In 2006, the mortgage interest deduction cost the US $76 billion. Although it affects a broad population, most of the benefit went to the moderately or conspicuously rich. Half of the benefit went to 12% of the taxpayers, those making more than $100,000 per year. If it was eliminated, house prices would fall probably 10-15%. This article says it promoted the then-housing bubble.
Bowles-Simpson proposed replacing the current deduction with a 12% tax credit (so you don't have to itemize to benefit, since usually only the wealthy itemize). A 15% credit proposed by a GW Bush panel would have produced $388 billion from 2013 to 2019, i.e., about $65 billion/year additional revenue.
A CBO analysis points out that the capital gains tax includes a tax on any change in value due to inflation, which is not real income, but is similar to any tax on interest which does not account for inflation. On the other hand, capital gains tax is not imposed until the item is sold, which may delay taxes for years. Capital gains over time have produced between 4-7% of revenues for individuals, although they were over 10% for the latter half of the 1990s. Changes in tax rates affect behavior, but usually for a short time, a few years. If capital gains taxes are going up, people will sell assets sooner to be taxed at the lower rate, but once they are sold, the spike in selling is over. In general it is hard to predict capital gains revenues.
The conservative Heritage Foundation position is that raising capital gains will stifle the economy. It implies that rich entrepreneurs will not work if they have to pay the same taxes as plumbers or engineers. They just won't get out of bed in the morning.
A Wall Street Journal article has some specific numbers for capital gains tax receipts in fairly recent years. In 2003, receipts were $51.3 billion. In 2007 they were $137.1 billion. A rough estimate is that if these rich people (and they are almost all rich) paid at the regular tax level (35%) rather than the current capital gains level (15%) the receipts would roughly double, i.e., to $100 billion in 2003 and $250 billion in 2007. This is very rough, because rich people hold some assets for a long time, and only sell them when the capital gains tax is relatively low. If there were no special capital gains tax, sales of assets would smooth out; with they special, lower tax they tend to bunch up either just before the rate goes up, or just after it comes down. But it seems like you could estimate that the lower capital gains rate cost the US treasury about $100 billion per year during the first decade of the 2000s, or about $1 trillion over the last 10 years.
A Wall Street Journal article has some specific numbers for capital gains tax receipts in fairly recent years. In 2003, receipts were $51.3 billion. In 2007 they were $137.1 billion. A rough estimate is that if these rich people (and they are almost all rich) paid at the regular tax level (35%) rather than the current capital gains level (15%) the receipts would roughly double, i.e., to $100 billion in 2003 and $250 billion in 2007. This is very rough, because rich people hold some assets for a long time, and only sell them when the capital gains tax is relatively low. If there were no special capital gains tax, sales of assets would smooth out; with they special, lower tax they tend to bunch up either just before the rate goes up, or just after it comes down. But it seems like you could estimate that the lower capital gains rate cost the US treasury about $100 billion per year during the first decade of the 2000s, or about $1 trillion over the last 10 years.