Two first class economists have written a paper showing that economic theory can be used to justify high personal income tax rates. They settle on a number in the 70% marginal tax rate as reasonable.
Their main argument is premised on the social welfare function.
In taking my Masters degree examination at UC Berkeley (I failed twice, even though my essay showed far greater knowledge of economics than any of the boobs reading it) I used the social welfare function, with all the mathematical details, to argue that the postal service should be sold. The Berkeley extreme leftist students grading my paper didn’t like what I wrote. The US government didn't spin the postal service off until four years later. No connection that I know of to my twice written essay.
A social welfare function is nothing more than the sum total of all the individual utility functions in the society. An individual has declining utility for anything. The more the individual has the less he'll pay for more. That applies to food, sex and dancing. Since the marginal utility of an additional anything is presumably declining it is easy to see how the social welfare function could be used to justify very high progressive income taxes.
I simply want my readers to know that there are arguments for high progressive taxes.
Since I consider the population of people who make a positive contribution to society by starting their own businesses and introducing significant innovations to be very small, I really don't care too much about taxation. As I have said in an earlier blog, just stay out of these peoples way and their drive for independence and creativity will be unleashed.
However, as with states in which it is obvious that these creative businesspeople are willing to move to more favorable commercial environments, it is reasonable to assume the same is true of nations. Individuals who are creative businesspeople, though a tiny number, may be influenced by the business climate and the opportunity in some other country to make a fortune.
We cannot know how true this is and therefore an argument of the form presented in this article has no testable validity.