Matthew Yglesias writes
In an excellent column, Stan Collender makes the point that it does no good to talk about cutting spending in pure numerical terms. If you don't spell out which actual things you want the government to do less of then you're not really doing anything. If we have a "hiring freeze," for example, what tasks currently undertaken by civil servants are going to go undone?
My colleagues Michael Ettlinger and Michael Linden try to take this on in a new report they call "A Thousand Cuts." It starts by looking at the quantity of deficit reduction that would be needed by 2015 in order to achieve the goal of "primary balance" (i.e., receipts equal non-interest spending) by that year. And it spells out a few different scenarios. One entails doing 33% of the reduction through spending cuts, one entails doing it with 50% cuts, one entails 75% cuts, one entails 100% through cuts, and one features 100% cuts and doesn't count tax expenditures as spending. The report is both a guide to specifically which programs are relatively low value and also to how extraordinarily painful the reductions would have to be to do this all or exclusively on the spending side.
This is, of course, equally true on the other side: people don't want to pay more taxes for all the services they say they want, and which will indisputably require more taxes to provide. It turns out that when Americans say they want spending cut, what they really mean is the trivial sum we spend on foreign aid; it also turns out that Americans actually don't know much about the structure of the US tax code. Certainly, during the Clinton-era tax levels, a large majority thought taxes were too high, and poll questions about what fair tax rates would be tend to produce answers that don't line up very well with our actual tax code.
I'm not sure what this proves except that Americans have a very fuzzy conception of how the government raises and spends money, and that poll questions asked in isolation (asking people whether they support programs, without informing them about relative costs) gives you responses that aren't very useful.
As we await the pronouncements of the budget commission, and more importantly, the reaction of the policymakers who set it up, it's not enough to merely say that Social Security and Medicare poll well. Low taxes also poll very well. What's going to matter is the strength of the relative preferences.
I'd guess that when push comes to shove, Americans prefer lower taxes to higher spending (though to be sure, I'm basing this on the fact that they punish politicians who raise taxes, and since politicians virtually never cut spending, we don't really know how people would react).
However, the folks who prefer higher spending tend to be VERY motivated, because their entire incomes often depend on it. I don't know how that will shake out in the end. But I am pretty sure that vaguely worded poll questions are not enough to protect programs from cutting.