Monday, April 16, 2007
Fuzzy Math Returns!
Jon Chait has a low tolerance for morons. In this case, former Bush press flack Ari Fleischer, who is seemingly unaware that 60 > 40, 31 > 30, and 37 > 31. Such are the mathematical contortions one needs to make in order to argue that our tax system actually fleeces the rich. I'd chide the Wall Street Journal for publishing such tripe, except that after their infamous "lucky duckies" editorial (the ducks, in the WSJ's opinion, are the impoverished Americans who don't pay taxes), I don't think they have any standards I could possibly appeal to.
Subscribe to:
Post Comments (Atom)
7 comments:
Dave,
I agree that Ari left out payroll taxes which drastically misstates what he is trying to accomplish. He should have looked at total taxes paid rather than just federal income taxes.
However, if you take the top x%(where x can probably be anything from 1-10, but I haven't checked how robust it is) of the population and make their share of GDP the independent variable and taxes as a share of total taxes paid as a dependent variable. The hypothesis of the writer you quote is that the slope of the line should be less than one which would mean that as the share of the rich has increased, the amount of taxes they have paid has decreased. Unfortunately, the slope is greater than one which indicates that the the share of taxes paid by the highest income individuals has increased relative to their share of income.
John Hall
Hall,
What do you mean by "share of GDP" and "share of income"? I don't think these are equivalent terms.
GDP is one measure of national income, you could use real disposable income if you prefer, but GDP is a suitable statistic for this kind of comparison.
How is GDP a measure of "national income," and in particular an equivalent of the income taxed by the IRS? GDP refers to *where* the money was made, i.e. in the U.S. I think you mean GNP, which refers to *who* makes the money, i.e. entities taxable by the IRS. I'm failing International Tax, but I did learn that Americans are taxed on their worldwide income, whereas reciprocal tax treaties allow foreign companies and foreign citizens who are taxed by their own governments to escape most U.S. taxes.
But even GNP isn't a good measure. For example, if I am an associate in a law firm, my labor is charged at $300/hour, but I am compensated at less than $100/hour. Meanwhile, a frequently-absent partner in the firm, because he is a partner, gets a share of the profits generated by my labor even if he has done nothing to help produce the service as counted in GDP/ GNP. Obviously at these wages, we're both in the current top tax bracket; it's just an example to demonstrate that how a good or service is priced in the market is not reflective of how much the person who produced the good or service is paid. Since GDP and GNP both depend on the market price of goods and services produced, they don't have much relationship to individuals' income. Personally, I'm inclined to tax the hell out in areas where people aren't producing goods or services. That the capital gains tax rate is lower than most federal income tax rates is ridiculous. The rhetoric of "double taxation" should be reversed -- let those who are doing the work to produce goods and services be taxed less, and those who profit from the workers' activity be taxed more.
pg,
The point of looking at shares instead of income and taxes as a percent of national income is to correct for the residuals not having a constant variance when you look at absolute levels. (the technical term for this in econometrics is heteroskedasticity) The point is to put each time period relative to each other and GDP is one of a multitude of national income statistics that you could use to do this.
Do I think GDP is the perfect statistic to measure national income? No. But that's not the point weighting them. Furthermore, the difference between GDP and GNP is so small for the U.S. that it wouldn't even matter for what I was doing. And I just want to clarify that you are aware that GDP and GNP only measure the final value of goods and service, the total value added.
Your example of the partner's income being reflected in GDP doesn't make any sense complaining about national income because every statistic that an economist would refer to as a national income statistic would include that as income. Does not the partner create a service by helping to bring in new clients and retain old ones? Is that not important? Even still, let's say the general partnership spend 33% of the amount they bill the client on overhead, 33% they give to you, and keep the remainder for themselves. All of that money shows up in national income statistics and all of that money is eventually taxed. What exactly is the problem? If you make 100 dollars an hour with 50 weeks a year and let's say you work your ass off for 75 hours a week. That's an income of 375k. Explain exactly why using a national income statistic that includes the final value of the goods/services you produce is bad because your argument isn't clear.
Finally, as someone who works in finance, I'm glad that you think that our industry doesn't produce any goods or services. What you're basically saying is that a company like Vanguard isn't offering a service by creating mutual funds that people can use to save for retirement. Furthermore, you're arguing that people who save responsibly for retirement using the non-service that Vanguard offers should be taxed extra because they wanted to plan for their retirement. I'm not going to get into an argument about why you're incredibly naive. I just think you are and will leave it at that.
John Hall
Hall,
If you work in finance, I assume you know what the capital gains tax is, and are aware that it taxes the *result* of some of the services provided by the finance industry; it does not tax the *wages* that people working in that industry make. So when I say that income from capital gains and other "unearned income" ought to be taxed at a higher rate than the income that is a product of the taxpayer's own labor, that in no way is saying that the finance industry doesn't produce any goods or services. That you would jump to the conclusion that I am saying that gives the strong impression that you have heard it many times over, and it's now your automatic reaction to any disagreement with your ideas.
Remember that none of your points have anything to do with my criticism about Dave's post. I could just as easily plot absolute levels of the variables on the x and y axis, but I would suffer from heteroskedasticity and the results wouldn't make sense. Dividing each by the same factor didn't change the relationship, but corrected for unequal variances over time. At the point that we're arguing about this stuff, it really does nothing except take time out of my day.
Post a Comment