In which I take a comment thread debate at Angrybear and front page it here mainly to check my html coding (haloscan is much less idiot proof user friendly than blogger).

[snip]
Since 1990 tax federal receipts as a percent of GDP have averaged around 18.9 percent. During the Bush years this number fell below 18 percent and during the Clinton years it went over 20 percent. Looking at this number going backwards in time I think setting a target rate at 19 percent seems reasonable.
[snip]
Aaron | 07.24.08 - 8:52 am |

This 19% of GNP number, it seems to be pure true conservativism. The average in the past must be optimal, because ... it is in the past. I think decisions about optimal G/Y should be made based on uhm you know costs and benefits. In particular, Baumol's disease implies that the relative price of health care and education is increasing compared to the price of goods. I think that this means that optimal G/Y is increasing. Also an aging population means that TR/Y will naturally increase. I note that the vast majority of our fellow citizens understand that and want it to happen They want to make the social security trust fund actuarily sound by increasing taxes on rich people and not by cutting benefits

"Currently, people pay Social Security taxes only on the first $90,000 of their annual income. If it were necessary to keep the Social Security program paying benefits as it does now, would you favor or oppose increasing the amount of income that is subject to Social Security taxes?"

6/10-15/05
Favor 63
Oppose 30
Unsure 7

, that is, they agree with Obama (except the majority supports a much more extreme reform). The argument that people don't know what they want, because they don't know that the US has changed since 1950 is absurd. The argument that people shouldn't get what they want, because they want higher taxes on the rich (search for upper-income) is undemocratic. The claim that economists know that the people are wrong is not based on the data.