At our sister blog, Henry writes about the relation between welfare policy and tax rates. I have no thoughts on the substance of the matter, but it reminded me of something I’ve wondered for awhile. Aren’t the terms “progressive” and “regressive” for taxation are a bit loaded? I’d think it would be better to use a value-neutral term such as “graduated tax rates” or “redistributive taxation.”
In a word, no. "Progressive" and "regressive" when used with regard to taxation are exemplars of the perfect match between denotation and connotation. The terms denote specific technical meanings; replacing them would merely obfuscate the truth. At the same time, the connotations of the two words perfectly reflect the position of tax policy on the political continuum, at least in the U.S. Why on earth would we want to change that?
Why change it? Because, first of all, it fails to communicate. Many people who hear "progessive tax" must interpret it to mean a tax designed to promote whatever goals they think are currently considered by many to be "progressive", which may not be a tax with rates increasing with income. And since those "progressive" goals change, any alignment is temporary. Most people today wouldn't call eugenics "progessive", for instance.
But I think Andrew's point is that the term, by accident or design, appears like an attempt to win an argument by just defining the other side's position as bad. Sort of like the disingenuous terms "pro-life" and "pro-choice". (In case anyone doesn't see that, I'm sure that just about every "pro-life" person kills and eats carrots, and that just about every "pro-choice" person denies that people should have the choice of killing and eating homeless people.)
"Graduated tax rates" does not tell us whether the rates are increasing with respect to income or decreasing.
How is "redistributive" not a heavily loaded term?
First of all, it implies just as much about how taxes are spent as how they are collected. ALL taxation is redistributive, but if you mean "taking from the rich and giving to the poor", it's not clear at all that's what happens in a progressive tax scheme: the lion's share of federal taxes are spent on the military, social security and interest on the debt, none of which is tilted toward the poor.
The term also implies that the _purpose_ of a progressive tax scheme is to redistribute wealth, which I think most proponents of progressive scheme would take strong issue with. The logic behind progressive taxation is not that of Robin Hood.
I also agree that calling it "graduated" doesn't help because it doesn't say whether the tax goes up or down with income.
But isn't this all academic? The overall federal tax burden is much more than the income tax — payroll and capital gains taxes heavily skew the overall tax rate away from the federal income tax schedule, to say nothing of the heavily regressive effects of sales taxes. In terms of his personal federal return, Warren Buffett is famously taxed at a lower effective rate than his secretary.
I skeptical that you can make this debate more neutral with a couple of carefully chosen adjectives.
Sure the terms are loaded. But which way?
Liberals think that "liberal" and "progressive" are good words.
Conservatives (especially lately) think that they are both bad words.
This is probably a great example of having your cake and eating it too; the terms are loaded just they way that people think they should be!
They are technical terms and people in the business understand that. The problem is when people outside the business see these terms they may misunderstand. Its our (economists) job to communicate.Its natural in economics to use everyday words (like efficiency, utility etc) in a technical sense. The alternative is to use something convoluted. Then we are accused of using jargon, being obscure etc. You can't win!
Statistics does the same (terms like efficiency, consistency, bias etc) but since statistical debates seldom enter the public arena the same problem does not arise.
Brandon Berg argues that a consumption tax is just as progressive as an income tax here:
http://distributedrepublic.net/archives/2006/12/0…
Public finance specialists usually prefer income elastic and income inelastic to progressive and regressive. Where log T = a + b (log Y) + e, taxes are revenue elastic where b > 1, and inelastic where