Capital is a key determinant of growth, along with labor and technology. Yet, the U.S. tax system systematically penalizes saving, which finances capital accumulation. The tax penalty on saving could be eliminated by moving to progressive consumption taxation.

The Tax Burden on Saving

Income taxation is inherently biased against saving, which arises when individuals consume later than they work. Under income taxation, individuals who save part or all of their earnings pay tax on the return on their saving, in addition to the tax already paid on the earnings. This double taxation imposes higher percentage tax burdens on those who save than those who immediately consume their earnings.

Wage taxation avoids the double tax and the resulting saving penalty because only the original earnings are taxed, with no tax on the return to saving. Consumption taxation also avoids the double tax and the saving penalty because the saver is not taxed on the amount that is saved and is taxed only on the future consumption ultimately financed by the saving.

It is sometimes argued that the current U.S. tax system taxes savings lightly and therefore imposes only modest saving penalties. A common claim holds that income from saving is taxed at a special maximum rate of 20 percent, well below the 39.6 percent maximum tax rate that applies to wages. Also, capital gains are not taxed until realization. Moreover, retirement saving and some other types of saving can be done through tax-preferred accounts.

But the actual tax treatment of saving is quite different, as documented by Carroll and Viard (2012, 14–17) and Viard (2013, 3–6). The 20 percent maximum tax rate applies only to two types of income from saving, long-term capital gains and qualified dividends Full ordinary income tax rates apply to other income from saving, including the profits of S corporations and non-corporate businesses, interest income, short-term capital gains, and nonqualified dividends. Also, since 2013, most income from saving by high-income households has been subject to a special 3.8 percent “unearned income Medicare contribution” tax, over and above regular income taxes. Although capital gains are not taxed until realization, they are also not indexed for inflation, so investors must pay tax on nominal gains that merely offset the rise in the price level. And, tax-preferred savings accounts feature complexity and restrictive rules that reduce participation. Due to contribution limits, the accounts also offer no marginal incentives for those making the largest contributions.

The profits of C corporations are also subject to a separate corporate income tax, featuring a 35 percent statutory tax rate. The much-vaunted tax savings that corporate shareholders reap from the 20 percent maximum tax rate on their dividends and capital gains provide only a partial offset for the burden of the corporate income tax. And estate and gift taxes also apply to the savings of the nation’s wealthiest households. State and local tax systems impose further taxes on saving, including individual and corporate income taxes, property taxes, and sales taxes on business purchases of capital goods.

If reforms are not undertaken, the tax burden on saving may increase in upcoming years. The Obama administration has proposed an increase in estate and gift taxes and a “Buffet rule” that would effectively raise the individual income tax rate on some high-income households’ long-term capital gains and qualified dividends. More sweeping proposals are likely to arise from the fiscal pressures posed by the rise in Medicare, Medicaid, and Social Security spending. Viard (2013, 7–8) and Cowen (2013) called attention to calls for an annual wealth tax, a tax that gained added support when subsequently advocated by Piketty (2014, 515–534).

Progressive Consumption Taxation

As mentioned above, consumption taxation does not penalize saving. Replacing income taxation with consumption taxation would therefore be beneficial. One approach, which may well be adopted at some point, would replace part of the income tax system with a value added tax (VAT). That approach raises a number of concerns, which are not discussed here.

The greatest gains could be achieved by replacing the entire income tax system (the individual and corporate income taxes, the unearned income Medicare contribution, and the estate and gift tax) with a consumption tax. For distributional reasons, though, the consumption tax would need to be progressive, rather than a regressive tax like the VAT.

One progressive consumption tax is the personal expenditure tax, sometimes called a “consumed income” tax. This tax starts with a framework similar to the individual income tax, but with a crucial modification that strips out the income tax’s saving penalty. Households would report their income on annual tax returns, but would then claim a full deduction for all saving and add back in any dissaving. What’s left would be the household’s before-tax consumption for the year. That consumption amount would be taxed at graduated rates, with higher tax brackets for households with higher consumption. Exemptions and refundable tax credits would be provided to low-consumption households.

Under a personal expenditure tax, households would deduct deposits in savings accounts, the costs of purchasing stocks and other investments, any loans that they made, and any payments (interest or principal) that they made on their debts. Households would pay tax on withdrawals from savings accounts, the full proceeds from selling stocks and other investments, any payments (interest or principal) received from loans that they made, and the full amount of any borrowing that they did. Seidman (1997) provides a thorough analysis of the personal expenditure tax.

Another progressive consumption tax is the Bradford X tax, which is a graduated-rate variant of the Hall-Rabushka flat tax. The X tax and the flat tax start with a framework similar to the VAT, with a crucial modification that removes the VAT’s regressivity. The VAT tax base would be split into two. Households would be taxed only on their wages and business firms would be taxed on value added minus their wage payments, so that the total tax base would be the same as the VAT. The household tax would have graduated rates, with higher tax brackets for households with higher wages. Exemptions and refundable credits would be provided to low-wage households. The business tax would have a flat tax rate, equal to the top wage tax bracket.

Although the household tax might look like the individual income tax and the business income tax might look like a corporate income tax, they would actually be quite different in economic terms, with no vestige of those taxes’ savings penalties. The household tax would be imposed only on wages, thereby exempting interest, dividends, capital gains, and all other income from saving. The business tax (which would apply to both corporate and non-corporate businesses) would be imposed on business cash flow, not income. The key difference is that investments would be immediately deducted, as they are under a VAT, rather than depreciated. As under a VAT, there would be no tax penalty on a new break-even investment. Carroll and Viard (2012) provide a thorough analysis of the X tax.

Neither the personal expenditure tax nor the X tax is familiar to the general public and neither is likely to be enacted in the near future. The personal expenditure tax may draw criticism because it does not include a tax collected at the business level and because individuals are taxed on the proceeds of borrowing. The X tax may be criticized because it does not look like a consumption tax and because it does not collect tax at the household level from people who are consuming the proceeds of past investments. And, each tax would have to address a variety of challenges, including transition issues, the treatment of financial institutions, and international issues. Nevertheless, each tax would allow revenue to be raised in a progressive manner without penalizing saving.

Conclusion

The rise in the federal government’s revenue needs is likely to prompt efforts to increase the already-high tax burden on saving in upcoming decades. This threat to growth can be averted by moving to progressive consumption taxation.

References

Carroll, Robert, and Alan D. Viard. 2012. Progressive Consumption Taxation: The X Tax Revisited. Washington, D.C.: AEI Press.

Cowen, Tyler. 2013. Wealth Taxes: A Future Battleground. New York Times, July 20.

Pikkety, Thomas. 2014. Capital in the Twenty-First Century. Cambridge, MA: Harvard University Press.

Seidman, Laurence A. 1997. The USA Tax: A Progressive Consumption Tax. Cambridge, MA: MIT Press.

Viard, Alan D. 2013. Capital Income Taxation: Reframing the Debate. AEI Economic Perspectives, July.


The opinions expressed here are solely those of the author and do not necessarily reflect the views of the Cato Institute. This essay was prepared as part of a special Cato online forum on reviving economic growth.