The nonpartisan Congressional Budget Office doesn't have a good record of predicting what the government's finances will look like a decade from now. No one does, really. But every year, the agency publishes a forecast of the government's books for the coming decade. Policies can change and assumptions can be wrong, but, for better or worse, it's the most detailed estimate we have to chart the course of the nation's debt.
The CBO's latest forecast was released Tuesday. Those lacking a social life can read the whole thing here, but I've provided some highlights.
Debt held by the public as a share of GDP is forecast to peak in 2014, decline for a few years, and by 2023 be about where it was a decade before:
As I wrote last month, this flattening off of the national debt -- contrary to earlier forecasts of runaway growth -- is thanks to three factors: stronger economic growth, the 2011 budget deal to impose a spending sequestration, and the recent fiscal cliff deal to raise tax rates on those earning more than $400,000 per year.
Adding some detail to spending and taxes, here's the path we're on:
One thing that sticks out is that growth in so-called mandatory spending -- entitlements such as Social Security and Medicare -- is a full 6% a year, or faster than the rate of economic growth. Meanwhile, the all-else category of "discretionary" spending inches along at 1% annual growth, which is less than the rate of inflation, let alone economic growth. The federal government is becoming a giant pension fund with a few declining services on the side. By 2023, discretionary spending as a share of GDP is forecast to be one-third lower than it is today, and by far the lowest it's been in half a century.
Here's another way to look at it:
Even with lower discretionary spending, entitlement growth is large enough that total government spending as a percent of GDP is forecast to be above the historic average for the next decade. After 2015, tax revenue as a percent of GDP could be above the historic norm, too:
Now some of the finer details.
Interest rates at all-time lows have saved the government hundreds of billions of dollars in interest costs. Since 1995 we tripled the national debt without paying a penny more in annual interest. But interest rates will eventually rise, and when they do, the interest bill on the national debt will balloon.
The CBO's forecast assumes annual interest costs will nearly quadruple over the next decade, from $223 billion this year to $857 billion by 2023. That assumes short-term interest rates will rise from 0.1% today to 4% by 2017, with the yield on 10-year bonds rising to 5%:
This is manageable. But what if these assumptions are wrong? What if interest rates rise to 6%, or 9% -- or 15%, as they did in the 1980s? The CBO gives some context (emphasis mine):
If interest rates on all types of Treasury securities were 1 percentage point higher or lower each year from 2014 through 2023 and all other economic variables were unchanged, cumulative outlays projected for the 10-year period would be about $1.1 trillion higher or lower (excluding the additional costs of servicing the federal debt).
One silver lining here is that higher-than-expected interest rates would probably be accompanied by higher inflation or faster economic growth, which would bring in more tax revenue than the CBO projects, offsetting part of the burden.
But that goes the other way: If economic growth is slower than the CBO projects, tax revenue falls and the deficit rises. And slower-than-projected growth seems like an inevitability, as one sharp-eyed analyst noted. "CBO budget forecasts assume 13 straight quarters of 3%+ growth," political analyst Sean Trende Tweeted on Tuesday. "Happened 2x since 1947. Haven't had 13 Qs at 3%+ TOTAL since 90s."
The CBO details how much tax revenue could fall if its growth estimates don't pan out (emphasis mine):
If the growth of real GDP and taxable incomes were 0.1 percentage point lower per year than in CBO's baseline projections, revenues would be lower than in the baseline projections by roughly $275 billion over the 2014-2023 period.
For perspective, $275 billion is about double what the federal government spends annually on education and training. Even slightly slower-than-expected growth can cripple a long-term forecast.
But we often forget that bad forecasts can also lead to better-than-expected outcomes, when spending is lower than originally estimated. That's exactly what's happening now with medical costs, as the CBO notes (emphasis mine):
In recent years, health care spending has grown much more slowly both nationally and for federal programs than historical rates would have indicated. (For example, in 2012, federal spending for Medicare and Medicaid was about 5 percent below the amount that CBO had projected in March 2010.) In response to that slowdown, over the past several years, CBO has made a series of downward technical adjustments to its projections of spending for Medicaid and Medicare. From the March 2010 baseline to the current baseline, such technical revisions have lowered estimates of federal spending for the two programs in 2020 by about $200 billion -- by $126 billion for Medicare and by $78 billion for Medicaid, or by roughly 15 percent for each program.
This is huge, because effectively all of the long-term projected budget deficit is tied to the assumption that health-care costs will balloon. If that assumption turns out to be incorrect -- and it already looks like that's the case -- everything about the budget debate changes.
"It's difficult to make predictions, especially about the future," Yogi Berra once said. Repeat those words every time you look at a budget forecast.