Peter Orszag on Middle-Income Tax Increases, Plus Broader Thoughts

by Reihan Salam


Rather than focus on the budget proposal and its manifold flaws, I prefer to focus on those making the strongest arguments for the president’s policy priorities. I suspect that former OMB director Peter Orszag gives us a clearer indication of where the Obama administration might go in a second term in a recent column:

[T]o significantly reduce the deficit over the next decade, additional revenue will be needed. The administration’s budget proposal projects revenue to reach 20 percent of gross domestic product by 2022, about 1 percentage point of GDP more than what is projected with no policy change. The administration deserves credit for proposing even that, given the antipathy to any tax increases. But in the end more revenue will be needed. And since the administration’s budget probably shows the outer limit of what’s plausible in terms of taxing high-income households, the implication is that middle-income households will have to pay more, too.

One wonders if Orszag is referring to the outer limit of what’s politically plausible in terms of taxing high-income households or if he is acknowledging the implications of new research on the elasticity of taxable income that factors in the extensive margin as well as the intensive margin, a subject that Arpit Gupta has discussed in this space. Regardless, it is heartening to see a veteran of the Obama White House, who did so much to frame the case for the new health law and to shape the president’s approach to the future size and shape of federal expenditures, establish that the president’s approach will necessitate tax increases for middle-income households. It would be nice to see if Jack Lew or President Obama would contest Orszag’s claim. 

To be clear, we may well need to raise taxes on (some) middle-income households even if we implement serious entitlement reform. We can at least imagine Medicare and Medicaid reforms that wouldn’t necessitate tax increases, but such reforms would be very unlikely to prove politically sustainable and would raise debt levels to dangerously high levels during the transition. Politically sustainable entitlement reform, e.g., the implementation of a transparent Medicare premium support system tied to a defined benefit, might generate dynamic cost savings, yet even then would probably require tax increases in light of the changing demographic structure of the population, at least in the medium-term.

The question is how we should approach this larger tangle of issues. A few thoughts:

(1) Federal taxes aren’t the whole story. Strategies that keep federal taxes low while encouraging state and local taxes to rapidly increase aren’t particularly useful if our goal is to increase the net disposable income of households. Transfers are also an important part of the story, e.g., are we transferring from the nonpoor to the poor or from the nonpoor to the nonpoor? Entitlement reform isn’t just important because it has the potential to restrain spending growth; it might also rationalize our transfer regime by making it more progressive. So the progressivity of the overall tax-and-transfer regime is important to keep in mind, as is the extent to which federal taxes can balance out state and local taxes. Because the federal government has an obvious advantage in taxing highly-mobile tax bases, state and local taxes will (and should) tend to tax less-mobile tax bases, e.g., property, retail sales, etc. We’d thus want the federal tax system to be a bit more progressive than we would if the regressivity of state and local taxes weren’t an issue.

(2) We should emphasize curbing tax expenditures rather than raising marginal tax rates or raising taxes on capital income. So a broader shift towards a progressive consumption tax like the Bradford X tax would be ideal, as Glenn Hubbard has argued (and also here). This would tend to mitigate the growth impact of any tax increase. Moreover, it would help improve the allocative efficiency of the economy.

(3) Our debates concerning the future of the federal government tend to underemphasize the importance of the diffusion of responsibility caused by the proliferation of joint state-federal programs, the foremost example of which is Medicaid. Michael Greve is the most lucid guide to this issue. Medicaid block grants, the most popular approach to containing cost growth and properly aligning incentives on the right, aren’t likely to work very well, as states will constantly dare the federal government to not bail them out by behaving recklessly. It would be far more sensible to revive the Reagan swap and federalize Medicaid while downloading other shared programs to the states. Coupled with the elimination of the state and local tax deduction, which helps insulate affluent and influential households from the cost of profligate state and local governments, this measure would go a long way towards restoring a competitive dynamic between states that could redound to the benefit of citizens.  

If we embrace this approach, some middle-income households will have to bear more of the cost for making choices that are currently subsidized (owning a home, living in a high-tax jurisdiction), but middle-income renters, residents of low-tax jurisdictions, and non-itemizers would generally be better off.