Update: turns out Ryan is in fact using the CBO's exact GDP projections. See the update at the end of the post.
And he's been guilty of this sort of thing in the past. As Annie Lowrey pointed out at the time, his 2012 budget relied on a Heritage Foundation analysis which implied that, if the budget passed, unemployment would be 6.4 percent in 2012 and 2.8 percent in 2021. Unemployment hasn't been that low (2.8 percent) since the Korean war and it's doubtful the Fed would ever let it get that low again, for fear of inflation.
So is the fiscal 2014 iteration of the budget problematic in this way? To be sure, some of the growth numbers in the piece are fishy. The Center for American Progress's Michael Linden notes that Ryan cites a CBO report as saying that a debt plan the size of Ryan's would grow the economy by 1.7 percent by 2023, even though that same report suggests that his plan would cut growth by 0.6 points next year.
In three years — 2017, 2019, and 2021 —Ryan's numbers are higher than the CBO's. He exceeds the CBO growth rate by about 1.4 points in 2017, by 0.3 points in 2019 and by 0.5 points in 2021. But on average, Ryan's real GDP numbers are only 0.2 points higher than the CBO numbers.
That's something, but it's not a real game changer. To see why, here's how the deficit looks as a percentage of GDP under Ryan's plan using both the CBO economic numbers and Ryan's economic numbers**:
Update: Though Ryan's budget is unclear on this point, Marc Goldwein of the Committee for a Responsible Federal Budget emails in to say that the Ryan economic numbers are, in fact, identical to the CBO ones. The above differences are entirely due to rounding error. So the budget isn't just honest about the growth figures, it's literally by-the-book.
Ryan's office notes that he's using CBO-proposed multipliers for dynamic scoring in the second appendix, under which we'll actually have a surplus by 2023. But even using static scoring, the budget is balanced (just barely) by then.
* Methodology note: Ryan's budget gives outlays, revenue and deficit numbers in both nominal dollar terms and as a percent of GDP. If you divide the former by the latter, you get his assumed nominal GDP numbers, and if you compare year to year, you get growth rates. That said, because his budget doesn't include 2013, the 2014 growth rate can't be computed, so I left that year off.
From there, I reduced the numbers by the CBO's year-by-year inflation projections to get Ryan's real projections. The CBO puts out year-by-year real GDP projections, so to get their nominal GDP projections I added the inflation projections to the real GDP projections.
** Methodology note II: The CBO doesn't put out nominal GDP projections (update - they do, I just didn't find them in time). But they do have (a) nominal GDP in 2005 (b) inflation rates from 2005 to 2023 (c) real GDP growth rates from 2005 to 2023. Using (b) and (c), you can get nominal GDP growth rates, and using that and (a) you can get nominal GDP from 2005 to 2023, and from there you can get what the Ryan deficits as a percent of GDP would be using the CBO's numbers.