This column originally appeared in Tax Notes, a subscription-only publication, http://www.taxanalysts.com/ .
By Diane Lim Rogers
Budget baselines might be boring, but they really matter -- especially now. According to the Congressional Budget Office's just-released update of the 10-year federal budget outlook, the difference between the current-law baseline and the extended policy baseline, just in terms of the tax side of the budget, is more than $4.7 trillion over 10 years.1The Bush tax cuts alone -- without alternative minimum tax relief -- account for $2.5 trillion, because under current law they are scheduled to expire at the end of 2012. And despite President Obama's continued emphasis on his "fiscally responsible" stance regarding the high-end Bush tax cuts, his own budget proposals still would produce revenues that are $2.3 trillion below current-law levels. And those amounts don't include the associated added interest costs.
Even though our lawmakers are responsible for making policy choices relative to current law, they have tried repeatedly over the past several years to make their proposals appear fiscally responsible by measuring them against a policy-extended, business-as-usual baseline. Measured against the CBO's current-law baseline, it's not just the "No New Taxes" Republicans who are proposing tax changes that would actually increase the deficit. The Obama administration and congressional Democrats continue to propose reducing revenues relative to current law, making their call for a more balanced approach with more revenues consistent only with the policy-extended baseline.
As a result, policymakers have set extremely low standards for tax policy's role in deficit reduction. Rather than using revenues to start filling in the deficit hole, they've used tax policy to just keep digging. William Gale offered his spot-on, common-sense explanation of this phenomenon at a recent Brookings Institution event on the debt-limit deal's supercommittee, comparing deficit reduction to weight loss and the policy-extended baseline to a dieter's continued bad habits:
- In terms of an example, think about this the following way: Suppose you've been eating badly the last, let's say, 10 years, and you've been gaining a lot of weight and you want to lose weight and you want to lose 15 pounds. Well, we won't go with 1.5 trillion pounds. You want to lose 15 pounds. The question is, compared to what?
Now, the way we'd usually think about it is, well, compared to where I am right now I want to lose 15 pounds. There's another way to think about it, though, which is to say, well, I've been eating badly for 10 years. If I continue to eat badly the next 10 years I'm going to gain 45 more pounds, so I'm going to lose 15 pounds relative to that increase of 45 pounds that I'm going to do over the next decade.
Now, nobody that's serious about losing weight builds in a 45-pound weight increase and then says, "I'm going to lose 15 pounds relative to that." But using one of the baselines, the policy-extended one, as the standard for a deficit reduction goal would be the equivalent of increasing the deficit by $4.5 trillion and then saying, "I'm going to cut it by $1.5 trillion [in other words, increase it by $3 trillion relative to current law]."2
Meanwhile, Republican leaders ironically view prospects of an official CBO baseline standard for revenues with glee, claiming it sets such a high bar for taxes that the policymakers on the supercommittee won't even approach it, let alone try to jump it. So much for hoping for superheroes on the supercommittee.
But we do need that higher bar. For revenues to play any meaningful role in deficit reduction, and for us to actually lose some weight by changing our poor diet and exercise habits, policymakers will have to set a revenue target somewhere far closer to -- ideally, precisely at -- current law. The best outcome would be if policymakers agree to strict, no-exceptions, pay-as-you-go rules on expiring tax cuts, which by definition would require current-law levels of revenue. That would allow enough deficit reduction to achieve fiscal sustainability over the medium term of the next 10 to 20 years -- long before any entitlement reforms would significantly affect spending levels.
Achieving current-law levels of revenue does not necessarily require letting current law play out -- which would be easy, by the way, because Congress and the administration could just go home and do nothing. Here are three broadly different tax policy strategies that could each be consistent with the current-law baseline:
- 1. Do Nothing. Allow all expiring tax cuts to expire as specified under current law. That would mean reverting to Clinton-era marginal tax rates. (Hmmm, what was so bad about those tax rates for our economy?)
2. Do It Big. Extend some or all of the marginal tax rates under the Bush tax cuts, but fully offset the costs of extending the low rates by broadening the tax base and reducing some tax expenditures (for example, limiting itemized deductions or reducing the exclusion of employer-provided health benefits). This is the fundamental tax reform approach.
3. Do It to the Rich. Extend some or all of the Bush tax cuts -- particularly those that affect middle-income taxpayers (lower tax rates, child tax credit, marriage penalty relief) -- and fully offset the costs by imposing an extra tax on the very rich, such as a surtax on households with incomes in excess of $1 million.
How do the three strategies compare in terms of economic effects? Theoretically it seems they would not be as different in their effects on the shorter-term, demand-driven recovery as on longer-term, supply-side growth. Pursuing the second or third option might do less damage to near-term demand than the do-nothing approach. But none of those income tax increases would threaten the recovery as much as the alternatives of spending cuts or letting the payroll tax cut expire.
All three tax policies would achieve the same amount of deficit reduction based on static revenue estimates. But are there potential differences in their dynamic effects on the economy's productive capacity? The higher marginal tax rates under the do-nothing approach are not likely to have much of a disincentive effect on labor supply or saving -- as we learned the last time we lived through the Clinton tax rate increases. Those marginal income tax rates, peaking at 39.6 percent, are still relatively low by historical and international standards. Option 2's fundamental tax reform approach of keeping much of the marginal rate structure while broadening the tax base is best for supply-side effects and overall economic efficiency, because the distortionary effects of taxes on economic decisions would be minimized. Option 3's strategy of raising rates just at the very high end of the income distribution means marginal rates at the top would have to go up substantially more, but it's unclear that the economic incentives of the rich to earn more income would be changed that much as long as their marginal tax rate were still far from 100 percent. Warren Buffett certainly disputes this worry.3
Combining all three approaches is possible, too, and makes the revenue target easier to achieve without having to take any one option to an extreme. For example, we could let the top tax rates expire or at least come up a bit (rather than letting all rates expire), reduce some tax expenditures in progressive ways (without eliminating them), and even add a new top bracket around the millionaire level, as suggested by Bruce Bartlett,4 without having to raise the top rate to the problematic levels suggested by the Tax Policy Center's "Desperately Seeking Revenue" analysis.5 Increasing the capital gains tax rate to something closer to that on ordinary taxable income could represent a combination of options 2 and 3 -- reducing a tax expenditure (the preferential rate) that disproportionately benefits the rich.
What are the potential differences among those paths politically? Of course, option 1 emphasizes a do-nothing Congress (so why bother keeping them on their job?); option 2 (fundamental tax reform) means policymakers will have to work much harder at actual policymaking and requires more public education, engagement, and support of those efforts; and option 3 is susceptible to the class warfare and redistribution criticisms and partisan battles. I think the politics suggests that some combination of all three ways of getting to current-law baseline revenue levels is probably best.
Must we go all the way to the current-law revenue baseline? Achieving the grand bargain on deficit reduction and going bigger than required with revenues would give us more of a cushion to allow for future waivers of pay-as-you-go rules for truly stimulative tax policies (or spending) that would better qualify as emergency spending. We need to change the question from, "Do we want the Bush tax cuts?" to, "Are the Bush tax cuts the best way to spend $2.5 trillion over 10 years -- for any purpose, be it short-term stimulus or to encourage longer-term economic growth?"
Whichever way we get there, setting our revenue standards to the current-law baseline would get us the more balanced approach to deficit reduction that Americans desire, bring revenues high enough to keep deficits at an economically sustainable level over the medium term (while we continue to work on gradual entitlement reforms), and leave us more flexibility in tax and spending policy to better address and not worsen the ailments in the near-term economy.
Note that revenue levels have huge potential in deficit reduction but aren't the most important factor in determining tax policy's role in stimulating (or contracting) the economy. Big overall revenue losses don't necessarily translate into big increases in the demand for goods and services (and hence the creation or maintenance of jobs in a recessionary economy), if the tax cuts go disproportionately to high-income households and businesses that are least likely to immediately spend their tax cuts. Setting a goal of boosting revenues to current-law levels to be achieved by increasing the overall progressivity of the tax system is therefore likely to both reduce the deficit and provide more effective support for the still-fragile economy. In fact, if some combination of the different approaches to getting to the current-law baseline were taken, we could easily gain enough revenue to be able to apply it to both reducing deficits to economically sustainable levels and temporarily providing more stimulative deficit-financed tax cuts or spending.
In conclusion, sticking to the current-law revenue baseline really isn't that hard to do, and there are lots of opportunities for our policymakers to actually make better tax policy while doing it. It starts with a simple commitment to strict pay-as-you-go rules on expiring tax cuts, something very easy for members of the debt deal's supercommittee to do right away.
2 Aug. 17. For information, see http://www.brookings.edu/events/2011/0817_deficit_committee.aspx.
3 Warren E. Buffett, "Stop Coddling the Super-Rich," The New York Times, Aug. 15, 2011, at A21.