Consumer Behaviour in the Digital Economy

Vardit Landsman
Close the Plant, Lose the Brand? Sales and Advertising Consequences of Collective Factory Layoffs

[Vardit Landsman; Stefan Stremersch]

Collective factory layoffs occur frequently in Western economies. However, their commercial consequences have received little attention from marketing scholars. Using a difference-indifferences model, estimated on an international data set capturing 74 collective factory layoffs from the automotive industry, this paper reports the following new findings. First, sales of brands that collectively lay off factory workers are, on average, 15% lower in the layoff country during the year following the event than in control countries. Second, the decline in sales is larger for foreign brands than for domestic brands. Third, if the layoff brand states that layoffs are motivated by demand considerations, it undergoes a larger sales decrease than if it states other reasons (e.g., costs). Fourth, a layoff brand’s advertising elasticity increases after the layoffs, as compared to before the layoffs. A survey indicates that these findings extend beyond managers’ present knowledge. The findings may assist managers seeking to implement collective factory layoffs, particularly in their decisions on where a layoff event should take place, and in efforts to prepare for the event’s commercial consequences. 

Simona Botti
Control and Well-Being: Knowing About the Future Can Hurt the Present

The sense of being in control, of perceiving experiences as caused by one’s own deliberate actions and not by external forces, is a basic individual need. In addition to choice and power, information is a source of perceived personal control because it increases the predictability of future events. Consumers who feel in control are usually happier, and when control is threatened they seek to reinstate it. We challenge this notion by proposing that, in certain conditions, having information about how the future is likely to unfold decreases well-being. We investigate both pleasant (i.e., a vacation) and unpleasant (i.e., a genetic illness) upcoming events and study individuals’ preference for and experience with having information about these events, as well as the relationship between information and perceived control.
Box Office, self-reported valence was not and the variance in Box Office explained by self-report measures was relatively low (R2=3%). Adding our neural decoded response measures to this model showed that both decoded valence as well as arousal significantly added to the model and resulted in a significant increase in explained variance (R2=24%).
Our findings show that, using neuroscience methods, we are able to obtain an implicit measure of the emotional experience evoked by marketing stimuli, and that these measures can be used as a neural marker for commercial success. As such, we provide evidence that neural activations reflecting the emotional experience in response to marketing stimuli are related to real-world outcomes, and that such neural measures significantly add to prediction models of choice behaviour in the population at large.

Maarten Boksem
Emotions and commercial success – Looking into the brain

[Maarten Boksem, Hang-Yee Chan, Vincent Schoots, Alan Sanfey, and Ale Smidts]

Neuromarketing, applying neuroscience methods to marketing, has gained considerable popularity in recent years amongst scholars and practitioners alike (Ariely and Berns 2010; Smidts et al. 2014). It is believed that neuroimaging methods could provide marketers with information that is not obtainable through conventional marketing methods. This idea is based on the assumption that people cannot fully articulate their preferences when asked to express them explicitly (Nisbett and DeCamp Wilson 1977), and that consumers’ brains contain hidden information about their true preferences. These issues with self-report data are particularly problematic in the measurement of emotional responses, which have been shown to be important determinants of commercial success and sales. That is, advertisements that manage to evoke the strongest (usually positive) emotions are considered to be the most effective (Batra & Ray, 1986). However, accurately measuring these emotions is not straightforward due to the limited capacity for introspecting on sub-conscious processes. Moreover, reporting on these processes with a high temporal resolution (for example every second for a video stimulus), might actually alter the ongoing emotional processes (Wilson and Schooler 1991).
For the present paper, we extracted the emotional experience (in terms of valence and arousal) from brain activity from subjects viewing movie-trailers, and we subsequently used this decoded emotional experience to predict individual preferences and choices as well as sales in the population at large. We found that the neural measure of valence was predictive of the participants’ individual preferences. In addition, we found that both the decoded emotional valence, as well as arousal significantly predicted choice for movies in the population at large (US Box Office). Importantly, while self-reported arousal was also significantly associated with Box Office, self-reported valence was not and the variance in Box Office explained by self-report measures was relatively low (R2=3%). Adding our neural decoded response measures to this model showed that both decoded valence as well as arousal significantly added to the model and resulted in a significant increase in explained variance (R2=24%).
Our findings show that, using neuroscience methods, we are able to obtain an implicit measure of the emotional experience evoked by marketing stimuli, and that these measures can be used as a neural marker for commercial success. As such, we provide evidence that neural activations reflecting the emotional experience in response to marketing stimuli are related to real-world outcomes, and that such neural measures significantly add to prediction models of choice behaviour in the population at large.

Bart J. Bronnenberg
Changing Their Tune: How Consumers' Adoption of Online Streaming Affects Music Consumption and Discovery

[Hannes Datta; George Knox; Bart J. Bronnenberg]

Constructing a unique panel data set of individual consumers’ listening behavior on digital music platforms, e.g., iTunes and Spotify, we study the effect of adopting streaming services on individual music consumption. Achieving quasi-randomization via a matching procedure, we estimate the changes in quantity, variety and new music discovery after adopting a streaming service like Spotify. Adopting streaming services leads to substantial increases in quantity, variety in any measure, plays of new content, and discovery of new favorites. It is also associated with a large drop in concentration. We document that adopting Spotify leads to more discovery of highly valued music. Relative to using iTunes, adopting Spotify raises repeat listening for consumers’ best new discoveries, although —consistently with the marginal variety on Spotify being free—lowers repeat listening for the average new discovery relative to non-adopters. We discuss the implications for platforms, labels, artists, and consumers.