New research shows that warning of dangerous side-effects can, over time, lead to higher sales.

This phenomenon, which we term “the ironic effect of warnings”, is predicted by construal level theory—the idea that the more distant we are in time, the more we focus on high-level information rather than on details.

We tested this theory in a series of experiments and found that in the short-term warnings (that cigarettes can cause cancer, artificial sweeteners may damage the immune system, erectile dysfunction medication can result in heart disease and stroke) predictably reduced the appeal of the product and its sales. Over time, however, the content of the warnings faded and fear of side-effects gave way to a feeling of trust in the company that revealed drawbacks of its product. As feelings of trustworthiness increased so did demand for the product.

Our research tested effects of warning of health hazards in four experiments with diverse products.

For each experiment we divided the participants into four groups. Two groups saw an ad for the product that included a warning; one of those groups evaluated the product shortly thereafter (near-future condition) and the other group evaluated it after a delay of either days or weeks (distant-future condition). The remaining two groups saw the same ads but without the warning and, again, evaluated the product either shortly thereafter or after a delay. We tested two types of delay: one was between the time participants ordered the advertised product versus when they received it; the other delay type was between the time that the participants saw the ad versus when they evaluated the advertised product.

Disturbing results

In one experiment involving cigarettes, two of the four groups of participants (all smokers) saw an ad for a new brand, which included a warning of the risk of lung cancer, heart disease and emphysema, whereas the other two groups saw the same ad but without the warning. All four groups were then given an opportunity to buy the product. Half (one group who had seen the warnings and one group who hadn’t) were told they would receive any packs they might order within 24 hours (near-future condition), whereas the others were told that they would receive the packs three months later (distant-future condition). Participants in the near-future condition who had seen the warnings ordered 75 percent fewer packs of cigarettes, on average, than those who hadn’t. In the distant-future condition, however, participants who had seen the warnings ordered 493 percent more than those who saw the same ad without the warning.

In a second study involving an artificial sweetener, two of the four groups of participants were shown an ad for the product that included a warning of potential hair loss, headaches and damage to the immune system.  Two groups (one who had seen the warnings and one who hadn’t) were then asked to place an order for the product. The remaining two groups were contacted two weeks later and allowed to order the product.  Again the backfiring of the warning was extreme.  Compared to those who had seen the ad without the warning, those who had seen the warning and made an immediate order purchased 94 percent fewer packets of sweetener, while those who made a delayed choice ordered 265 percent more.

Similar results emerged in two studies in which participants rated the attractiveness and trustworthiness of erectile dysfunction and hair loss medications: some having seen the warning, some not, in near-future and distant-future conditions.

All experiments demonstrated powerful effects when it came to warnings and risky products.

Message for regulators

Warnings are ever more common and prominent in advertising today. In fact, it was watching strikingly clear and explicit warnings in some television ads, which linked products and such dreadful consequences as aneurysm, cancer and death, that prompted us to wonder how warnings can be spelt out so clearly, yet products still sell so well?

One of the points we make in our paper Warnings of Adverse Side-Effects Can Backfire Over Time is how the effects of warnings may ‘fly under the radar’ when regulators test the effectiveness of a product warning. In such tests, ads are typically shown to people, who shortly thereafter are asked whether they would buy the product.

Because in real life there are often significant delays between reading the warning and making the decision, it’s important for regulators to incorporate this into their research. While this will make their studies more cumbersome, failing to do so may result in incorrect conclusions.

Companies who genuinely wish to warn consumers, should ensure that their warning message is conveyed or repeated shortly before the decision about whether to buy or use the product is made.

Broader implications

This research has implications for the continued use of advertising for controversial and dangerous products like alcohol and tobacco. Common sense suggests that it may be acceptable to allow companies to advertise such products if they’re also warning people about the dangers. Yet our research suggests that not only can ads make these risky products more attractive, but the warnings of the risky side-effects can ironically make them even more attractive. Thus, this research suggests that allowing firms to advertise products whose consumption is associated with significant risks is should be more closely considered.

More broadly, our finding that warnings of adverse side-effects can backfire, has significant implications in areas such as health, finance and law. Depending on timing, informing people of the risks associated with say, a medical procedure or an investment option, may increase the likelihood that they adopt the risky behavior.

Follow us on twitter @INSEADKnowledge or Facebook https://www.facebook.com/Knowledge.insead.