i dont see the logic behind any argument that would replace data on income with data on consumption rates/levels--but i would have no problem with datasets that juxtaposed the two--the difference between the lines you might draw would be a nice image of credit and it's role--and that seems to me one level of what this question turns on, really: what to do and how to think about the pervasive role played by consumer credit in driving expenditures.
there's a second problem:
the relation of these expenditure levels to the problems of structural inequality in the distribution of wealth is different, it seems to me: IF one were to replace data about income with data about consumption levels, it would have the effect of minimizing the appearance of economic inequality. but everyone would know--at least for a while--until they forgot about it--that nothing particular has actually changed about the distribution of wealth except what is now being used to index it.
on this, the question seems to me not to be whether you imagine inequalities in the distribution of wealth to be a problem or not, but rather what statistics are to do, what they measure and why they measure it--and whether it is a good idea to be cavalier about switching indices in order to generate or reinforce ideological biais or for political advantage (think about the reagan redefinition of inflation rates by excluding from them what causes inflation rates to rise...what good has it done, beyond enabling reagan to say he "did something" about inflation)---if you think statistics are an extension of politics, then you'll land in one place on this--if you think that it's a good idea for policy-makers to have something approaching an accurate picture of the socio-economic realities they are supposed to administer/interact with, even if that picture poses problems--then you'd land in another.
==========
as a counter to the edito quoted in the op, here's an editorial from this morning's ny times by robert reich.
make of it what you will in general, but it sure raises problems for the op edito:
http://www.nytimes.com/2008/02/13/opinion/13reich.html