Analysis of Fiscal Therapy: Conventional and Dynamic Estimates

By John Ricco, Rich Prisinzano, and Sophie Shin

In his new book, Fiscal Therapy, William Gale argues that the long-run trajectory of the US federal debt is unsustainable. He also offers a comprehensive tax and spending reform proposal with the aim of fixing this problem. He writes:

“Taken together, the proposals would put debt on a stable course by reducing the growth of spending and raising taxes. In the absence of reform, the federal government’s debt will rise inexorably and unsustainably…

Under the proposals, debt would be stabilized at a sustainable level….The plan would boost economic growth. By controlling the debt, it would release huge amounts of capital for private investment. Corporate tax changes, as outlined below, would boost business investment.”

In this post, we analyze these claims using the PWBM platform. We find that Gale’s proposal would reduce federal debt substantially – even beyond his estimates – and increase long-run economic growth.

The proposal

Gale’s comprehensive policy package results in a net increase in taxes and a net reduction in spending. However, taken individually, some policies cut taxes and some increase spending. Fiscal Therapy’s chapters are organized by policy area. Table 1 lists these chapters and their respective policies, showing how we modeled each provision.

Table 1: Provisions in Fiscal Therapy by Chapter

Chapter Provision Full model Modified estimate
Taxing People Tax capital gains at death X
Raise the top two capital gains rates to 20 percent and 24.2 percent X
Tax like-kind exchanges X
Treat carried interest as ordinary income X
Eliminate $10 billion of various loopholes X
Phase-out mortgage interest deduction X
Create a homeowner credit X
Restore pre-TCJA income tax rates and brackets X
Increase IRS funding to bolster enforcement X
Convert the estate tax to an inheritance tax X
Taxing Business Raise the corporate rate to 25% X
Allow full expensing of all investment X
Eliminate the net interest deduction on new debt X
Close the Gingrich-Edwards loophole X
Eliminate the pass-through deduction X
Tax like-kind exchanges X
Eliminate FDII provisions X
Saving Social Security Raise the retirement age X
Change the benefit formula X
Add new bend point X
Raise minimum benefit level X
Raise survivor benefits X
Reinstate benefits for college-aged children X
Limit spousal benefits for high income households X
Increase taxation of benefits for high income households X
Use chained CPI-U X
Gradually raise payroll tax rate to 6.7% X
Raise payroll tax cap X
Change treatment of government workers X
Investing In People Expand the Child Tax Credit X
Expand the Earned Income Tax Credit X
Double federal TANF funding X
Double SNAP funding X
Enact universal pre-K X
Expand tax credits for education X
Enact paid family leave X
Taxing Consumption Phase in a 10% value added tax with a rebate X
Improving The Environment Phase in a carbon tax with a rebate X
Investing For Growth And Security Increase infrastructure spending X
Increase R&D spending X
Increase defense spending X
Reforming Health Care Reinstate the ACA's individual mandate penalty X
Approriate the ACA's cost sharing reduction payments X
Fix the ACA's "family glitch" X
Create a public option in the insurance marketplace X
Transition to a bundled payment system for Medicare X
Allow Medicare to have the same negotiation power for prescriptions as the VA X
Introduce a premium support system to Medicare X
Limit the tax expenditure for employer-sponsored insurance X
Expand Medicaid X
Increase and standardize the excise tax on alcoholic beverages X

Note: For "modified estimate" items PWBM applies PWBM's macroeconomic forecast to existing budget estimates (for example from the Joint Committee on Taxation (JCT) or the Congressional Budget Office (CBO).

“Full model” means Penn Wharton Budget Model (PWBM) directly analyzes a policy using the PWBM platform. For “modified estimate” items we apply PWBM’s macroeconomic forecast to existing budget estimates (for example from the Joint Committee on Taxation (JCT) or the Congressional Budget Office (CBO), but do not include growth effects except from changes in government debt. For example, Fiscal Therapy includes proposals to increase certain health and education spending. These policies could conceivably increase labor productivity in ways other than through government debt. As the empirical support for a more direct impact other than debt becomes more established in the literature, PWBM will include these direct effects in future analysis.

All of Gale’s proposals are assumed to take effect in calendar year 2021.

Budget effects

We find that the proposals outlined in Fiscal Therapy would dramatically reduce debt held by the public. On a conventional scoring basis, we estimate enacting these policies would cause the debt-to-GDP ratio to fall from 188 percent to 17 percent in 2050. Factoring in macroeconomic feedback effects, those numbers are 197 percent and 15 percent, respectively.

Gale projects the budget effects of every policy in the book, taking estimates from a wide variety of sources. Table 2 compares PWBM’s budget estimates with those presented in the book.

Table 2: Revenue, Outlays and Debt-to-GDP Ratio Under Current Policy and Fiscal Therapy Policy in 2050

Estimate Current Policy Fiscal Therapy Policy Difference
Revenue Conventional 20.2% 27.9% 7.7%
Dynamic 21.0% 27.8% 6.9%
Fiscal Therapy 18.4% 23.6% 5.2%
Noninterest outlays Conventional 22.3% 23.3% 1.0%
Dynamic 22.6% 23.3% 0.6%
Fiscal Therapy 21.3% 21.6% 0.3%
Interest outlays Conventional 7.6% 0.9% -6.7%
Dynamic 7.9% 0.8% -7.1%
Fiscal Therapy 6.7% 2.2% -4.5%
Debt-to-GDP Ratio Conventional 187.9% 17.4% -170.5%
Dynamic 196.6% 15.3% -181.3%
Fiscal Therapy 179.0% 57.0% -122.0%

Note: Consistent with our previous dynamic analysis and the empirical evidence, the dynamic projections above assume that the U.S. economy is 40 percent open and 60 percent closed. Specifically, 40 percent of new government debt is purchased by foreigners.

While our baseline debt-to-GDP ratios are similar, they are compositionally different. For individual income tax revenue projections, PWBM uses a current law baseline. Gale, however, assumes that TCJA’s provisions are extended rather than expiring as scheduled at the end of 2025. As a result, PWBM has a higher revenues-to-GDP ratio under baseline.

PWBM also projects higher interest rates under baseline. This fact helps explain why our estimate of long-run debt reduction is larger than what is presented in Fiscal Therapy. A greater share of projected debt is due to interest outlays, so the marginal dollar of primary deficit reduction does more to reduce debt in the future, all else equal.

Economic effects

PWBM projects that this package would increase capital, labor, and overall output. However, consumption would fall somewhat due to the value added tax and to a lesser extent the carbon tax. Table 3 presents our estimates of changes in key macroeconomic aggregates in 2050 relative to current policy, including those under alternative assumptions about the openness of the economy.

Table 3: Fiscal Therapy Policy Effects on Key Macroeconomic Variables Relative to Current Policy in 2050 - Partially Open, 0 Percent Open and 100 Percent Open

Partially open 0% Open 100% Open
GDP 7.3% 23.1% -0.7%
Capital services 20.6% 75.1% -2.7%
Hours worked 1.0% 3.2% -0.1%
Labor income 7.3% 23.1% -0.7%
Consumption -5.1% -0.2% -7.9%

Note: Consistent with our previous dynamic analysis and the empirical evidence, the Partially Open projections above assume that the U.S. economy is 40 percent open and 60 percent closed. Specifically, 40 percent of new government debt is purchased by foreigners.

Under PWBM’s default openness parameter (40 percent open), long-run output increases by 7 percent, driven primarily by an increase in capital services (21 percent). The intuition is that by reducing the long-term fiscal imbalance, crowd-out of private capital is lower under the proposal. Full expensing also increases the incentive to invest, building up the capital stock over time.

Under alternative assumptions about international capital flows, the story changes. We model two extremes: a scenario with 1-for-1 crowd-out, and a scenario with no crowd-out. Assuming a fully closed economy -- the scenario where crowd-out is the highest -- the policy package has the biggest economic impact.

Conversely, if we assume a fully open economy, thereby turning off the channel for government debt to affect growth, then the long-term effects turn slightly negative. This directional change is due to higher taxes on labor and domestic savings.