Increasing Reuse of Eco-bags at a Large Chain Supermarket

Executive Summary

To reduce the use of plastic bags, Nudge Lebanon, in collaboration with a group of students from the American University of Beirut, increased the use of eco-bags among customers who regularly shopped at a large chain supermarket, by sending them weekly SMS reminders. Regular customers who received the reminders were 8 percentage points more likely to reuse their eco-bags compared to the control group. These effects persisted several weeks after receiving the reminders. The SMS reminders also led to a 3.9 percentage points increase in the purchase of eco-bags, relative to the control group. However, this increase didn’t carry on to the weeks following the intervention.


It has been estimated that 500 billion to 1 trillion disposable plastic bags are consumed each year[1], most of which end up in landfills, polluting both our land and water and causing significant harm to our health, as well as the health of wildlife and marine life.

Therefore, in an effort to abolish plastic bags, many countries around the world have either banned their use, or have introduced a bag-tax.  For example, France, Kenya, Taiwan, China, and Australia, among others, have completely banned the usage of plastic bags, while other countries including the UK, Portugal, Greece and some cities in the US like Washington DC have put in place a plastic bag tax.

In Lebanon, most supermarkets and grocery stores still offer disposable plastic bags to their customers for free. While this policy is unlikely to change in the foreseeable future, some supermarkets have been making an effort to promote the use of durable eco-friendly bags instead. Indeed, many customers have purchased such eco-bags with the intention to use them, however, the utilization rate among these customers remain very low.[2]

There are many behavioural challenges that could lead to low utilization rates, including social norms, habit, inconvenience, and forgetfulness due to cognitive overload from everyday life.

In an attempt to address the latter, Nudge Lebanon, in collaboration with a group of students enrolled in a Behavioral Economics course jointly offered with the American University of Beirut, designed two SMS reminders to nudge customers who have previously purchased branded eco-bags, to reuse these eco-bags when they shopped at the supermarket.


A Randomized Controlled Trial (RCT) was conducted in three branches during the months of April and May 2018. The targeted sample consisted of customers who had already purchased an eco-bag during the past six months across the three branches (circa 1,400 customers).

Customers were assigned to a treatment or control group using a stratified randomization, by branch and frequency of visits[3]. This was done to account for any systematic differences between customers who visit a certain branch, and/or with a certain frequency.

The trial involved four phases implemented over a period of seven weeks:

  1. Baseline period: For two weeks, the reuse of eco-bags was tracked across the three select branches, to control for any baseline differences between the control and treatment groups prior to our interventions[4].
  2. Intervention I: For two weeks, the treatment group received weekly SMS messages reminding them to bring their eco-bags when they visit the supermarket, while the control group received nothing – see the message on the left-hand side in Figure 2.
  3. Post-intervention I period: The reuse of eco-bags was tracked for two weeks following intervention I to test for the sustainability of the intervention (i.e. whether the intervention succeeded at creating a habit to reuse eco-bags).
  4. Intervention II: In week 6, the treatment received an additional reminder + a tip to place the eco-bag in a visible place in the car.[5] The intention behind the tip was to reduce the likelihood that customers would misplace their eco-bags.

timeline.pngFigure 1 – Trial Timeline

Accordingly, the treatment group received two types of SMS messages; a simple reminder in weeks two and three, and a “reminder + tip” in week six[6].

SMS message.pngFigure 2 – SMS Messages 


Reuse was tracked by cashiers who inserted a line code into the system each time they observed a customer reusing their eco-bags[7]. The outcome of interest was whether customers reused their eco-bags at least once per week. In addition, we collected data on the number of new eco-bags that were purchased during the trial period.

Reuse of eco-bags

Overall, neither the simple reminder, nor the reminder + tip significantly improved the proportion of reuse among customers receiving them even after controlling for reuse during the baseline period, the branch, the frequency of visits and the period of intervention (B = 0.007; 95% CI = [-0.015, 0.03]; p-value =0.520).[8]

However, when we considered the behavior of customers who visited the supermarket with different frequencies separately (frequent, regular and occasional), we found the following:

  1. Intervention I
    • Regular customers who received the simple SMS reminder were 8 percentage points more likely to reuse their eco-bags compared to the control group (B = 0.08; 95% CI = [ 0.014, 0.14]; p-value = 0.017).
    • This finding was robust to adding covariates (number of bags owned, the frequency of visits, the duration in days since the purchase of an eco-bag and the number of items purchased; B = 0.09, 95% CI = [0.025; 0.15]; p-value = 0.007).
  2. Post-intervention I period
    • The observed effects in intervention I persisted during the post-intervention period (B = 0.07; 95% CI = [0.0026, 0.13]; p-value =0.041). Regular customers in the treatment group were 7 percentage points more likely to use their eco-bags relative to the control group.
  3. Intervention II
    • Regular customers who received the reminder + tip were 7 percentage points more likely to reuse their eco-bags compared to the control group. However, this effect was not statistically significant (B = 0.07; 95% CI = [-0.014, 0.15]; p-value = 0.104) – most likely because of the relatively small sample size.[9]

None of the other groups (frequent or occasional customers) revealed similar effects.

chart.pngFigure 3 – Proportion of Regular Customers Reusing Their Eco-bags

Purchase of new eco-bags

In addition to reuse, we also assessed the impact of the reminders on the purchase of eco-bags. Overall, customers who received the simple reminder (intervention I) were 3.9 percentage points more likely to purchase an eco-bag compared to the control group (B = 0.039; 95% CI = [0.007 ,0.071]; p-value = 0.017). This finding was robust to adding covariates (number of bags owned, the frequency of visits, the duration in days since the purchase of an eco-bag, the number of items purchased; B = 0.035, 95% CI = [0.006, 0.064]; p = 0.018). However, these effects did not carry over to the post-intervention period or intervention II period.


The lack of success in increasing reuse among frequent customers is likely due to the resistive effects of habit, especially since these customers are accustomed to using the free disposable plastic bags provided by the supermarket. Therefore, a weekly SMS reminder might not be strong enough to change such habits.

On the other hand, the success of the reminders in increasing the reuse among regular customers could be due to the fact that these customers visit the supermarket less often, hence the habit of using disposable plastic bags is less rigid, and therefore more prone to change.[10]

This suggests that the role of habit should be considered in attempting to influence people’s behavior, and in bridging the gap between intention and action. This is consistent with Triandis’ Theory of Interpersonal Behaviour which suggests that, in addition to intention, habitual responses have a strong influence on behavior[11].

Furthermore, regular visitors were likely to buy more items than frequent ones. Therefore, it is more worthy for them to bring their eco-bags when shopping. In fact, our analysis revealed a significant positive association between the number of items purchased and the likelihood of eco-bag use.

To conclude, future work should investigate whether sending messages to customers more frequently, and over a longer period of time is more effective at changing habits, especially among frequent customers. In addition, future work should examine whether the timeliness of the SMS reminder matters, by using predictive analytics to detect the time most likely for a customer to visit the supermarket. Finally, the content of the reminders could be modified to test whether different messages or tips can prompt the desired behavior.


[1] Retrieved from

[2] Based on discussions with the management of the supermarket.

[3] Customers were differentiated according to their frequency of visits; frequent customers, who visit at least once a week, regular customers, who visit a least once every two weeks, and occasional customers, who visit at least once a month.

[4] The supermarket did not collect any data on the use of eco-bags prior to this trial.

[5] It is customary for consumers to drive to the select branches of the supermarket.

[6] All SMS messages were sent at 10am on Saturdays as the baseline data showed that the weekends are the most crowded shopping time of the week.

[7] There was no formal mechanism to track reuse of eco-bags. As a result, a new process was introduced by the supermarket for the duration of the trial.

[8] Limiting the analysis to the weekends – following the assumption that the reminders will be more salient and accessible – revealed a positive, but statistically insignificant results (B = 0.034; 95% CI = [-0.008, 0.077]; p-value = 0.119).

[9] Data collection for Intervention II was limited to one week. As a result, it is likely that the intervention was underpowered.

[10] Note that the sample of occasional customers in the time frame of the interventions was too small to detect any significant effects of the reminders.

[11] Triandis, H. C. (1977). Interpersonal behavior. Monterey, CA: Brooks/Cole