In “Miles Kimball, the Supply-Side Liberal,” Noah Smith summarizes my post “Can Taxes Raise GDP?” with this passage:

Many people are familiar with the fact that people work about the same  amount whether taxes are low (as in the 2000s) or high (as in the 1960s). Miles agrees with this, but points out that high taxes hurt people in a different way, by making them feel so poor that they have to work more, and thus depriving them of leisure.

This makes it clear to me that I didn’t get some of my point across in “Can Taxes Raise GDP?” since what Noah is saying only applies to the case when the government is wasting the money.  What if the government isn't wasting the money?  Let’s take the case of the government redistributing money from the rich to the poor (which is a big element of what is going on in Medicaid and Medicare, for example).  The first thing that happens, as Karl Smith emphasizes in “Welcome Miles Kimball: Now Set the Record Straight on Taxes” is that the poor receiving the money work less:

benefits lead people to work less, a point I have long tried to make. 

This is a big effect.  And if it means that fathers and mothers don’t have to work second jobs to make ends meet and can spend more time with their children, this is a good thing, even though it makes GDP go down.  (One of the many problems with using GDP as if it were the measure of how well a society is doing is that child-rearing by parents doesn’t get counted in GDP.)

So far so good, but what about the side effects of taking the money from the rich to give to the poor?  The “Occupy Wall Street” movement and its spinoffs paint a picture of rich people who got rich by some kind of chicanery or evil–a picture embedded in the negative connotation now carried by the phrase “the 1%.”  But the typical case is someone who got rich–or whose parents got rich–by providing a useful service.  And I am not just talking about the wonderful services provided by someone like Steve Jobs or Oprah Winfrey. For vividness I need to use famous people as examples, but before their current fame, in the relative anonymity that most rich people live in, Duchess Kate’s parents rose to become part of the “the 1%” in England by providing people with party supplies and decorations.   There were many parties that were better because Carole and Michael Middleton provided these supplies and decorations.  Aside from the rare cases (think Bernie Madoff) in which someone gets rich by pretending to provide a service they are not really providing, or by means of some other evil–we all benefit from having the rich work hard to provide us with those services. Also, many of the rich provide jobs to employees.  

Now let’s look at the bad side-effects of tax distortions. To do this, we need to compare distortionary taxes (such as income taxes, labor earnings taxes or consumption taxes) to taxes on the rich that do not depend on the level of their income, earnings or consumption.  If we could tax the rich in a way that did not depend on their income, earnings or consumption, they would fulfill their obligation to contribute toward helping the poor in part by working harder to make more money to pay the taxes.  This would benefit all of the people who receive the services they provide.   And they might hire additional employees to work with them in the extra time they are devoting to work.  

By contrast, distortionary taxes such as income, earnings or consumption taxes push the rich toward getting the money to pay their taxes out of their existing income rather than in part by raising their before-tax income to soften some of the hit to their after-tax income.  We are all then worse off as a result: the rich because they would rather work harder than take the tax hit entirely as a reduction in after-tax income, and everyone else who would have benefitted as a customer or employee from the extra services and jobs they would have provided.  

So why don’t we use non-distortionary taxes that don’t depend on income, earnings or consumption to raise the taxes to help the poor?  It is because it is not easy to tell the rich and the poor apart without looking at income, earnings or consumption.  

In principle, if we could look at genes, say, to figure out who is inherently blessed with the talents to be rich, we could avoid distortions.  This approach raises a host of troubling philosophical issues.  But it is taken seriously in the academic literature.  See for example my University of Michigan colleague Joel Slemrod’s paper with Kyle Logue: Genes as Tags: The Tax Implications of Widely Available Genetic Information.  At this point, I am not willing to recommend such a gene-based approach.  But without some new approach, we are stuck with identifying who is rich by looking at income, earnings or consumption.  So we use taxes that depend on income, earnings or consumption, with all their attendant distortions. 

My overall message in this post is given well by Noah’s summary of my first post “What is a Supply-Side Liberal”:

Taxes bad, redistribution good.  

Some messages are worth repeating.