A Nudge Toward Better Health

Diets hinged on sweeping, systemic changes to the way you eat are often the least sustainable. Dieters frequently gain back the weight as soon as willpower wears thin. Sustainable weight loss is achieved through small, sustainable tweaks in behavior over the long term. That might be the case not only for those of us trying to lose weight after the holidays, but for the overweight populations of entire countries. 
 
For nations seeking to address the epidemic obesity and the looming health problems and costs it brings, taxes on sugary drinks may provide a helpful nudge, changing behavior for the better. Increasingly such taxes come up as a possible solution in the US, where 69 percent of Americans are overweight. 
 
Jeffrey Grogger, the Irving Harris Professor in Urban Policy at the Harris School of Public Policy, wanted to know exactly how much of the problem such taxes could realistically address. To find out, he analyzed price and purchase data to trace the impact of such a tax as implemented in Mexico.

Grogger discussed some compelling early findings with MBA students at the Becker Friedman Institute on Nov. 9. His calculations show that soda taxes raise prices enough to cut consumption by a small but potentially significant margin.
 
In Mexico, 73 percent of the adult population is overweight, and 33 percent is considered obese. Making matters worse, fewer citizens suffering from diabetes control their condition with preventive care and blood sugar monitoring, making their overall medical expenses higher. 
 
The Mexican government launched the National Strategy to Prevent and Control Overweight, Obesity and Diabetes in 2014. This includes a whole suite of national programs targeting various trouble spots for overeaters, including taxes on various junk foods, limits on ads for sugary foods aimed at children and a national tax on soda. Sugary drinks seemed a natural target, given that they represent 20 percent of adult energy intake in Mexico; as implemented, the 9 percent tax represented a roughly 1 peso increase per liter of soda. 
 
The policy thinking is pretty simple: raise the price to constrict the demand for it. But Grogger wanted to quantify exactly how much such a tax could lower the demand. How much of that price increase would actually pass through to consumers, as opposed to being absorbed by producers? Would Mexican consumers facing higher soda prices substitute other beverages and limit the overall impact the tax could have on weight loss? 
 
Getting price data was easy—the Mexican Consumer Price Index survey contained all the highly detailed market data Grogger needed. A chart comparing the price jump before and after the tax went into effect really drew his interest. "When I saw this graph was when I got excited about this project. The number of times I've seen [a graph that dramatic] I can count on one hand." 

The problem was the lack of a control product. Since the tax went into effect nationwide all at once, Grogger didn't have data on what soda prices would look like without the tax over the same time period. To compensate, he used a "synthetic cohort approach" wherein a fictional product price index is generated using aggregated marketwide price data from before and after the tax went into effect. Essentially, he constructed a forecast of what soda prices would look like without being taxed, so that he could compare it to what actually what happened. 
 
He compared those results to an ARIMA model aimed at achieving a similar (but painstakingly crafted) forecast. "I don't know how many of you have fit data to an ARIMA model by hand, but there's art to this as well as science," said Grogger. 
 
The synthetic approach pointed to a 14 percent increase in prices under the 9 percent tax; the ARIMA model pointed at a 12 percent increase. There were also no noticeable increases in other beverage prices, meaning the tax didn't simply raise prices across the board. The demand response to that price increase? On average, each consumer decreased soda consumption enough to lose between roughly 2 to 4 pounds, depending on which model is used.
 
That doesn’t sound significant, but other research shows that a Body Mass Index drop of just 5 percent can make a significant difference in health outcomes. It translates to 21 percent fewer new cases of Type II diabetes, 8 percent fewer new cases of hypertension and 8 percent fewer new cases of coronary heart disease and stroke. If Mexico were to target a 5 percent reduction in BMI , the results of the tax would represent roughly 25–40 percent of that goal. 
 
Framed as part of a larger incremental health program, these results are promising, said Grogger. Soda taxes are not the whole solution to the problem of obesity, but they certainly seem to be an effective tool.

—Mark Riechers

This article has been republished with permission from the Becker Friedman Institute for Research in Economics at the University of Chicago, where it originally appeared in a slightly different form.