> Since the notion of income for the very rich is not very well defined, it should not be expressed as a function of income but of wealth.

> Our proposal is to have a minimum tax equal to 2% of the wealth of billionaires each year.

I think this is pretty reasonable.

To me the more interesting part is a structure for tax-avoidance where he says each country should be able to look at a company's total global tax bill, compute a "deficit" of how much more tax they would have paid if that country's tax rate applied to its global business, and pro-rate based on how much actually occurs in that country ... which presumably then updates their "deficit" calculation.

> There can always be countries that play the role of tax collectors of last resort. If Spain decides that it wants multinationals to pay at least 25%, it can calculate the tax deficit of each multinational that has access to its market, defined as the difference between what they are taxed globally and what they would have to pay if they were subject to a minimum tax of 25% in each country where they operate. Some will pay at least 25% everywhere and will not have an extra tax in Spain, but for those that do not [pay]? Spain could collect a portion of its fiscal deficit. For example, if Apple has a global fiscal deficit of 10 billion and does 10% of its global sales there, Spain could say, “If you want to continue to have access to our market, you must pay 10% of 10 billion, an additional €1 billion in taxes.” The logic is very simple: you have to link market access to minimum taxation.

Am I wrong, or does this mean that if multiple countries took such a policy, calculating taxes then requires a fixed-point process to converge on a value?