The global soda tax experiment

They’re cloyingly sweet, nutritionally empty — and, increasingly, subject to taxation. More than 35 countries and seven cities in the US — starting with Berkeley, California, in 2015 — now impose a tax on soda and other sugar-sweetened beverages, and several more places are considering it.

Public health researchers and organizations such as the American Heart Association and the American Academy of Pediatrics see these taxes as low-hanging fruit in the battle against obesity and the health problems such as diabetes that often come with it. In the United States, nearly 40 percent of adults are obese, which adds $147 billion to the nation’s annual healthcare spending, according to the Centers for Disease Control and Prevention. The problem is complex, but the widespread consumption of foods packed with added sugars — which add calories but no essential nutrients — plays a major role, and beverages account for nearly half the added sugar in the American diet.

“It’s really hard to shift these behaviors, and taxes are, if not the single most, one of the most impactful and important policies to move the needle on unhealthy eating habits,” says Christina Roberto, a behavioral scientist at the University of Pennsylvania in Philadelphia. Taxes have helped to reduce the public health impact of alcohol and tobacco, and many public health researchers say there’s good reason to think they can mitigate the harms of sugary beverages, too.

At the same time, there are also reasons why soda taxes might not have the impact on public health that advocates hope for. The current taxes may be too low to affect purchasing behavior. People could switch to other unhealthy foods. Or, in some cases, they could simply buy their soda in a neighboring city that doesn’t tax them.

Definitive answers won’t come fast: Chronic conditions like obesity and diabetes take years to develop, and so, too, will any health benefits resulting from a new tax. But an emerging body of research suggests that beverage taxes have already reduced consumption of sugary drinks in some communities — an encouraging and essential step.

Taxing bad habits

The use of taxes to compel people to make healthier choices has a long history with tobacco and alcohol, which are taxed by nearly every country in the world. “There’s decades of work now on tobacco, hundreds of studies from around the world, showing that if you raise prices you induce adults to quit smoking and prevent kids from taking it up,” says Frank Chaloupka, an economist at the University of Illinois at Chicago. Research has linked higher cigarette taxes to reduced mortality from throat and lung cancer and other respiratory diseases, Chaloupka and two coauthors wrote earlier this year in the Annual Review of Public Health. Other studies have linked higher taxes to lower rates of hospitalization for heart failure and lessened severity of childhood asthma.

With alcohol, it’s more like dozens of studies, but the conclusions are similar, Chaloupka says: Alcohol taxes have been linked to lower frequency and intensity of drinking and reductions in unhealthy consequences of alcohol abuse, from cirrhosis of the liver to motor vehicle injuries to alcohol-related violence. The higher the tax, as a rule, the greater the impact.

Sugary beverages may seem more innocuous than cigarettes and alcohol, but there’s strong evidence tying them to a host of chronic health problems, says Barry Popkin, an economist and nutrition researcher at the University of North Carolina, Chapel Hill. Sugary drinks cause sharper spikes in blood sugar than most types of food, studies find. Over time, they may be more apt to disrupt the body’s insulin regulation. And sugar dissolved in a drink doesn’t trigger the brain’s satiety mechanisms the same way that sugar in solid food does. As a result, “what we’ve learned in the last 20 years is that what you drink doesn’t affect what you eat,” Popkin says.

Those extra liquid calories (roughly 250 in a 20-ounce bottle of many popular sodas, or 10 percent of the recommended daily total for an adult male), add up. Studies by Popkin and others have linked habitual consumption of sweetened beverages to an elevated risk of weight gain, obesity, type 2 diabetes, cardiovascular disease and other health problems. A 2010 meta-analysis of previous studies that tracked a total of 310,819 participants, for example, found that people who drink one or more sugary drinks a day have a 26 percent higher risk of developing type 2 diabetes than those who drink no more than one sugary drink per month.

This research has focused on beverages containing calorie-adding sweeteners such as sucrose (table sugar) and high fructose corn syrup — not just sodas but also sports and energy drinks, fruit juices with added sugar, and sweetened coffee and teas. There’s less research, and more expert disagreement, on the health effects of pure fruit juice (which can contain as much sugar per serving as soda, but has vitamins and other nutrients too) and beverages with artificial sweeteners that don’t add calories.

Sugary beverages certainly aren’t the only culprits. Sugary foods are, too, but they’re more difficult to define and regulate, says Kristine Madsen, a pediatrician and research scientist at the University of California, Berkeley School of Public Health. “If you start getting into foods that could be classified as junk food you get into huge debates,” she says. Take granola bars. Some are loaded with fat and sugar — essentially cookies masquerading as health foods. Others might be packed with nuts and dried fruit and contain little added sugar, making them legitimate sources of protein and dietary fiber. But a typical beverage with added sugar has no nutritional value, Madsen says. “There’s nothing it adds to someone’s diet that benefits them.”

The idea behind sugary-beverage taxes is rooted in basic economics: Raising the price on a product tends to discourage people from buying it, especially if it’s not something they deem essential in the first place. One encouraging sign for soda taxes, Chaloupka says, is that economists find that the price elasticity for sugary beverages — that is, the degree to which people respond to price increases by reducing their purchases — is at least as great as it is for alcohol and tobacco.

In wealthier countries, that sugary-beverage price elasticity averages about -0.8, meaning that for every 10 percent increase in the price of soda, purchases decline by 8 percent. (Price elasticity averages about -0.4 for tobacco and ranges between -0.5 and -0.8 for alcohol.) Not surprisingly, people with less money tend to be more sensitive to price increases, and research in lower-income countries and communities reports even higher price elasticity, so that a 10 percent price increase results in more than a 10 percent reduction in purchases.

Public health researchers and economists bored into these data and more at a 2015 meeting convened by the World Health Organization to review soda tax research and make recommendations. Along with price elasticity, the experts considered actual purchase data — what little were available at the time — from countries where taxes had been implemented, along with a small number of computer modeling studies estimating how calories saved from reduced soda consumption might translate to reduced risk of obesity and diabetes. The WHO’s resulting report acknowledges the need for more research, but it concludes that taxes of 20 to 50 percent are most likely to be effective, based on the available evidence.

That’s in the same ballpark as existing taxes on alcohol and tobacco, note Chaloupka and colleagues. Alcohol taxes range from 0.3 percent in Kyrgyzstan to 44.9 percent in Norway, with an average of 17 percent worldwide. Tobacco taxes average 48 percent in high-income countries and 32 percent in low- and middle-income countries.

Only a few countries have levied beverage taxes at the higher end of the WHO’s recommended range: Saudi Arabia and the United Arab Emirates levy a 50 percent tax on sweetened beverages, for example, and a 100 percent tax on energy drinks. (The goal in Saudi Arabia was raising revenue, not improving public health.) In other places it’s more complicated.

A few countries, including the United Kingdom and South Africa, have implemented tiered or graded beverage taxes that increase with sugar content. In the UK, where the nationwide tax went into effect in April 2018, several beverage manufacturers responded by reformulating their drinks to contain less sugar (adding artificial sweeteners, at least in some cases), thereby avoiding the highest tax rate. (Coca-Cola refused, deciding instead to reduce serving size and pass some of the tax to consumers.) The impact on sales, not to mention public health, remains to be seen.

In the US, beverage taxes range from 1 to 2 cents an ounce. Structuring a tax this way makes it easy to implement, but it means that the percentage of the price increase varies for different products

Researchers who support the taxes acknowledge that such small price increases are unlikely to dissuade occasional soda drinkers, but those aren’t the people at greatest risk. The hope is that taxes will make a dent in consumption by people with more serious habits — such as the 5 percent of Americans who report drinking roughly 600 calories worth of sugary beverages (more than four 12-oz cans) on any given day.

Soda studies

One of the best-studied taxes is in Mexico, which in January 2014 became the first country in the Americas to implement a significant sugary-beverage tax. Like many middle-income countries, Mexico has seen the health risks associated with overconsumption surpass the health risks of undernutrition. Roughly two-thirds of Mexicans are overweight or obese, and diabetes has become the country’s leading cause of death and disability.

The Mexican tax adds one peso per liter to the price of all beverages with added sugar. That typically works out to about 10 percent, says Arantxa Colchero, a health economist at the National Institute of Public Health in Cuernavaca who has studied the tax. Drinks with artificial sweeteners are excluded, as are pure milk and fruit juices, but unlike many places, Mexico taxes milk and yogurt drinks with added sugar. (Elsewhere, policymakers have decided that the benefits of getting children to drink milk outweigh the downsides of added sugar in beverages like chocolate milk — a point of debate among public health researchers.)

To assess sugary-beverage purchases before and after the tax, Colchero and colleagues used a nationwide survey of more than 75,000 Mexican households. According to their analyses, purchases dropped 6 percent in the tax’s first year, and more in households that were low-income, had children or were heavy consumers to begin with. Bottled water purchases, on the other hand, increased 16 percent — an encouraging sign, Colchero says, that people were switching to a healthier alternative. A follow-up study using additional data found similar effects, and suggested that the drop in sugary-beverage sales grew to nearly 10 percent in the second year of the tax.

Can such modest decreases translate to better health? Computer modeling studies based on the Mexican purchase data suggest that they could. In one study, researchers used a simulation to predict the prevalence of cardiovascular disease and related conditions. The model was developed using the Framingham Heart Study in the US, which uses public health data on age, sex, smoking, body mass index and more to predict cardiovascular health trends, but the scientists plugged in Mexican public health data wherever available.

That study predicted 189,300 fewer new cases of type 2 diabetes and 20,400 fewer heart attacks and strokes over a 10-year period, assuming a sustained 10 percent decrease in sugary-beverage consumption in Mexico (and estimating that people would make up 39 percent of those lost calories elsewhere in their diets). “The impacts would be much higher if the tax was 20 percent,” says Colchero, who was not part of that study but collaborated on another study that also predicted substantial reductions in diabetes resulting from the tax.

The second modeling study also estimated the impact of the tax on Mexico’s obesity rate by converting numbers on reduced soda purchasing to calories saved, and using a computer model to predict changes in body mass index. After 10 years with the current tax, the scientists predicted that Mexico’s obesity rate would drop 2.5 percent, potentially corresponding to several million fewer obese people.

Both modeling studies suggested that doubling the tax would roughly double the public health benefits. The Mexican legislature is considering legislation that would do that.

In Berkeley, which implemented a penny-per-ounce tax on sweetened beverages in 2015 — the first such tax in the US — researchers have also seen reduced beverage purchases. One study examined millions of checkout scanner transactions for two supermarket chains in the area and found a 10 percent drop in sales of the taxed beverages. Sales of bottled water, which isn’t taxed, rose 16 percent during the same time period; sales of untaxed vegetable, fruit and tea drinks rose 4 percent.

A recent study from Philadelphia found an even greater reduction in sugary-beverage sales. That city’s beverage tax went into effect in January 2017 — to evaluate it, behavioral scientist Roberto and colleagues used a dataset of sales at supermarkets, pharmacies and big-box stores like Walmart. Sales of sweetened beverages dropped 51 percent the year after the tax was implemented, the team reported in May in the Journal of the American Medical Association. Sales in Baltimore, a nearby city with similar demographics and no beverage tax, remained flat during the same period, suggesting that the tax was responsible for the drop, as opposed to some regional trend or societal shift.

About a quarter of that decline was offset by an increase in sales in three surrounding zip codes, suggesting that some people were willing to drive across the city line to get their soda, or at least pick some up when they were passing through. But even factoring in that cross-border shopping, Philadelphia has seen a 38 percent decline in the purchase of sweetened beverages, the researchers conclude. That’s equivalent to an annual reduction of 78 million 12-ounce cans of sugary drinks, or 49 cans per person in a city of 1.6 million.

Several factors could account for the larger drop in sales in Philadelphia compared with Berkeley, Madsen says. Philadelphia’s tax is greater (1.5 cents per ounce, versus 1 cent per ounce in Berkeley) and its population is poorer, on average, and so might have felt more of a pinch from the price increase. In addition, Berkeley residents drank relatively little soda to begin with. “It is harder to see a large drop in sales if you start with low baseline sales,” Madsen says.

Other researchers also have found evidence that Philadelphia’s beverage tax is changing consumer behavior. “All these studies use different datasets, but the nice thing is we’re getting some confirmation,” says John Cawley, an economist at Cornell University. Cawley and colleagues surveyed hundreds of Philadelphians before and after the tax was implemented, initially approaching people as they exited stores to ask about their purchases, then following up by phone with more detailed questions.

Adults who participated in the study reported drinking about 10 fewer sodas a month after the tax, amounting to a reduction of about 31 percent, according to a study recently published by Cawley and colleagues in the Journal of Health Economics. The study also provides the first data on how beverage taxes affect children, Cawley says. The Philadelphia tax did not reduce soda consumption by children as a whole, the researchers found, but it did reduce consumption among those who were frequent soda drinkers to begin with.

Healthy outlook?

Despite the growing evidence that beverage taxes reduce sales, there is so far no direct evidence that the taxes have the intended health effects. Gathering such evidence won’t be easy. Ideally, researchers would like to monitor the health of a representative group of people before and after the tax, says Lisa Powell, a health economist at the University of Illinois, Chicago. “You need to plan those studies and recruit people well in advance of the tax and track them over time, which is extremely expensive to do,” she says. So far this has not been done, although Roberto has applied for funding for a study that would use electronic health records for thousands of patients in the University of Pennsylvania hospital system to look for changes in body mass index, and possibly indicators of diabetes, before and after the enactment of the Philadelphia soda tax.

The alternative, looking for changes in the overall population — say, in the prevalence of obesity or diabetes — requires more data and more sophisticated statistics. Powell and other researchers suggest that 10 years would be a reasonable time frame to expect to see a payoff in reduced rates of diabetes and cardiovascular disease. That’s about how long it took for lung cancer rates to drop after states started implementing tobacco taxes, Popkin says. “We didn’t have the hard biological health outcomes for a long time,” he says.

In the meantime, the taxes are raising significant revenue. The seven US cities with beverage taxes currently raise a total of $133 million per year. Although not all of those taxes were passed as public health measures, most of the revenue goes to improve community welfare in some way. Exactly where the money goes depends on local politics and perceived needs in the community. In Philadelphia, for example, the tax was passed as a means to raise money to expand early childhood education. In Berkeley, the money has gone to local organizations that promote nutritional education and exercise, including the Edible Schoolyard project initiated by restaurateur Alice Waters to build kitchen gardens at middle schools to teach children about food and nutrition.

In Seattle, which implemented a 1.75-cent-per-ounce soda tax in 2018, the revenue has been used for a variety of programs aimed at improving health equality, such as subsidizing fruit and vegetable purchases for low-income people, says Jim Krieger, a former chief of chronic disease prevention for the city and executive director of Healthy Food America, a research and education nonprofit. Partly as a result of targeted marketing by beverage companies, Krieger says, low-income communities have higher rates of sugary-beverage consumption and higher rates of disease associated with those drinks. “The tax revenue is being invested where it will do the most good in relation to the harms being caused by the sugary drinks.”

Culture shift

The beverage industry is strongly opposed to these taxes. In 2016, it spent $30 million in California alone to oppose new ballot measures to impose beverage taxes in Oakland and San Francisco (both passed). Industry-funded ads present the taxes as attacks on consumer freedom, unfairly burdensome to low-income people, and bad for employment and the overall economy. Studies by independent researchers in Philadelphia and Mexico have found little or no evidence for negative economic impacts.

The industry has lobbied effectively for state laws banning new local beverage taxes. Michigan passed the country’s first such law in 2017; Arizona, California and Washington followed suit in 2018. The California law leaves in place the existing beverage taxes in Berkeley, Oakland, Albany and San Francisco, but it upended plans to put soda taxes on the ballot in at least two other cities, Santa Cruz and Richmond. In the face of industry opposition, the California legislature in April shelved discussions of a bill that would impose a statewide beverage tax.

If the goal is improving public health, taxes that cover a larger geographic area would be advantageous, Cawley and colleagues write in a recent paper in the Annual Review of Nutrition. “Optimally, this would be something that’s happening not at the city level but at the state or national level, so that there’s less incentive to just drive a mile or two to evade the tax,” Cawley says.

Public health researchers who advocate for the taxes see them as just one part of a larger strategy to tackle obesity and diabetes. Several countries are trying a more comprehensive policy approach. In Chile, which has the highest obesity rate in Latin America and has led the world in sugary-beverage sales per capita in recent years, lawmakers have passed a suite of policies since 2012 that include a small sugar-sweetened-beverage tax, warning labels on foods with high levels of added sugar (similar to the labels on cigarette cartons warning of the health risks of smoking), bans on sugary beverages in schools, and limits on marketing foods and beverages with added sugar to children. “The more comprehensive you can make the laws, the bigger health effect you’re going to have,” says Popkin, who has advised the Chilean government on these policies.

But in addition to new policies, what has to happen is a cultural shift, says Laura Schmidt, public health researcher at the University of California, San Francisco. “With tobacco, the number one thing that made the difference was norms,” she says. “The policies and the debate and the education campaigns made smoking unpopular.”

Countermarketing — media campaigns that undermined tobacco company advertising by pointing to negative health effects or industry manipulation of consumers — may have played a role as well. It’s a strategy already being tried with sugary beverages, for example with the “Berkeley vs. Big Soda” campaign launched in 2014 to counter industry-funded ads trying to prevent voters from approving the tax there, and New York City’s “Pouring on the Pounds” campaign, which emphasized the connection between sugary beverages and weight gain (one ad, for example, showed a man opening a soda can and pouring out chunky, gelatinous fat).

Cultural changes may be underway in the US, where consumption of added-sugar beverages has been declining steadily since the early 2000s. One study, based on nationally representative data from the CDC, found that the proportion of American adults who reported drinking at least one sugary beverage a day dropped from 62 percent to 50 percent between 2003 and 2014 (and from 80 percent to 61 percent for children).

With additional nudges from soda taxes and other policies, advocates say, that decline could develop into significant health benefits in the years ahead. And as public perception shifts, lawmakers will feel emboldened to pass more aggressive policies, Schmidt says. “It’s a virtuous cycle.”

Editor’s Note: A graphic in this article was updated on October 22, 2019, to correct an error. Morocco is part of Africa, not Europe as originally stated.