In other words you want not to support their incomes, but certain aspects (those that you think should be encouraged more than others) of their consumption.

In the same vein, why not give a bigger EITC to people who buy larger houses? Or bigger cars?

This might help supporting the prices of housing and the jobs of car workers, and both of these are things that seem to be important to many voters.

As for paying for it, there’s a great deal of evidence that consistent financial well-being at the lowest and medium-lowest levels of society–reduced inequality–results in superior economic performance (i.e. growth in GDP per capita). Think: sustained aggregate demand.

There’s also a lot of evidence that when that well-being and relative equality do not exist, we risk meltdowns like 1929 and 2008. Viz:

http://www.asymptosis.com/robert-livingston-once-again-sheds-serious-light-on-our-current-situation-as-illuminated-by-the-great-depressioncondensed.html

Economists on all sides agree that government policies which spur growth (and prevent meltdowns) do the best job of paying for themselves. I’m not sure if you’re planning to address this kind of “dynamic scoring” approach, but it would have merit here.