BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Sir, Perhaps Some Perrier In Your Benzene?

This article is more than 4 years old.

(Part 2 of 2 -- inspired by yesterday's column on Consumer Reports' spurious report on arsenic in bottled water -- which still can't compete with the 1990 Perrier scare, the all-time case study in how businesses can be knocked silly by media math-mangling.)

On February 10, 1990, the front page of the Saturday New York Times reported on a slew of stories gripping America. Negotiations with the Soviet Union over the possible reunification of West and East Germany. Gangster John Gotti getting acquitted yet again. Ronald Reagan testifying about his role in the Iran-Contra scandal.

These were mere aperitifs, though, compared to the most arresting headline of all.

The New York Times

Perrier sparkling water had spent the 1980s becoming the hip beverage of yuppies nationwide. The drink’s marketing promoted – flaunted – that it was absolutely “pure,” bottled straight from an underground mineral spring in southern France. Pretentious commercials showed glistening glasses of delicate bubbles wafting through the voiceover: “It’s perfect. It’s Perrier.”

Well, not quite perfect. In early 1990, during some routine quality checks, North Carolina health officials examined a few Perrier bottles and discovered elevated levels of benzene, a naturally-occurring liquid that nonetheless, all things being equal, one would prefer not sip over lunch. Benzene makes for nice varnish, gives gasoline its freshly-pumped bouquet, and, you guessed it, increases risk of cancer. That being said, traces of benzene exist in many liquids – including natural water. The U.S. Environmental Protection Agency says that benzene levels in drinking water should not exceed 5 parts per billion, so that “no adverse health effects are likely to occur … with an adequate margin of safety.”

[Note how that statement has even more elevated levels of numerical squishiness. If the goal is “no adverse health effects are likely” -- and "likely" means "probable," more than 50 percent -- does that green-light some serious risks to still be 1-in-6 shots? And what is an “adequate margin,” anyway? Thanks, E.P.A.]

The bad news was that Perrier’s levels really were roughly 3 times the E.P.A.’s guidelines. The good news? Those standards referred to drinking water exposure over a lifetime -- like more than a quart every day for 50 years. Popular as Perrier was, no one subsisted on the stuff.

Perrier exposed no one to any health risk. Not one. The only harmful exposure, turns out, was to the company -- from the predictably breathless media hysteria. Newspapers all over America did report the key number correctly: the benzene levels were 3 times what they should have been. Three. What that number connoted would, as always, be determined by the words people placed around it.

Headlines and articles turned that 3 into Perrier being “contaminated,” “laced,” and “tainted” with benzene. Benzene was proclaimed at various points as a “component of crude oil,” “cancer-causing,” and “flammable poison” (as if “poison” weren’t sufficient). The Los Angeles Times invoked a term even more toxic than benzene -- Tylenol, capsules of which had recently killed seven people in Illinois after some psychopath laced them with cyanide.

Perrier had no choice. It issued an immediate recall of every bottle in the United States. Store owners cleared the shelves; restaurants dumped cases at a time; Beverly Hills hotels had maids remove both bottles from every in-room refrigerator. In all, 72 million bottles were destroyed in a matter of days, lest America turn into one walking carcinoma.

What did the U.S. Food and Drug Administration have to say about this? How did America’s primary hey-don’t-eat-that agency address this “scare” that posed such a “danger” to the American public? Acting F.D.A. commissioner James Benson stepped to the press-conference microphone.

“If I had a bottle in the refrigerator,” Benson said, “I would drink it.”

Benson strained to explain how “3 times,” however accurate, still put no one at any meaningful risk. He presented other numbers to explain the triviality of that 3: Someone needed to drink a pint of Perrier a day for 70 years to raise their risk of cancer by only 1 in 1 million -- or roughly the chance of their getting killed by a meteor or asteroid, according to NASA. Other officials noted that Perrier’s benzene levels were still lower than those in coffee or apple juice.

Too late. Perrier ultimately issued a worldwide recall and had to halt production for three months, costing it an estimated $40 million. Its internal investigation found that a filter hadn’t been changed properly; they promised it would get replaced more often. The company avoided mentioning that Perrier always had, and would continue to have, some benzene -- because that’s just inevitable with natural spring water.

Perrier did bring this on itself with its own language. Before others exaggerated with words like “contaminated” and “tainted,” conveying danger that didn’t exist, Perrier had sold its water with equal hyperbole. “Perfect”? The company might as well have taped a “Sue Me” sign on its back. (And, right on cue, lawyers filed suit against Perrier for misrepresenting its product and causing – this is priceless – “anguish” among its devotees.) “Pure” is slightly more fuzzy. There are different standards. Ivory Soap advertised itself throughout the 20th century as “99 44/100% pure,” dazzling its customers who remained unconcerned with what in the world that last 56/100% of impurity could be. (Perhaps impurity can be its own selling point -- one can imagine a gas-station energy drink with the tagline, “Now with more benzene!”)

Then again, no one other than foul-mouthed children ingests Ivory Soap; with Perrier, the stakes were higher. The “pure” Perrier promised was a word just begging for mathematical debate. Did it mean a literal 100 percent, meaning that the water had been run through so many filters that not one non-H20 molecule remained? Pretty unrealistic. A better word would have been "authentic". Water comes out of the ground, for crying out loud.

The lesson? Choose your number words carefully. Whatever you sell, don't call it "perfect". Even if people expect you to be.

 

Send me a secure tip