Wednesday, April 27, 2016

Soda taxes: pros and cons

Soda tax supporters would happily pay an extra 12 cents for this artisanal obesogenic toxin
In another sign that the current Democratic presidential primary is unusually substantive and policy-focused, Clinton and Sanders recently skirmished a bit over the merits of Philadelphia's proposed soda tax. This is a perfect wonky issue for liberals to discuss, because it shows real differences between the candidates. Unlike topics such as gun control, minimum wages or climate change, where both Clinton and Sanders are directionally in sync compared to Republicans, it's not clear which candidate is 'more liberal' on policy involving the demand-side of health care.

While this article claims Sanders' opposition to soda taxes marks him as more conservative, I'm not sure soda taxes have received enough national attention and scrutiny to clearly define the issue's partisan dimensions. While soda taxes are opposed by conservatives, the legislative action so far has mostly been by liberal mayors in cities--it hasn't yet acquired status as a partisan 'signal' among national Democrats. Sanders' explicitly liberal argument against it suggests the issue's partisan valence isn't settled just yet.

On its face, soda taxes appear quite similar to other 'sin tax'-style public health consumption taxes like those on alcohol and cigarettes. Setting aside the merits of policy, soda taxes are much more politically marginal: alcohol taxes receive support from religious conservatives, and cigarette taxes received popular support from legal outcomes and from fears over secondhand smoke harms. The public health benefits of soda taxes are much more vague and complicated.

Ultimately, the debate over soda taxes boils down to a comparison of its perceived pros and cons. Let's take a look at the big ones:

Pro 1: Public health. Nutrition science is increasingly converging around refined carbohydrates and sugar as the primary driver of metabolic syndrome and diabetes. By raising the price of soda, policymakers hope to reduce its overall consumption and increase overall health. Having healthier people would be good for its own sake, but it would also provide a big boost to the economy and might reduce the demand for health care services, perhaps lowering costs. Poor people are disproportionately affected by diabetes, bad health, and high health care costs, so part of this argument is that soda taxes are progressive (in a causally complex way).

Con 1: Regressive. Bernie Sanders opposes the Philly soda tax on the grounds that poor people spend a disproportionate amount of money on soda and food generally, and so would be harmed disproportionately via higher soda prices.

Pro 2: Efficient taxation. Economists generally view taxing 'bad' things (like cigarettes) as preferable to taxing 'good' things (like income) because of the incentive effects. Additionally, consumption taxes have the added benefit of encouraging savings and investment, which the average US consumer probably should do a bit more of. If the revenue from soda taxes allows for lower taxes in other areas, the economy's tax structure will be more efficient.

Con 2: Rent-seeking. Of course, Pigouvian consumption taxes are only efficient in theory. In practice, soda tax legislation might create opportunities for businesses and interest groups to insulate themselves from competition and acquire extra income at the expense of consumers. For example, sugary juices or double iced mocha no-whip frappuccinos have typically been excluded from Soda taxes, creating a weird market distortion that helps some businesses, hurts others, and does nothing for public health if people simply switch their sugary drink selection.

Pro 3: Template for better future policy. Much of the public policy that affects demand-side health care (nutrition, physical fitness, etc.) is hopelessly kludgey, ineffective, and filled with rent-seeking opportunities. Public health is overwhelmingly driven by huge forces like poverty, culture, land use policy, and trade policy; tiny programs like Let's Move surely have little to no effect. If tax policy can emerge as the primary lever of federal action on this issue, many of these ridiculous programs can go away. Soda is as good a candidate as any for normalizing tax policy as an effective, less partisan, more acceptable form of government action in the domain of demand-side health care.

Con 3: Paternalism. This critique is the most classically 'conservative': by singling out soda, the liberal elite is devaluing the decisionmaking ability of poor, uneducated, lower-class sodadrinkers. This could be specific ("why soda and not other unhealthy foods") or general ("why tell people what they can or cannot eat"), but the overarching theme is that government is taking a step too far into people's lives in exchange for uncertain benefits.

I'm generally supportive of soda taxes, and see Pro #3 as particularly persuasive. The costs of rent-seeking and paternalism certainly deserve to be taken seriously, and in a perfect world the government would have little explicit role in demand-side health care policy because of these concerns. Unfortunately, the world we live in is filled with ineffective health programs, and so the alternative to soda taxes isn't no government intervention--it's simply the status quo setup. More soda taxes would be a marginal improvement if they displace current or future bad policies.

Surge pricing... get it?
Sanders' charge that soda taxes are directly regressive is interesting, but not decisive. A policy's distributional effects are important, but they are only one component of a holistic accounting of total impact. Soda taxes are regressive, but not so much so that their other, positive effects should be immediately discounted. Many policies that are regressive accomplish socially useful things, and we accept these in part because other policies with compensating distributional effects exist (or can exist).

For example, Sanders' free college plan is poorly targeted and if implemented could very plausibly end up being regressive (by greatly helping the already well-off and not helping the truly desperate). Yet there are compelling arguments for why it might still be worthwhile: its simplicity makes it more politically and bureaucratically advantageous than Clinton's plan, and it would greatly help poor and disadvantaged people.

For another example, the mortgage interest rate deduction is a terribly regressive policy that overwhelmingly benefits wealthy people. But it also costs a ton of money, has socially destructive effects on economic behavior, and has miniscule beneficial effects. It's the sum of its total impact that makes this policy atrocious, not any single piece.

I think it's well-understood that specific policies like SNAP, EITC, Obamacare etc. are designed with distributional effects as their explicit rationale, and most policymakers would admit that in the abstract this differentiation is mostly a good thing. Requiring every policy to be progressive or neutral in its distributional effects would greatly limit the range of policy alternatives, and isn't realistic.

I like that Sanders raised this somewhat wonky critique of a new public healthy policy trend, but worry that the isolated demand for distributional progressivity will become a new tool in the far-left's rhetorical toolbox when it comes to opposing specific policies (Sanders supports the mortgage interest rate deduction). The deep concern is that our federal legislative gridlock has made it difficult to make multiple policies synergize and work with each other. If congressional Republicans block all explicitly redistributive policies, some Democrats may increasingly require progressivity in policies where doing so would result in less effective policy.

No comments :

Post a Comment