VAT taxes can end up being progressive?

The Economist Free Exchange blog has a brief discussion on consumption/VAT taxes:

...[A]n OECD study... finds, contrary to perceptions, that America has
one of the world's most progressive tax structures.
Even before the rise in
top marginal rates promised by Mr Obama, the US income tax collects 45 per cent
of its revenues from the highest-income decile. Compare that with Britain at 39
per cent, Canada at 36 per cent, France at 28 per cent, Sweden at 27 per cent
and an OECD average of 32 per cent.
That is because consumption taxes (VAT)
in Europe undermine the progressivity of income taxes. There exists a healthy
strand of the public finance literature suggesting that consumption taxes are
more efficient than income taxes. They can encourage more saving which leads to
more investment. Mr Crook believes that rather than steepening the tax schedule
America should consider a VAT.
Actually, this needn't even undermine
progressive taxation; consumption taxes can be implemented in a manner which
taxes higher earners (or spenders) more. Glenn Hubbard and
Bill Gentry have found that broad consumption taxes can be progressive, for
instance.

Someone writing in the Nikkei's Economics Classroom column (not online, and I neglected to save the copy...) argued in line with that last argument that raising the consumption tax could end up being progressive. In particular they estimated that a higher Japanese consumption tax with the revenue used to fund social programs would end up being progressive as the poor would spend less yet use more social services, resulting in a net progressive redistribution. In their estimate, the middle class earners were squeezed the hardest as their net taxation/service usage was about even, even compared to the rich. Compared with the deeper analysis of the 48 page Hubbard and Gentry article, the Nikkei article seems kind of oversimplified.