Misinformation is one of the most discussed issues of the last few years (“misinformation” and “fake news” were declared as the Dictionary.com words of the year in 2018 [Strauss 2018]), having far-reaching consequences in many aspects of society—including the economy. A 2019 report by the University of Baltimore suggests that the impact of fake news on the global economy is about $78 billion annually. Settlements as high as $191 million have been reached in cases filed by the FTC between 2015 and 2020 (FTC 2019, 2020). The FTC states that “an ad must be truthful, not misleading, and, when appropriate, backed by scientific evidence” (FTC 2023). Despite such legal regulations, the prevalence of misleading information supported by competitive brands is commonplace despite inadequate scientific evidence. For example, many sources conclude there isn’t enough evidence to support claims that “natural deodorants” are safer than antiperspirants containing aluminum (Duggal 2020, Palus 2019, Penn Medicine 2019).
A recent Journal of Marketing Research study seeks to understand the impact of misinformation on consumers’ willingness to pay (WTP), approaches to debunking such misinformation, and how debunking attenuates the impact of misinformation. Specifically, the authors attempt to understand
- the effect of misinformation on WTP by creating misbeliefs,
- if debunking can alleviate this effect,
- the differential impact of the source of the debunking, and
- how consumers learn from debunking (i.e., biased or unbiased updating).
The authors conduct two studies, one to understand the effects of misinformation and debunking and the other to understand the heterogeneity in responses across different beliefs. Both studies are large-scale incentive-compatible choice-based conjoint experiments for three product categories (deodorant, toothpaste, nutrition shakes) in which misinformation about a controversial ingredient is prevalent (aluminum, fluoride, and GMOs, respectively). The studies include a total of ~7,000 and 14,000 participants, respectively.
The authors find that misinformation decreases willingness to pay for consumers with correct prior beliefs, whereas misinformation has no significant effect on the WTP of consumers with incorrect opinions. The research also finds that debunking helps reduce the impact of such misinformation across all the product categories, with the most impact on the deodorants with aluminum as an ingredient, wherein there is a robust negative baseline preference for the ingredient. In addition, the studies demonstrate that the regulator is the only source (as opposed to media and competitors) that is effective at debunking misinformation in all three product categories, possibly because of the source’s trustworthiness. The research also suggests that debunking can alleviate the impact of misbeliefs formed before the study on the consumers’ willingness to pay. The authors find that the debunking effect is through correcting the misbeliefs and do not find any backfiring effect.
Misinformation decreases willingness to pay for consumers with correct prior beliefs, whereas misinformation has no significant effect on the WTP of consumers with incorrect opinions.
The following figure lists the impact of debunking the purported harmful effect of three ingredients across the three product categories: deodorants with aluminum, toothpaste with fluoride, and nutrition shakes with GMO. The increase in the WTP due to the debunking of the misinformation by the regulator is large for deodorants, followed by that for toothpaste. The impact of debunking by media is significant for deodorants and toothpaste, whereas that by the competitor is significant for toothpaste and nutrition shakes.
What Types of Consumers Are Impacted Most?
The second set of studies suggests that the debunking impacted a particular set of consumers the most: those with a prior belief that the ingredients were harmful. The debunking works through unbiased learning for these consumers, and the change in belief is the highest for this group. The authors also conduct an analysis of equilibrium strategies by incumbents. The research considers three approaches: coordinated debunking, unilateral debunking, and introducing an ingredient-free product without any debunking action being taken. The simulations suggest that the approach of unilateral debunking is beneficial over no action being taken. However, introducing the ingredient-free product by all incumbent players is the most beneficial approach from the firms’ perspective. Through this approach, the incumbents can expect a profit of about $1.42 for a deodorant, $1.28 for a toothpaste, and $0.25 for a nutrition shake per consumer.
The authors provide additional insights in the following interview:
Q: There are cases where the misinformation about some products or categories is more widespread than others. Do you think the magnitude or severity of misinformation impacts consumer responses (willingness to pay in this case)?
A: Yes, we do see heterogeneity in the effect of misinformation in our study, which is consistent with the extent of misinformation prevalent in the marketplace. For example, misinformation about fluoride decreases WTP on average, but not misinformation about GMOs and aluminum, which is more widespread.
Q: Are there specific characteristics or elements of the debunking messages that have been observed to be more effective? How can marketers optimize the content and delivery of debunking messages for maximum impact?
A: This is a great question. While we did not test for message characteristics in our study, we designed the message to be from three different sources: competitor firms, media, and regulators. We found that debunking by regulators is the only source that has a statistically significant effect for all three product categories, likely because regulators are perceived as the most trustworthy. We also found preliminary evidence suggesting that the novelty of the source predicts for the effectiveness of debunking: In the GMO study, where debunking by competitor firms is much less seen than in the other categories, debunking by competitors appears to be very effective as well.
Q: The research highlights the importance of the source of debunking messages. Given the exponential rise of social media and influencers, how can marketers strategically select multiple debunking sources and combine them together (e.g., messages from the regulator promoted by brands’ social media pages) or use influencers to increase the effectiveness of the debunking?
A: This is interesting but is outside the scope of our study. That is, we did not test how combining debunking sources with influencers can interact with the strength of the debunking messages. On one hand, assuming that debunking messages by influencers is effective, influencers could potentially be used as a way to target certain consumers, in that some influencers may have more followers who believe that aluminum is harmful. Our paper suggests that debunking is effective at correcting misinformed beliefs. But on the other hand, one has to consider whether followers would view the debunking message from the influencer as credible. For example, a debunking message from an influencer who has followers that believe aluminum is harmful may be seen as inconsistent with the influencer’s brand and, therefore, be viewed as less trustworthy. However, this is speculative, as our paper does not test for this. Testing the effectiveness of influencer debunking is an interesting direction for future research.
Q: In the absence of a coordinated effort, it’s best for all incumbents to subscribe to misinformation and create the X-free product, but might that not hamper their own X product/line? How can an incumbent strike a balance between subscribing to these two ideologies?
A: This is a good question. Our simulations account for the potential cannibalization by considering the firm’s total sales from both X-containing and X-free products. We assume that the introduction of X-free products cannibalizes sales from X-containing products to an extent that reflects consumer choices documented in conjoint analysis studies. We do not presume a positive or negative synergy between products subscribing to two ideologies. This phenomenon is evident in the market strategies of brands like Dove, Speed Stick, and Tom’s of Maine, which support the plausibility of our model’s assumptions. These brands validate our approach by successfully marketing both product types simultaneously. Our model’s predictive validity, supported by real-world data, suggests that firms can strategically navigate product cannibalization to optimize their overall product portfolio.
Q: Given that a new brand introduces an X-free product, one option for the incumbent is to introduce an X-free product as well. The other option is to undertake an information campaign, debunking the misinformation. Are there circumstances where you would suggest one approach over the other?
A: When a single incumbent firm undertakes the task of debunking, it bears the cost alone, while other firms gain the benefits of the debunked myths without contributing to the effort. Because campaigns are expensive while information is free to all, there is no incentive for other firms to introduce additional debunking campaigns in the simulation. A similar cost-benefit analysis is crucial when evaluating alternative strategies like introducing an ‘X-free’ product. Given that the existing preference for the X-free product is sizeable, and assuming that the marginal cost for an X-free product is greater than the marginal cost for the X-containing product by a reasonable amount, our simulations show that introducing an X-free product always generates more revenue than debunking the misinformation. However, our lack of access to specific firm cost structures limits our ability to prescribe unequivocally the best strategy. The optimal decision would depend on the precise costs involved in launching an ‘X-free’ product and the potential market share gain from competitors who are invested in debunking efforts.
Q: As a naïve observer, one might expect respondents to be more cautious with regard to ingredients in foods versus personal care, which is observed in the a priori preference of the ingredients in the paper. However, in the ingredient studies, debunking misinformation about GMOs could not correct prior beliefs. Could the prevalence of misinformation across all products in the category be a reason for the subdued effect of the debunking?
A: We do find that debunking can undo the damage caused by both the experimental dose of misinformation and the misbeliefs formed prior to our GMO study. Moreover, for aluminum and GMOs—where most consumers have incorrect prior beliefs—we find that not only does debunking increase WTP but that this increase is significantly greater for those with incorrect prior beliefs than for those with correct prior beliefs.
Read the Full Study for Complete Details
Read the full article:
Jessica Fong, Tong Guo, and Anita Rao, “Debunking Misinformation About Consumer Products: Effects on Beliefs and Purchase Behavior,” Journal of Marketing Research. doi:10.1177/0022243722114708
- Duggal, Neel (2020), “Aluminum-Free Deodorant: Is It Safe?” SpeedStick (accessed November 2023), https://www.speedstick.com/en-us/sweat/aluminum-free-deodorant-is-it-safe.
- Federal Trade Commission (2019), “FTC obtains record $191 million settlement from University of Phoenix to resolve FTC charges it used deceptive advertising to attract prospective students,” Press release (December 10), https://www. ftc. gov/news-events/press-releases/2019/12/ftc-obtains-record-191-million-settlement-university-phoenix.
- Federal Trade Commission (2020), “Legal Library: Cases and Proceedings,” (accessed November 2023), https://www.ftc.gov/enforcement/cases-proceedings.
- Federal Trade Commission (2023), “Truth in Advertising,” (accessed November 2023), https://www.ftc.gov/news-events/topics/truth-advertising.
- Palus, Shannon (2019), “Let It All Sweat Out,” Slate (June 22), https://slate.com/human-interest/2019/06/rise-of-all-natural-deodorants-not-necessary-sweating-antiperspirant.html.
- Penn Medicine (2019), “Is Deodorant Harmful for Your Health?” Health and Wellness (June 6), https://www.pennmedicine.org/updates/blogs/health-and-wellness/2019/june/deodorant.
- Strauss, V (2018), “Word of the year: Misinformation. Here’s why,” The Washington Post, 10.
Go to the Journal of Marketing Research