Josh Barro's column bashing the so-called "Buffett Rule" proposal makes an excellent (but largely separate) post about the political economy of failing to index capital gains taxes to inflation:
Let’s say that you bought an asset last year for $100 and sold it today for $104, but there was 2 percent inflation. Your real gain was only $2, but you paid taxes of 60 cents (15 percent of $4), for a real tax rate of 30 percent.
Whether you think 15 percent or 30 percent is the correct rate, the point is that this winds up giving asset owners a strong reason to fear inflation. Absent this distortion, a rich guy sitting on a pile of shares should be indifferent to between 2 percent real growth plus 2 percent inflation and 2 percent real growth plus 3 percent inflation. But under the system of taxing capital gains on the basis of nominal increases in value, an additional percentage point of inflation amounts to a tax increase. And as we all know, rich people are pretty good at lobbying against tax increases. This is probably not an explicit element of Federal Reserve Open Market Committee members' thinking as they continue to plot a course of unduly tight money, but the objective interests of wealthy people tend to have a large influence on policy and this is very plausibly a causal factor in driving tight money.