What's the Number?

Matt Miller has a new column in Fortune, which I don't usually read, so I'm a little late on this, but it takes up a topic that we're likely to hear more about in the future: Just what percentage of GDP do we need in taxes, over the long term, to pay for what we expect from government?

The column is available to the un-Fortuneate here.

Miller puts the issue in rather starkly political terms:

The challenge for Democrats, and for anyone else seeking a coherent view of political economy in the era ahead, is this: How do we propose to make the health and pension programs for seniors sustainable while also paying for needed nonelderly initiatives? And how do we do all that while keeping overall taxes as a share of GDP at levels that don't hurt economic growth (without pushing taxes beyond levels Americans are likely to support)?

"I don't think that conversation has yet taken place in the heads of most Democratic economists," says Robert Litan, a former Clinton aide who heads up policy research at the Kauffman Foundation. Many who urge tax hikes to fix Social Security assume we can also raise taxes later to achieve those other goals. But they've never laid out a macro budget that adds this all up. Still, there has to be number locked inside their heads that captures how high taxes as a share of GDP should go to cover the full progressive monty. Think of it as the Democrats' secret number.

So what is it? Triangulating from chats with economists, plus a glimpse of a coming book edited by Isabel Sawhill and Alice Rivlin of Brookings that will (finally) frame these issues explicitly, I'd say the number is somewhere around 28% of GDP.

Well! Federal taxes today are at 16.3 % of GDP, historical lows not seen since the 1950s. At the end of Clinton's tenure, taxes reached nearly 21% of GDP, a historical high. As the boomers retire, everyone knows taxes are going up. Newt Gingrich recently told me so himself. But 28% of GDP? Yikes.

Revenues as a percentage of gross domestic product is a very strange, abstract concept to base policy around. Outside of the extremes (Kim Jong Il=100%/Grover Norquist, foraging for nuts and berries=0%), it's hard to know how to argue for one alternative over another. I found myself in a debate the other day which degenerated into shouting out percentages, and finally settling on X.5%. (The X is priveleged insider information, but it's not "28," trust me.) It's a very weird experience -- what are you really discussing?

But we're going to be discussing it a lot. It's the sum of all the promises we've made and all their implications. Matt's is a very good way to put it. And to see how far Democrats are from acknowledging it, consider, for example, that the Center for American Progress, where Miller himself hangs his hat, recently put out a tax plan, which pointed with pride to the fact that it would raise revenues to only 17.2% of GDP on average over the next ten years. (Miller's time horizon is presumably much longer, though.)

I think it's a little unfair of Miller to call this "the Democrats' secret number." It's the number of anyone who doesn't want to cut existing programs and who believes that eventually we will have to deal with health care. What's the Republican secret number? Who knows? And what's the gap between the Republican secret number on spending and the Republican secret number on what they think revenues should be?

In short, what Miller is saying is that even if we didn't need to recover the 5% of GDP lost to the Bush tax cuts, even if we stop the giveaway of trillions more, even if we dealt in some reasonable way with the modest long-term Social Security funding problems, the years ahead are going to be a long ugly slog of compromises and cuts. That's exactly what reasonable economists were warning about in 2000, and we should have heeded the warning then.