The Rich Can't Pay for ObamaCare. By ALAN REYNOLDS
The president intends to squeeze an extra $1.2 trillion over 10 years from a tiny sliver of taxpayers who already pay more than half of all individual taxes. It won't work.
WSJ, Mar 03, 2010
President Barack Obama's new health-care legislation aims to raise $210 billion over 10 years to pay for the extensive new entitlements. How? By slapping a 3.8% "Medicare tax" on interest and rental income, dividends and capital gains of couples earning more than $250,000, or singles with more than $200,000.
The president also hopes to raise $364 billion over 10 years from the same taxpayers by raising the top two tax rates to 36%-39.6% from 33%-35%, plus another $105 billion by raising the tax on dividends and capital gains to 20% from 15%, and another $500 billion by capping and phasing out exemptions and deductions.
Add it up and the government is counting on squeezing an extra $1.2 trillion over 10 years from a tiny sliver of taxpayers who already pay more than half of all individual taxes.
It won't work. It never works.
The maximum tax rate fell to 28% in 1988-90 from 50% in 1986, yet individual income tax receipts rose to 8.3% of GDP in 1989 from 7.9% in 1986. The top tax rate rose to 31% in 1991 and revenue fell to 7.6% of GDP in 1992. The top tax rate was increased to 39.6% in 1993, along with numerous major revenue enhancers such as raising the taxable portion of Social Security to 85% of benefits from 50% for seniors who saved or kept working. Yet individual tax revenues were only 7.8% of GDP in 1993, 8.1% in 1994, and did not get back to the 1989 level until 1995.
Punitive tax rates on high-income individuals do not increase revenue. Successful people are not docile sheep just waiting to be shorn.
From past experience, these are just a few of the ways that taxpayers will react to the Obama administration's tax plans:
• Professionals and companies who currently file under the individual income tax as partnerships, LLCs or Subchapter S corporations would form C-corporations to shelter income, because the corporate tax rate would then be lower with fewer arbitrary limits on deductions for costs of earning income.
• Investors who jumped into dividend-paying stocks after 2003 when the tax rate fell to 15% would dump many of those shares in favor of tax-free municipal bonds if the dividend tax went up to 23.8% as planned.
• Faced with a 23.8% capital gains tax, high-income investors would avoid realizing gains in taxable accounts unless they had offsetting losses.
• Faced with a rapid phase-out of deductions and exemptions for reported income above $250,000, any two-earner family in a high-tax state could keep their income below that pain threshold by increasing 401(k) contributions, switching investments into tax-free bond funds, and avoiding the realization of capital gains.
• Faced with numerous tax penalties on added income in general, many two-earner couples would become one-earner couples, early retirement would become far more popular, executives would substitute perks for taxable paychecks, physicians would play more golf, etc.
In short, the evidence is clear that when marginal tax rates go up, the amount of reported incomes goes down. Economists call that "the elasticity of taxable income" (ETI), and measure it by examining income tax returns before and after marginal tax rates claimed a bigger slice of income reported to the IRS.
The evidence is surveyed in a May 2009 paper for the National Bureau of Economic Research by Emmanuel Saez of the University of California at Berkeley, Joel Slemrod of the University of Michigan, and Seth Giertz of the University of Nebraska. They review a number of studies and find that "for an elasticity estimate of 0.5 . . . the fraction of tax revenue lost from behavioral responses would be 43.1%." That elasticity estimate of 0.5 would whittle the Obama team's hoped-for $1.2 trillion down to $671 billion. As the authors note, however, "there is much evidence to suggest that the ETI is higher for high-income individuals." The authors' illustrative use of a 0.5 figure is a perfectly reasonable approximation for most purposes, but not for tax hikes aimed at the very rich.
For incomes above $100,000, a 2008 study by MIT economist Jon Gruber and Mr. Saez found an ETI of 0.57. But for incomes above $350,000 (the top 1%), they estimated the ETI at 0.62. And for incomes above $500,000, Treasury Department economist Bradley Heim recently estimated the ETI at 1.2—which means higher tax rates on the super-rich yield less revenue than lower tax rates.
If an accurate ETI estimate for the highest incomes is closer to 1.0 than 0.5, as such studies suggest, the administration's intended tax hikes on high-income families will raise virtually no revenue at all. Yet the higher tax rates will harm economic growth through reduced labor effort, thwarted entrepreneurship, and diminished investments in physical and human capital. And that, in turn, means a smaller tax base and less revenue in the future.
The ETI studies exclude capital gains, but other research shows that when the capital gains tax goes up investors avoid that tax by selling assets less frequently, and therefore not realizing as many gains in taxable accounts. In these studies elasticity of about 1.0 suggests the higher tax is unlikely to raise revenue and elasticity above 1.0 means higher tax rates will lose revenue.
In a 1999 paper for the Australian Stock Exchange I examined estimates of the elasticity of capital gains realization in 11 studies from the Treasury, Congressional Budget Office and various academics. Whenever there was a range of estimates I used only the lowest figures. The resulting average was 0.9, very close to one. Four of those studies estimated the revenue-maximizing capital gains tax rate, suggesting (on average) that a tax rate higher than 17% would lose revenue.
Raising the top tax on dividends to 23.8% would prove as self-defeating as raising the capital gains tax. Figures from a well-know 2003 study by the Paris School of Economics' Thomas Piketty and Mr. Saez show that the amount of real, inflation-adjusted dividends reported by the top 1% of taxpayers dropped to about $3 billion a year (in 2007 dollars) after the 1993 tax hike. It hovered in that range until 2002, then soared by 169% to nearly $8 billion by 2007 after the dividend tax fell to 15%. Since very few dividends were subject to the highest tax rates before 2003 (many income stocks were held by tax-exempt entities), the 15% dividend tax probably raised revenue.
In short, the belief that higher tax rates on the rich could eventually raise significant sums over the next decade is a dangerous delusion, because it means the already horrific estimates of long-term deficits are seriously understated. The cost of new health-insurance subsidies and Medicaid enrollees are projected to grow by at least 7% a year, which means the cost doubles every decade—to $432 billion a year by 2029, $864 billion by 2039, and more than $1.72 trillion by 2049. If anyone thinks taxing the rich will cover any significant portion of such expenses, think again.
The federal government has embarked on an unprecedented spending spree, granting new entitlements in the guise of refundable tax credits while drawing false comfort from phantom revenue projections that will never materialize.
Mr. Reynolds is a senior fellow with the Cato Institute and the author of "Income and Wealth" (Greenwood Press, 2006).