Because medicine is very labour intensive, it suffers from baumol's cost disease (
http://en.wikipedia.org/wiki/Baumol's_cost_disease) - so it is reasonable to expect that the cost of providing care should rise in line with gdp. I do not beleive that significant savings can be made without some innovation that will make doctors more productive - automated dignostic machines, or machines that allow surgery to be performed more quickly. We are at least 50 years away from that level of automation. And its not clear how comfortable people would be getting diagnosed by a machine. If you add to this new treatment options, it is inevitable that healthcare costs will grow faster than gdp.
All this talk about higher income tax causing greater gdp growth seems to miss the broader point, which is that the relevant number you want to examine is total governemnt revenue. By this I mean that income brackets on their own are a incomplete measure. Say i reduce income tax but increase co-operation tax in a revenue neutral manner, then I would not expect any growth. The cost of all taxes is ultimately borne by the consumer, and often large changes in the tax code are done in such a way that revenue is only slightly affected. In reality total government revenue as a % of gdp (I.e. the sum of all taxes) grew continuously from WW2 until about 2000 (at 35% gdp), there was a big drop (4-5%) in 2002, and then they went back on an upwards trajectory. Thus the classic interpretation should be that growth fell with time from WW2 until the present. Obviously life is not so simple and the economy tends to have cycles etc. Not to mention sudden burst of growth brought on by a particularly profound technological advancement. It looks to me like that thesis is broadly supported by awm's data. 3.5% gradually decreasing to about 2.5 %, if we make a few adjustments, eg 88-92 is a 4 year period containing a large recession, so it is artificially low, whereas 98-02 would have picked up both teh rebound form the early nineties recession and the dot com bubble (which marked the rise of the internet and like many new technologies it probably caused above average growth when it was first utilised.)
Also, lets not forget that awm's analysis misses out several pertinent points. Eg, what was the bracket for the top rate? If you look at my sources before you will see that the total %gdp paid in income taxes has been remarkably stable between 10 and 15% for over 50 years - if you lower the tax but increase the applicability in a revenue neutral way then you haven't really changed the tax burden at all. It should be clear that when one talks about tax cuts stimulating growth one should only be talking about lowering the total tax burden. The idea that if i tax one group less and another more I can stimulate growth seems flawed to me.
FYI, income tax makes up slightly over a third of total government revenue in the USA. Vat-like taxes make up slightly less than a third, with the other large contribution being "social insurance". I got my figures from
http://www.usgovernm...local=s#usgs302
The physics is theoretical, but the fun is real. - Sheldon Cooper