Friday, April 19, 2013

Luck and taxes

Consider what makes a movie a blockbuster. Many factors contribute to a film's success: quality of the writing, acting, directing, marketing, etc., as well as the public's response to these inputs. Not all of these are within the control of any one person, including the film's producer. If a film is unusually successful, chances are that it is due to a combination of factors, including some that are not predictable or controllable. One way to see this is to note that sequels to very successful films are, on average, more successful than typical films but less successful than their predecessors, because it is generally impossible to replicate everything that made the original film successful. (It's also possible that a sequel is produced just to milk the success of the original, without any intent of comparable quality, or that a sequel happens to be better than the original; thus the qualification of "on average" above.) This is a well-known statistical phenomenon: regression to the mean. Observations at the extreme of a distribution tend to get there for lots of reasons, including some randomness. Successive observations tend to be closer to the mean of the distribution. For another example, the sons of extremely tall fathers tend to be taller than average but shorter than their fathers, whereas the sons of very short fathers tend to be shorter than average but taller than their fathers.

Now consider income distribution. Someone with a very high income probably got there for a number of reasons, not all of which were under the individual's control: i.e., such a person probably worked hard, made good choices, etc., but also got lucky in some respects. Similarly, someone in the very low end of the income distribution probably got there partly as a result of individual action (or inaction) but partly as a result of happenstance. It's also possible that one man becomes rich even though external factors are against him, and another man becomes poor despite having every advantage in life; but on average, those in the extremes of the income distribution are there because of a variety of factors pointing in the same direction. This to me is the most compelling justification for progressive income taxes. Life is not fair, and the tax system (as well as some of the ways in which the revenue is spent) mitigates the inequities. The justification is even stronger when we consider that some advantages or disadvantages are systematic but still outside an individual's control. Note also that the tax system is not coming anywhere close to equalizing incomes, because there are still plenty of factors contributing to income that would generally be regarded as fair and should not be discouraged.

Of course there are other rationales, including the idea that those who earn higher incomes are benefiting more from our whole economic infrastructure, and therefore should contribute more toward its maintenance; and this one from Mark Thoma.

No comments:

Post a Comment