Back in February, President Obama released a revenue-neutral framework for lowering the corporate tax rate from 35 percent to 28 percent, which he highlighted during last night’s presidential debate. And unlike Mitt Romney’s proposed overhaul of the income tax code, the Obama White House was willing to identify various loopholes and tax expenditures to make up for the revenue loss caused by the reduction of the statutory rate.
Yet as Robert Pozen of the Brookings Institution observed at the time, there was a problem:
Reducing the rate by seven percentage points would cost roughly $360 billion over five years, and the expansion of the manufacturing and R&D expenditures would add another $125 billion or so over five years. So the total cost would be $485 billion. (Most budgetary analysis is done on a 10-year basis, but tax expenditures are usually estimated over five years, so I’ll translate everything in this post into five-year numbers.)
Obama would pay for this by broadening the corporate tax base. Although he specifically mentions eliminating five tax expenditures, it’s clear that these represent only a token amount of savings:
* Favorable treatment of carried interest: $10 billion
* Accelerated depreciation of corporate jets: $1.5 billionOil & gas subsidies: $24 billion
* Moving from last-in-first-out to first-in-first-out: $21 billion
* COLI (a technical provision related to corporations that buy life insurance): $4 billion
That’s $60 billion total if you assume a 35% tax rate—with a 28% tax rate, eliminating these tax expenditures would only raise $50 billion. At this point, Obama is still $435 billion in the hole.
As Pozen goes on to explain, the Obama administration does go on to identify a “menu of options,” including reducing the bias towards debt financing. This strategy could raise a great deal of revenue.
Unfortunately, as Josh Barro has explained, reducing the bias towards debt financing is in tension with President Obama’s proposed “Buffett Rule”:
A corporation deducts interest payments before calculating its taxable income, and then an individual owner of corporate debt pays tax on interest payments at ordinary income rates. On the other hand, a corporation pays tax on profits after interest expense. These after-tax profits are either distributed to shareholders, who pay tax on the dividends; or they are retained, in which case the stock price rises and shareholders pay tax on capital gains.
Because interest is taxed only once and profits are taxed twice, corporations take on more debt than they would in absence of the tax distortion. The distortion is mitigated by the fact that dividends and capital gains are taxed at lower rates than interest income. Because the Buffett Rule would raise capital gains and dividend tax rates and, in many cases, lower the effective tax rate on interest, corporations would face even more incentive to overleverage themselves.
There is an irony here: one of the criticisms of Mitt Romney’s record at Bain Capital is that private equity firms put unhealthy amounts of leverage on the firms they acquire in order to exploit the favorable tax treatment of debt. The Buffett Rule would make that strategy even more attractive.
I understand why there has been little effort to address these questions. The corporate income tax is far less politically salient than the federal income tax, and the revenues involved are smaller. These two cases are not perfectly symmetrical by any means. But I’d suggest that there is a family resemblance. The Obama White House might have been wary of offering greater specificity on tax expenditures it would abolish to avoid intensifying political opposition from disfavored industries, and the Romney campaign presumably eschewed specificity on tax expenditures for the same reason.
There is one obvious rejoinder available to the president and his allies: if further cuts to tax expenditures prove unpalatable, the cut to the corporate tax could be smaller than 7 percentage points. But of course the same could be said of Romney’s income tax proposal.
(Thanks to Lucas Goodman for the pointer.)