I don't think with the [1] graph that use a LINEAR LOG SCALE for the Y axis we can talk about a LINEAR growth of GDP per capita as you do. It's more like an EX

I don't think with the [1] graph that use a LINEAR LOG SCALE for the Y axis we can talk about a LINEAR growth of GDP per capita as you do. It's more like an EXPONENTIAL increase!! Also, if you look at the trend, the recent decline is totally unprecedented - except maybe just before the great depression (the small cut over 1914) and during the great depression. Considering the decline happened even with a very strong stimulus - and was NOT fixed by the stimulus, I guess we have many reasons to be worried about this slowdown in the pace of growth (and technology). A tax change may not usually be a good idea - just like a keyneysian stimulus looked weird and was even initially though to be damaging during the great depression. Desperate times call for desperate measures. We are not there yet, but at least having a refreshing new proposal (regressive tax based on the duration investment spending is kept) is IMHO valuable, if only to have another tool we can use besides the traditional "progressive tax+redistribution to increase consumption". BTW the Australian example you give is quite interesting. May I ask you for some links ?