Spinning soda into gold

They’re talking about taxing sodas again. TigerHawk doesn’t sound happy; Conor Clarke asks what the big deal is. To me, the big deal is that I hate all sales and consumption taxes - actually all individual taxes except the income tax. If I assume, however, that the government is bound and determined to levy some kind of tax on “unhealthy” food and drink then what becomes important to me is making sure we understand what we should be taxing and why we’re taxing it.

Clarke has up an interesting chart that shows the results of asking people who and what they would be willing to tax “to help pay for health care reform and provide coverage for more of the uninsured.” Soda and soft drinks were a less popular target than (in order of popularity) cigarettes; families making more than $250,000 a year; wine and beer; and unhealthy snack foods. Clarke is puzzled about the unpopularity of taxing soda. I’m not.

It does seem illogical to support taxes - higher taxes - on cigarettes, wine, and beer while not also supporting taxes on “unhealthy snack foods” and “soda and soft drinks”. But it’s not surprising: relatively few people smoke and I suspect that relatively few people drink enough to make raising taxes on wine and beer an issue. (Plus we can all just start drinking bourbon instead. Although I like mine with Coke so I may be out of luck anyway.) An awful lot of people, however, like a cold Coke on a hot day at least occasionally. And since the survey didn’t specify only non-diet sodas, it’s not just those of us who are addicted to the real thing who are going to resist a soda tax.

Also, people are aware that cigarettes, beer, and wine are already subject to significant taxes so the idea that’s it’s “normal” to tax those products is firmly implanted in most people’s brains. Similarly, cigarettes and alcohol are firmly filed under “Vice” while snack foods and sodas are equally firmly filed under “Treat”. Finally, while people may agree with Clarke that soda isn’t good for people’s health, I don’t think most people think of soda as being bad for you in the same way that cigarettes are and alcohol can be.

One issue that Clarke gives weight to is where to draw the line in defining what is and is not a “soda and soft drink” and he bemoans the lack of a bright line to distinguish what would and would not be taxed. On that score, I have a suggestion. Well, two suggestions, actually.

Let’s start by looking at the rationale behind taxing sodas: sodas cause obesity and obesity is a big health problem so if we tax sodas we will improve health by reducing consumption and raise money for other health-related improvements. Interestingly, the Wall Street Journal article cited by both TigerHawk and Clarke does not provide any links to any evidence of the causal relationship between sodas and obesity. The WSJ article does identify The Center for Science in the Public Interest as the moving force behind the idea of taxing sodas to pay for health care reform so I checked their Website and found a one-page memo proposing the soda tax. The memo states - without hyperlink, citation, or footnote - that:

More bad news comes from researchers who are finding that soft drinks are especially good at making people gain weight. In fact, soft drinks are the only beverage or food that has been linked to a greater risk of obesity.

Perplexed but undaunted, I did a quick Google search for “soda causes obesity” and found the top hits dated from 2006 or earlier. An article in The San Diego Union-Tribune seems to sum up the state of play pretty fully. It presents four arguments advanced to support the causal link between soda and obesity.

Argument 1: While soft drink consumption was rising between 1977 and 1997 so was obesity. Somewhat haphazardly lumped into the same argument is a reference to two studies that found obesity increased in schoolchildren and nurses when they consumed more sodas.

The first part of this argument presents correlation but not causation. The second part is somewhat tautological: as students and nurses consumed more calories they gained weight. None of this answers the really important questions: Why did soft drink consumption rise and why does soda consumption apparently occur on top of existing calorie consumption rather than replacing it? In other words, when the students and nurses drank more soda why didn’t they cut back on calories from other sources? Answers to those questions are contained in the second argument.

Argument 2: Soda is sweetened with high fructose corn syrup (HFCS) which does not act the same way other carbohydrates do to reduce appetite. Thus while consuming “real food” causes the body and brain to figure out that they’ve eaten, consuming HFCS does not. (This is a remarkably sloppy section. It begins by talking about HFCS then cites evidence from a study of “caloric beverages”. One is left to assume without evidence that the caloric beverages were sweetened with HFCS.)

Unfortunately, Argument 2 is presented in such a way as to only address the question about why people who consume soda don’t reduce calories from other sources. A little thought, however, reveals that the problems with HFCS can shed even more light on Argument 1. If high fructose corn syrup doesn’t produce feelings of satiety, perhaps that’s why soda consumption rose from 1977 to 1997 - a period that matches up nicely with the introduction of HFCS into sodas:

HFCS was rapidly introduced to many processed foods and soft drinks in the U.S. from about 1975 to 1985.

In other words, this suggests that the causal relation is not between soda and obesity but between HFCS and obesity. Soda may be an intermediary - HFCS in soda causes increased soda consumption which causes obesity - but since HFCS is found in so many foods I'm not even convinced we can say that. Take away people's sodas and they will still be consuming a lot of HFCS. So if HFCS is the problem, people deprived of soda may simply increase calorie consumption from other sources in a continuing vain attempt to achieve satiety.

Argument 3: People who consume soda are more likely to eat less healthily across the board. There is absolutely no evidence - none, zip, zilch, nada - that consuming soda causes people to eat more fast food and fewer vegetables. It makes just as much sense to say that people who eat fast food are more likely to consume soda simply because it’s the most readily available beverage at fast food places. Or that people who are careful to eat healthily are more likely to avoid high-calorie beverages. Or that rich people eat vegetables and drink water while poor people go to McDonalds a lot.

Argument 4: I don’t even know how to describe this argument. The article says:

Many different types of studies link sugary drinks and weight gain or obesity. Some even show a “dose-response” relationship – as consumption rises, so does weight.

There’s that tautology again: as people consume more calories their weight goes up. And there’s that question again: why does soda consumption apparently occur on top of existing calorie consumption rather than replacing it? And notice the sloppy use of the word "sugary" when what's almost certainly meant is "HFCS-y". Argument 4 is just a restatement of Argument 1 with the addition of the words “[m]any different”.

As I said above, what I derive from this review of the evidence is not that soda causes obesity but that if there is any causal relationship at all it’s between high fructose corn syrup and obesity. So here’s my first suggestion. If the government is going to go ahead with this tax, don’t levy it on soda (or “unhealthy snack foods”): levy it on high fructose corn syrup. And since we don’t know whether HFCS’ bastard child, crystalline fructose, is any better, let’s tax that also. Think of the benefits. First, HFCS is in everything; the government will make a fortune in revenue. Second, since a HFCS Tax will advantage manufacturers who use real sugar in their products we can expect fewer items with HFCS and more items with real sugar. This will result in a real-life field test of the idea that HFCS is messing with our appestats and making us all fat.

My second suggestion is that the government tax artificial sweeteners also. There is research linking the consumption of diet sodas to obesity so if the goal is to reduce weight as well as raise revenue, there’s no logical reason not to tax artificial sweeteners right along with HFCS. People who cannot eat sugar, like diabetics, can submit their receipts for the purchase of artificially sweetened items along with a form signed by their doctor and get their taxes refunded.

And there you are. Tax high fructose corn syrup, crystalline fructose, and artificial sweeteners. Americans will become slender and thus healthy and the government will make a fortune in revenue. Er, or not. Actually you can only have one of those two outcomes. In order for people to become thin, they will have to eliminate the taxed items from their diet. But in order for the government to rake in the dough, people will have to continue to eat the taxed items. You can’t have it both ways. Which means that when The Center for Science in the Public Interest says:

The Obama administration needs to mount a comprehensive anti-obesity campaign, and slashing non-diet-soda consumption should be front and center. We need to get soda out of schools, install millions of water fountains across the country, require warning labels on soda containers, and sponsor a media campaign to counter the soda industry’s billion-dollar-a-year effort to maximize sales.

As it turns out, the quickest, most effective way to put a lid on soda sales would also give the government the money to do all that and more: slap a tax on carbonated and non-carbonated soft drinks.

what they really mean is either “We’re incapable of logical thinking” or “We don’t care about improving people’s health. We just want the government to have more money.”