A few notes on soda taxes (a footnote to @cjsnowdon’s work)

I was just reading this post by Chris Snowdon, about jurisdictions that impose taxes on soda, ostensibly in the name of fighting obesity. He notes that this is patent bullshit, since they typically impose the same taxes on energy-free versions of those drinks (“diet sodas”) also. In that post, he suggests that it is purely a money-grab. Even with that incentive, I suspect the “moralizing” aspect, the disdain from certain people toward those of us who indulge in the sins of soda, is not purely a rationalization. I suspect that those in power are true believers in the moralizing, not just taking advantage of it to find an easy revenue source. I also suspect Chris has said exactly the same thing in his extensive analyses of the point. If you are looking for points about the subject in general, in other words, look to him, not me. I just have a couple of additional observations that are basically a footnote to his writing.

Soda taxes, like any consumption tax for something that is not strongly a luxury good (in the technical sense), are highly regressive. Indeed, proponents often indirectly concede that they consider it a feature, rather than a bug, that cigarette excise taxes are brutally punishing for poor people even though wealthier people can shrug them off. They delight in how much pressure they are exerting on the poor in particular. One might think this means that these — soda taxes, cigarette taxes, and the like — are always pushed by the upper-class elitists, like those who comprise the global tobacco control jet-setters. These are the elites who are just oh-so-concerned about the plight of the poor, by which they mean the poor actually trying to seek fulfillment in their lives rather than acting like proper serfs.

Perhaps this is the case for the soda tax in France that Chris mentions. But one thing to realize about the U.S. situation is that when urban governments enact these measures, they are often led by, and certainly supported by, the black political machines that dominate parts of those cities. (I suppose I should say “African-American political machines”, but I am already going third-rail with this, so why bother?) Yes, those leaders are also are much wealthier than their constituents, but it is not really the same motivation. The black churches and urban black oligarchs are quite often extreme true-believer puritan moralists. Much more so than the jet-setter moralists who drink, smoke weed, fornicate, etc., even as they are punishing smokers and cola drinkers. This does not necessarily affect the analysis, but someone trying to understand the politics should keep it in mind.

So what actually happens in the cities? Here I can get more concrete based on geography. It turns out that the regressiveness is much greater than a simple analysis of “soda purchases as a portion of income” analysis would suggest. Chris mentions Philadelphia, Oakland, and Seattle. I have lived in the first two and spent a lot of time in the third. I can tell you that there is basically no normal middle (or higher) class person in the Philadelphia or Oakland who does not shop outside the city limits, at least a few times per month, and many do most of their shopping outside the city. (That “normal” is meant to exclude the car-eschewing and/or urban sophisticate type, people who are clearly not the target for these sin taxes. Note that I am not casting any aspersions here — I have been both of those types, as well as the typical commuter, at various points in my life.) So if they care about this tax at all, if the money matters to them or they are otherwise inclined to avoid paying it, they can buy all their retail soda outside the taxation zone. In Oakland they would also have to avoid buying when they shop in neighboring Berkeley also, but that does not change the equation much. For Seattle, I would guess that more of the middle class does all their shopping within the city limits, but they are still much more likely to be able to avoid this tax without burden than poor people.

So it is the poor — people who do not drive beyond the city limits for commuting or recreation, and who do not go shopping at the upscale malls, Wegmans, and Ikeas that are just out of town — who are left paying the tax when they shop locally. This includes the people we normally think of as the urban poor, along with the elderly, teenagers, and college students. Again, the perpetrators of these punitive taxes probably are not bothered by this. They know these are the people who might actually change their behavior in the face of the punishment, rather than shrugging it off.

If the taxes get high enough that they really hurt, we will get the same phenomenon we have with cigarettes. Someone will drive to wherever they can buy pallet-loads cheapest, within a few hundred miles, and then sell them out of the back of their truck for less than the local price would have been, even without the tax.

All this is about retail purchases. Restaurant or convenience purchases do not allow for such evasion. So we are back to the standard regressiveness calculation. Except we are talking about people who can indulge in such expenditures in the first place, and are probably not too price sensitive.

But wait. How are soda prices in restaurants set? (The analysis is similar for vending machines and other convenience sellers.) They are not the competitive prices for grocery retail purchases, where sellers cannot charge much more than their competitors. Restaurants have a monopoly on the soda purchases of their customers. The purchase price is almost pure profit. The marginal cost to the seller of a $2 soda is in the order of ten cents. So how do they set their price? The monopolist’s profit-maximization involves jacking up the price, losing a lot of buyers who would have made the purchase for $0.20, to the point that net revenue from lost buyers (who at that point are really not paying much attention to a few tens of cents more) exceeds the extra per-unit net revenue. But that actually seems like the naive economist talking. I suspect that restaurants really do not know what the monopolist price is, but instead raise the price to the level they can get away with without offending their customers, because it seems dickish to charge that much, causing them to stick to water on principle even though the cost is really no big deal to them, or even avoid the establishment entirely. Thus it is “just” $5 at an expense-account restaurant that charges $15 for a beer and $50 for an entree, even though $10 would increase their profits, and just $1.89 at McDonalds.

So what happens when the restaurants have to pay $0.20 for the inputs, due to the tax, rather than $0.10? Nothing. The dickishness threshold is not changed. Since these sales are to people who are less poor, this further increases the regressiveness.