Responding to the Smarter Liberal Critiques of Simpson-Bowles

by Josh Barro

There’s been a lot of screaming from the Left about the “catfood commission,” but not every liberal objection to the report is stupid or shrill. Let’s take a look at the Left’s better objections to Simpson-Bowles.

The smarter liberal critiques of Simpson-Bowles have generally focused around two areas. One is that it cuts Social Security too much and Medicare not enough. The other is that it sets a revenue target that is too low, and/or includes a tax system that is not progressive enough. If Democratic politicians don’t like Simpson-Bowles, it would be nice to see them sign onto these critiques and try to modify it into a plan of their liking.

I don’t have too much to say about the second critique, except that I think it undersells the extent to which abolishing most tax deductions and credits is a progressive reform, so long as you retain the Earned Income Tax Credit and the Child Credit. Paul Krugman bizarrely reacted to the Simpson-Bowles package of base broadening and rate cuts as though reducing marginal tax rates on high earners is itself a bad thing, independent of the total amount of collections. Derek Thompson of the Atlantic explains nicely why the commission’s plan is more progressive than it looks.

It’s true that a plan that trades lower rates for a broader base will generally mean lower tax bills for extremely high earners, but much of that burden is shifted down into the rest of the top quintile, which now benefits handsomely from deductions for home mortgage interest, health care expenditures, and state and local taxes paid. I don’t think that’s much of a social justice concern. And because Simpson and Bowles did not propose a VAT, I don’t think it’s likely that their plan would shift much of the tax burden down to lower-income people.

That said, it would of course be possible to conceive a tax reform plan that would be more progressive (or less), or one that would raise more revenues (or less), simply by moving the rates around. This is simply a question of tradeoffs between economic growth and equity. But in any case, base broadening and tax simplification are a good idea, as most commentators on the Left seem to agree, except Krugman.

On the first point, critics on the Left have rightly pointed out that the Simpson-Bowles approach to Medicare is a bit of a “hand-wave,” with its centerpiece being declaration that Medicare spending growth should be capped at GDP plus 1 percent. It’s true that this is maddeningly vague. It’s also true that Simpson-Bowles has very specific Social Security proposals even though medical entitlements are a bigger problem than Social Security. But I still think people like Kevin Drum are being unfair when they say the plan is therefore not “serious.”

Indeed, the hand-wave is awfully similar to certain provisions in the Affordable Care Act, which sets cost reduction targets for Medicare and then directs CMS to draw up plans to meet them, which Congress would have the power to disapprove. That’s not a plan either: it’s a plan to make a plan. This approach is used because saving money in Medicare is hard, and is going to involve a certain amount of trial and error. For example, ACA provisions such as the Cadillac Tax should bend the cost curve down and produce some savings, but we won’t know how much until we get there—so we can’t say today exactly what policies we should enact in the future to save money.

Basically, “what should Social Security policy be in 2035?” is a question that we can much more plausibly try to answer today than “what should Medicare policy be in 2035?” That doesn’t mean we couldn’t come up with more specific answers than Simpson and Bowles do. But I do think it means we sort of need to grade their Medicare proposal on a curve, and accept the fact that making Medicare sustainable will be a decades-long process that we cannot precisely map out in advance.

It’s also not true, as Drum claims, that discretionary spending and Social Security are not significant problems or that a “serious” budget plan should focus entirely on Medicare and Medicaid.

Take Social Security. Drum notes that CBO expects Social Security spending as a share of the economy to “[go] up very slightly between now and 2030 and then [flatten] out forever.” That “very slight” uptick is in fact about 1 percent of GDP, or a relative increase around 20 percent. This is a big adjustment that would require about 20 percent more revenues for Social Security if benefits are not reformed. Absolutely, Medicare is a bigger problem, but this is still a significant issue that requires a policy solution.

More broadly, as health care spending grows to take up a larger share of the economy, spending on things other than health care will have to take up a smaller share of the economy every year. We must take steps to control health care cost growth, but we cannot reasonably expect to slow it to the rate of growth in GDP. One reason is that the structure of the health care industry produces fewer productivity gains from new technology than other industries. Another is that we keep inventing new health care technologies that are useful and therefore worth spending money on.

The question, then, is whether health care spending will crowd out other government spending or other private spending. The correct answer is surely going to be “both,” and indeed the deficit commission’s revenue target of 21 percent of GDP, much derided on the Left for being too meager, would constitute a record level of federal revenue except for during World War II. This higher level is used to make higher health care entitlement expenditures possible—and would mean reduced private expenditure as a share of the economy.

But we should not expect the crowding-out effects of higher health expenditure to fall solely on the private sector. This is especially true because public expenditures must be financed with taxes, and the economic cost per dollar of taxation rises the more you tax. As discussed before, this is because the deadweight loss of taxation (the economic costs of distortions created by taxes) is a function of the tax rate squared. Discretionary Program X might be worth the deadweight loss it imposes when it means the difference between taxes at 17 percent of GDP and 18 percent, but no longer appropriate when it means the difference between revenues at 22 percent and 23 percent.

I am not necessarily a proponent of a hard cap on federal revenues as a share of GDP, but I do think it’s correct to say that part of the appropriate response to rising Medicare and Medicaid costs is to cut other areas of government spending. And therefore, even though medical entitlement spending is far and away the largest driver of our long-term budget problems, the search for budget solutions should cast a wide net.

Josh Barro is the Walter B. Wriston Fellow at the Manhattan Institute.

The Agenda

NRO’s domestic-policy blog, by Reihan Salam.