Now that the 10-year budget problem is (poorly, counterproductively) solved, Republicans are faced with a bit of a problem: How do they keep justifying an agenda based entirely around debt fears now that the debt doesn't look so scary?

The answer they've come up with is to change the timeline so the debt looks scarier again. Rather than judging the budget over the next 10 years, they want to judge it over the next 30. The problem, as Jonathan Chait points out, is that no one knows what will happen to the budget over the next 30 years. Imagine projecting the economy of 2013 in 1983. Even the best model would've missed the Internet revolution and the rise of China and India and the 2008 financial crisis, not to mention the two Iraq wars.

To put it another way, a central insight of conservatism is that the government isn't very good at predicting the future, and then is slow or even unable to respond when its predictions prove incorrect. If you think the government can usefully forecast the economy 30 years into the future, then that should unlock a deep confidence in the merits of central planning.

But I wouldn't advise it. Washington's forecasts aren't very good over five years, and they're only vaguely plausible over 10. Over 20 or 30 years, they're basically useless, as this New York Times graph shows:

Those light blue lines flitting off into nothingness? Those show what Washington thought would happen to the economy. The thick, dark line shows what did happen to the economy. And these errors are off of five- or 10-year forecasts. Thirty-year forecasts would mainly be useful for the lols.

Which isn't to say Republicans are wrong to worry about the long term. We know our demographics are getting worse and our health-care costs are likely to climb (if at a slower rate). But what we need to address aren't 30-year forecasts. They're 30-year policies. And that's what we'd have in a budget deal: Policies like chaining CPI and capping tax deductions and raising Medicare premiums for higher-income retirees all grow in relatively predictable ways over time. They save much more money in years 10-30 than in years 0-10, in fact. If our budget problems gets bigger in 2023, well, so are the polices we passed to deal with it.

But we don't have a budget agreement. Republicans won't agree to new tax revenues and so there's no deal. Instead, we have sequestration, which, according to the law, disappears entirely after 10 years. If you're worried about year 25, in other words, you should hate sequestration.

I'm getting to be a bit of a broken record on this, but repetition is called for: We've "solved" our deficit problem in the worst possible way. But we don't need a useless 30-year budget forecast to know that, and we definitely don't need one to fix it. We just need to look at this graph of the next 10 years:

As you can see, I have helpfully annotated this chart. The deficit falls fast — too fast — in the next few years. That's a headwind against the recovery. Then it rises again through the rest of the decade. So we're making our jobs problem worse while we're supposed to be focusing on jobs, and making our deficit problem worse when we're supposed to be focusing on deficits.

What we need is a deal that supports the economy now and puts in place smart cuts, reforms and tax changes to put us on a sounder footing for later. Instead, we've got a deal that hurts the economy now and probably hurts the economy in the future, too. We don't need a 30-year forecast to tell us that.