Design Matters When Raising Taxes to Reduce the Deficit and Stabilize the Debt

May 16, 2023 6:41 pm Published by Comments Off on Design Matters When Raising Taxes to Reduce the Deficit and Stabilize the Debt

More voices are entering the fray to call out the problem of growing U.S. deficits, which are on track to reach nearly 7 percent of gross domestic product (GDP) by the decade’s end. Rather than continue down the path of growing debt, lawmakers should craft a comprehensive solution. International experience cautions against tax-based fiscal consolidations, but modest tax increases may be part of a successful debt reduction package. In this blog post, we explore how to design tax increases to stabilize the debt. The next blog post in our debt and deficits series considers how to design spending cuts.

Design is important because raising a dollar of revenue through different taxes has different effects on the economy. A more harmful tax increase can shrink the economy, yielding less revenue from other taxes. Harm the economy too much and the solution may prove counter-productive, reducing the likelihood of successful debt stabilization.

These points bear out in the economic literature. Tax-based deficit reductions tend to have a more negative impact on the economy and less successful track record than spending-based ones. The difference is primarily due to the response of private investment, as business confidence falls to a greater degree and for a much longer duration after tax-based plans.

Overall, successful fiscal adjustments primarily cut spending and include modest tax increases. A rough guideline for an expenditure-based plan is for at least 60 percent of its savings to come from spending cuts and 40 percent or less from revenues. Applied to the U.S. context, stabilizing debt at its current share of GDP would require about $8 trillion in 10-year savings, with a $3 trillion limit on tax increases.

Research from the International Monetary Fund (IMF) and the Organisation for Economic Co-operation and Development (OECD) emphasizes avoiding economically damaging taxes in fiscal consolidation packages. Raising corporate or personal income taxes is a no-go. Instead, less harmful tax types, like consumption or environmental taxes, and base broadening efforts should be in the mix.

Tax Foundation similarly concludes that not all taxes are created equal. We can categorize taxes in a hierarchy of most to least harmful, largely based on how responsive an activity is to taxation. Corporate income taxes are the most harmful because capital is highly mobile. Individual income taxes come next, followed by relatively less harmful payroll and consumption taxes.

The table and chart below illustrate how modest tax increases with minimal harm to the economy can contribute to debt stabilization.

Compare increasing the gas tax by $0.35 and inflation-indexing it, broadening the individual income tax base by eliminating the exclusion for employer-sponsored health insurance (ESI), or raising the top individual income tax rate to 50 percent above $400,000 for single filers and $450,000 for joint filers and the corporate tax rate to 28 percent.

Illustrating the Trade-Offs of Different Options to Reduce the Deficit
 Raise the Gas Tax by $0.35Eliminate the Exclusion for Employer-Sponsored Health Insurance28% CIT and 50% Top Individual Rate above $400K/$450K
Long-Run GDP-0.2%-0.5%-1.3%
Conventional Revenue, 2024-2033 (Billions)$931$3,523$2,872
Dynamic Revenue, 2024-2033 (Billions)$797$3,117$2,151
Dynamic Improvement in Debt-to-GDP Ratio, 20332 percentage points8 percentage points5 percentage points
Dynamic Improvement in Debt-to-GDP Ratio, Long Run4 percentage points18 percentage points9 percentage points

Source: Tax Foundation General Equilibrium Model, May 2023.

The base broadening and excise tax options both result in relatively minor economic trade-offs for the revenue raised, while the economic trade-offs of raising marginal income tax rates are much harsher.

As a result, a less economically harmful option like eliminating the ESI exclusion results in a much more powerful reduction in long-run debt as a share of GDP. However, even with substantially higher tax revenues, long-run debt-to-GDP would still continue to grow, showing that spending needs to be the primary focus to successfully reduce debt.

Design Matters When Raising Taxes to Reduce the Deficit and Stabilize the Debt-to-GDP ratio US debt taxes deficits

That brings us to the options currently on the table. Unfortunately, President Biden’s plan to reduce the deficit, as described in his budget, depends entirely on net revenue increases from raising economically damaging taxes—an approach inconsistent with successful efforts to reduce debt.

The Biden administration’s estimates of nearly $3 trillion in deficit reduction under the budget are highly uncertain, particularly as they depend on a novel set of tax increases. On a conventional basis, Tax Foundation estimates the President’s budget plan would reduce the 10-year deficit by $2.5 trillion. Because the plan would reduce GDP by 1.3 percent, the deficit reduction drops to $1.9 trillion over 10 years on a dynamic basis.

By 2033, publicly held debt as a share of GDP will reach 118 percent under the baseline. Biden’s budget would reduce it to 112 percent conventionally or 114 percent dynamically. In the long run (by 2053), the plan would reduce debt-to-GDP from 195 percent under the baseline to 174 percent conventionally or 180 percent dynamically.

The smaller improvement on a dynamic basis highlights the importance of minimizing the economic costs of tax hikes and instead seeking efficient sources of revenue. For example, in contrast to Biden’s proposals, Tax Foundation’s Tax Reform Plan for Growth and Opportunity proposes replacing the current corporate income tax with a distributed profits tax and the current individual income tax with a much broader based flat tax, and reforming estate and capital gains taxes at death. Base broadeners in the plan help offset the costs of the reforms, including eliminating most tax expenditures: the exclusion for employer-sponsored health insurance, all itemized deductions, and many tax credits.

The plan is approximately revenue neutral, but because it raises revenue more efficiently, it increases long-run GDP by about 2.4 percent and reduces long-run debt-to-GDP by about 5 percentage points on a dynamic basis.

In lieu of a fundamental overhaul of the tax system, lawmakers may consider permanence for all or part of the expiring Tax Cuts and Jobs Act (TCJA) provisions. An across-the-board extension of the individual income tax expirations would be pro-growth but would significantly reduce revenue. Alternatively, lawmakers could build on the TCJA’s reforms to further broaden the base by, for example, using the options outlined in Tax Foundation’s Plan for Growth and Opportunity and curtailing new green energy tax credits.

Other reforms, such as permanent expensing for capital investments and research & development, would also enhance the efficiency of the tax system. On a conventional and dynamic basis, such changes would reduce revenue in the short term, but, over the long run, the fading revenue cost and permanent economic benefit would be enough to slightly reduce deficits and the debt-to-GDP ratio.

Lawmakers have a tough task ahead as they consider solutions to stabilize the deficit and debt. Though it may not be politically popular to raise the gas tax or broaden the tax base, to the extent that a comprehensive deficit reduction package includes modest tax increases, such options are the most advisable on economic grounds. Ultimately, a better-designed tax system should be a goal of any fiscal consolidation package.

Note: This blog post is the third in a series in which our experts explore the issues and potential solutions for America’s growing debt and deficits.

Categorised in:

This post was written by prismatax

Comments are closed here.