Well, at least about the health effects of smoking.

So say David Cutler and Edward Glaeser, economists from Harvard (paper here). The factor that best explains different smoking levels in the U.S. and Europe is the difference in beliefs about the consequences of smoking:

Public opinion surveys suggest that Americans have some of the strongest beliefs that cigarettes are extremely harmful. Furthermore, there is an extremely strong negative correlation across individuals between beliefs about the harms of smoking and smoking, and a somewhat weaker correlation between the same variables across countries....

A simple decomposition suggests that these belief differences can explain between one quarter and one-half of the difference in smoking rates between the U.S. and Europe. We present some evidence suggesting that these differences in beliefs are themselves the result of concerted government action emphasizing the harms of smoking.

Cigarettes are more expensive in Europe because of higher taxes on tobacco, so if cost were the driving force, Americans would smoke more than Europeans. Higher income levels in the U.S. might also suggest that smoking should be more popular here, as the monetary costs of smoking as a percentage of income would be lower. Then again, high income levels may cause people to quit smoking, as they have more to lose. Income levels have some explanatory value, but they are not the best predictor of smoking levels.

Regulation seems to play a role. Workplace bans raise the costs associated with smoking, and that seems to reduce consumption. But both Europe and the U.S. haves such regulations -- indeed, the authors conclude that Europe is more regulated than the U.S. in this regard, though that seems hard to believe.

In the end, the authors look to beliefs about health consequences as the most important determinant of smoking levels. This sort of evidence is startling:

Both smokers (83 percent) and non-smokers (94 percent) in the US strongly believe that smoking [is harmful to health.] By contrast, 52 percent of German smokers and 84 percent of German nonsmokers shared that belief. Beliefs appear to be specific to the society, much more than to the individual who smokes or not.

So why the difference? Perhaps public health efforts in the U.S. have been stronger than in Europe. The authors provide some interesting history of the U.S. anti-smoking campaign:

Popular knowledge about the harms of smoking almost certainly dates from Readers’ Digest, which ran an article in 1952 titled “Cancer by the Carton.” The news was picked up by other newspapers and media outlets, including even Edward R. Murrow’s (a particularly famous smoker) See It Now television program. These early reports and the related publicity created the first cigarette cancer scare in the early 1950s. The 1950s saw nascent public beliefs form about the harms of smoking. In a January 1954 Gallup survey, 41 percent of people answered ‘yes’ to the question “Do you think cigarette smoking is one of the causes of lung cancer, or not?”

... From 1933 to 1952, cigarette smoking rose every year. Indeed, between 1920 and 1952, cigarette consumption fell only during the bleakest years of the great depression (confirming the positive income elasticity of cigarette consumption at lower income levels). Between 1952 and 1954, cigarette smoking took its first dramatic drop. From 1952-1953 smoking dropped by 3 percent, followed by an additional 6 percent the following year. While it is possible that this drop was due to something other than changing beliefs about the health risks of cigarettes, contemporary observers certainly thought that the decline in smoking was the result of the health scare. For example, the treasurer of the American Tobacco Company said in 1954 that “there is a tendency to ascribe the drop in cigarette consumption almost entirely to the so called ‘cancer scare’.” (New York Times, May 7, 1954, p.35).

The reaction in the marketplace proceeded along several dimensions. On the one hand, manufacturers and the public responded by making cigarettes somewhat safer. Use of filtered cigarettes rose from less than 2 percent in 1952 to more than 20 percent in 1955. Edward R. Murrow on the air linked the rise of filtered cigarettes to heightened fears about the dangers of cigarettes. On the other hand, the cigarette industry fought back through advertising. Using their vast advertising budgets and spending on their own rival research (which unsurprisingly found that cigarettes were harmless), cigarette companies were able to overcome the negative publicity associated with these early studies. Smoking rose again from 1954 through 1963.

Increasing evidence in the medical community showed the harms from smoking, and in 1964, the Surgeon General issued his famous warning about the health consequences of smoking. In 1966, the Federal Trade Commission required cigarettes to be sold with a label warning that “cigarettes may be hazardous to your health.” Both the federal government and private groups like the American Cancer Society mounted campaigns meant to increase awareness of the health consequences of smoking. The fruits of this campaign are apparent in public opinion surveys. In 1960, 50 percent of Americans believed that cigarette smoking was one of the causes of lung cancer. By 1969, the share was 71 percent (Cutler and Kadiyala, 2003). One sees this in the consumption data as well. In 1964 alone, cigarette smoking fell by 3 percent, and smoking was down by 8 percent by 1970.

Since the 1950s, beliefs about the harms of smoking have cemented. The Gallup Organization (Gallup, 1981) documented a rising belief that smoking was dangerous. By the 1980s, Viscusi (1992) finds that people actually overestimated the health risks of smoking. As noted above, over 90 percent of Americans now believe that smoking causes cancer. A strong circumstantial case links the decline of smoking to the expansion of information about the harms of smoking. [D]ata on the share of people who have ever smoked show a decline beginning in cohorts coming of smoking age after 1964. Other evidence about the link between perception and smoking comes from individual correlations between beliefs and actions. Smoking rates among those who believe that smoking a pack or more of cigarettes per day has a great risk are only 23 percent, compared to 52 percent among those who do not believe that link. If beliefs causally affect smoking (and not the reverse, as with cognitive dissonance), it suggests a large impact of beliefs on actions.

All of this makes me wonder about the obesity problem in the U.S. The authors have another (older) paper on that. Blame your freezer, microwave, and dishwasher, they say: "technological innovations which made it possible for food to be mass prepared far from the point of consumption, and consumed with lower time costs of preparation and cleaning." Anti-obesity public health campaigns are more recent than anti-smoking campaigns, but the link between education and behavior is encouraging.

Thanks to NPR's "Talk of the Nation" for the tip.

Europe | Bookmark

TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8345157d569e200d83488842f53ef

Links to weblogs that reference Europeans are Ignorant: