>I knew there was some criticism of SimCity's simplified tax model in the U.S. some years ago (a simple inverse relationship between tax rates and business growth)
Isn't that pretty much accurate? I'm not aware of any serious theory postulating that raising taxes will have any non-negative effect on economic growth in and of itself.
Where it breaks down is once you account for what the tax dollars buy. If government takes $5000 from you but provides $6000-equivalent worth of health services (which it may be able to do through economies of scale or reduced transaction costs or solving a collective action problem etc.) then you have a net gain of $1000 rather than a net loss of $5000.
Imagine bifurcating the result: In one scenario, your country is the beneficiary of the efforts of a large international charity, you get $6000/citizen worth of health services without collecting any tax dollars from your population. Second, your country collects $5000 in taxes from each citizen and then sets the money on fire, or gifts the money to an international banking cartel in exchange for nothing whatsoever. Does anyone care to argue that the second could somehow be preferable to the first?
Which brings it back to the conservative position, which is that governments through incompetence or corruption will tend to convert $5000 of tax revenue to something less than $5000 worth of services to the population, resulting in a net loss to the economy and the country's people.
What so happens to be true as a factual matter is that a) governments can strike the balance in either direction depending on how well they're managed, but b) that the larger a government is, the more likely it is to allocate resources inefficiently, because as size increases economies of scale reach diminishing returns while the attention of legislators and their ability to perform due diligence is spread ever thinner over an expanding scope of government projects, leaving more opportunities for mistakes and corruption.
I always thought it would be an interesting experiment to eliminate 90% of federal spending but not lower federal taxes whatsoever, just give the money to the states whose residents paid the taxes, with no strings attached (i.e. the feds serving only to enforce uniform taxation and inhibit jurisdiction shopping), and letting the states individually decide how to spend the money. You can imagine a significant reduction in corruption: California residents would no longer be paying a pro rata share of the cost of boondoggles in Massachusetts (or vice versa), and Massachusetts residents may decide they don't want them either if they have to pay the entire cost out of their own pockets.