Trendspotting: The Fiscal Side of the CBO Long-Term Forecast
- According to the latest CBO budget outlook, the pandemic recession will push federal debt and deficit projections onto a much higher future trajectory. The publicly held national debt is expected to reach 104% of GDP by next year and climb to nearly nearly 200% of GDP by 2050. (Tax Foundation)
- NH: Future historians, when writing about US fiscal policy, will likely divide their chronicle into two eras: BC and AC, that is, before and after Covid-19. Yesterday, I wrote about the significant changes CBO has made in this year's long-term outlook in America's future economic performance. (See "CBO Projects Long-Term Slowing of US GDP Growth Rate.") Today we look at the equally spectacular changes in their assessment of our budgetary future.
- What are these changes? Are they plausible? Might they be (gulp!) overly optimistic? Are the CBO's economic and fiscal scenarios mutually consistent? Let's take a look.
- First, the numbers. The immediate jump from 2019 to 2020 in the projected deficit for 2020--from 4% to 16% of GDP--has hardly any precedent in US history. (The only bigger jump was the +15.7% surge occurring from 1942 to 1943.) That big whammo, plus several large post-recession deficits expected in the early 2020s, puts us on an entirely new debt path. And this new path is pushed even higher over time by the CBO's less optimistic economic assumptions. In 2019, the CBO was projecting a debt as 144% of GDP. In 2020, that projection has been ratcheted up to 189%. That's +45 percentage points in just one year.
- Btw, I get this question all the time: How does the CBO calculate deficit and debt? "Deficit" includes all on-budget and off-budget accounts. It is the unified budget balance. The CBO (rightly, IMO) regards "off-budget" balances as an accounting fiction. "Debt," likewise, means publicly-held debt: All debt sold to institutions outside the US government, and this includes the Fed. So, yes, "monetized" debt is included.
- All current spending and tax legislation, moreover, is put on autopilot. For example, all of Trump's 2018 tax cuts that are scheduled to expire do expire. For realism, the CBO ignores the exhaustion of any trust fund, so Social Security and Medicare continue to pay benefits even after the accounting entry stored on some Treasury computer file reads zero.
- Now take a look at the following four charts. The first two simply lay out the debt and deficit paths annually until 2050. In the second, we see that "net interest" due to accumulating deficits is the dominant driver of the accelerating debt burden. But it's not the only driver. The third chart shows the additional impact of growing outlays for Social Security and all federal health spending programs. And the fourth chart divides this growth into increases due to demographic aging and into increases (in healthcare only) due to age-adjusted costs per capita rising faster than GDP per capita.
- If you want to see an especially scary chart, look at the following. By the 2040s, per the CBO, current revenue will not only fail to cover interest outlays. It will not cover anything other than Social Security, healthcare benefits, and other mostly mandatory benefit programs. Virtually everything "discretionary" that government pays for--defense, infrastructure, R&D, civil service, NIH, national parks, you name it--will be funded with borrowed money. Boomer to Millennial: Après moi, le déluge.
- At this point, let me emphasize that the CBO is making a projection, not a prediction. It's a gigantic if-then statement. If current law remains frozen in stone and long-run economic and demographic trends play out according to our best expectations, then this will be the result. The future will almost certainly not unfold this way, but the exercise does clarify some of the policy choices awaiting us.
- Are the numbers plausible? The rising costs due to demographic aging are pretty mechanical and not really open to much fudging. The rising cost burden due higher per-capita healthcare spending is more open to adjustment. I have no doubt that, confronted by the threat of fiscal armageddon, policymakers will ration and rationalize federal health spending--hopefully in a way that will make it less wasteful.
- The biggest cost driver of course is net interest. And here, unfortunately, the CBO may be lowballing the challenge. Note (above) that the CBO expects net interest spending actually to decline as a share of GDP through most of the 2020s before rising rapidly in the 2030s and 2040s.
- Why decline in the 2020s--in the face of a much larger debt-to-GDP ratio? Well, because the CBO expects real interest rates to stay very low for many years to come. How low? Well, consider. From 1990 to 2008, the real interest rate on 10-year Treasury notes (measured as the ex-post historical differential from nominal) averaged 3.1%. Since the GFC, it has averaged 0.8%. Through the 2020s, the CBO expects it to linger at 0.9%, and then slowly rise thereafter until reaching 2.5% in 2050.
- The collapse of CBO's interest rate projections since last year has been dramatic indeed. Only about half of this decline is due to a smaller inflation premium; the other half represents a real decline.
- OK, what scenarios could justify another decade of very low real interest rates even in face of massive additional federal borrowing? I can think of only two. First, a decade of macroeconomic slack and a glacial recovery from the current recession. Or second, further decline in the marginal return on capital, which would be reflected in a further decline in the rate of total factor productivity (TFP) growth.
- The problem is this: The CBO's economic assumptions rule out either scenario. In fact, it assumes a fairly full and rapid recovery from the Recession of 2020. And (as we saw yesterday) it assumes a rise, not a fall, in TFP growth rates in future decades.
- So that's my qualm with the CBO long-term projection. You can't have it both ways. It's OK to assume a paper-thin real interest rates through most of the 2020s--a world in which borrowing is free and there is no such thing as crowding out. But in that case, you also need to posit a decade of relative economic stagnation and adjust your economic assumptions accordingly. Or assume the opposite. A decent macroeconomic recovery along with a healthy real return to new investment. But in that case, you need to posit a steeper and more fiscally dangerous interest rate rebound.
- Maybe there is yet one other scenario. That would be for global investors to be fundamentally mistaken about ex-post interest rates: So in the early 2020s they invest at very low rates, but by the late 2020s they are slaughtered by high and rising rates. With the average Treasury debt duration now at about five years, this whipsaw might save the CBO fiscal projection for the 2020s as a whole. But it could lead to a savage rise in inflation and risk premia in the decades thereafter.
© 2021 Hedgeye Risk Management, LLC. The information contained herein is the property of Hedgeye, which reserves all rights thereto. Redistribution of any part of this information is prohibited without the express written consent of Hedgeye. Hedgeye is not responsible for any errors in or omissions to this information, or for any consequences that may result from the use of this information.