that the billionaires can afford to pay the proposed wealth tax. If necessary, they could borrow against their stock to finance the payment and delay the extra capital gains hit.
But will they pay it?
Leaving from California with a tax bill on my mind
Returning to the disagreement in estimated wealth tax collections, what’s driving this $125B gap?
The key is how much avoidance each group assumes. GGSS apply a 10% avoidance estimate to the $2.2T in billionaire wealth. So, if the zero-avoidance revenue estimate was around $110B, they assume $10B would dissipate when the dust settles. They cite as justification the transparency of California’s billionaire wealth in public company stock, the small number of people who can be “easily” audited, and most important, the retroactive design of the tax.
What does 10% mean in terms of actual people? RJKDC report that, out of 212 billionaires they believe might have been subject to tax, the top 5 hold nearly half of the collective wealth of this group.
Put plainly, to wipe out 10% of the base ($218B), you just need one Zuckerberg to flee the tax. Sergey Brin and Larry Page alone account for a quarter of the base (around $500B). According to media reports, Brin and Page have already left. If true, this would wipe out more than double the 10% assumption. And that doesn’t account for avoidance by anyone else in the tax base.
GGSS have argued that the retroactive design captures the wealth of these purported movers. In other words, the rich people may think they moved, but GGSS disagree. Call this the Don Henley provision—you can check out any time you like, but you can never leave.
There are other assumptions driving the remaining difference between estimates. But this avoidance assumption is the most striking and consequential.3
We will have to wait until November to see if California’s voters want to test it.
One may argue that these companies pay corporate tax on their income, and the burden of that tax falls on these billionaires. For example, Meta reported a provision for income taxes of $8.3 billion in 2024. Mark Zuckerberg owned about 13.6% of Meta’s outstanding shares as of late 2024. But if companies raise prices or cut wages in response to such taxes, they will shift some of that burden to consumers and workers.
Note these are average tax rates, which are the ratio of taxes paid to their broad measure of income including unrealized capital gains in the Forbes 400 case. These are not marginal tax rates, which measure how much taxes go up if you earn one more dollar in income, nor are they average tax rates on the IRS definition of taxable income.
For instance, the reason the RJKDC estimate turns negative when they calculate lost income tax revenue in all future years is also their larger avoidance assumption.