The HBR article Reusable Bags Make People Buy Organic—and Junk: An Interview with Uma Karmarkar makes interesting reading. The same study is also quoted in a recent Economist article: Eco-waverers: When people feel good about themselves, they do bad things .
The study highlights the fact that we like to reward ourselves for good behavior and often that reward is contrary to the positive benefits of the good behavior (a few scoops of ice-cream after an extra mile on the treadmill- we've all been there).
Uma Karmarkar's (@uma_karma) study finds that people who take reusable bags to their grocers also buy more junk food. This comes about as a result of the "feel good" feeling they get from using reusable bags.
In my opinion what would be interesting is to see if the behavior changes if the perceived root cause goes away. Will the shoppers be as likely to "reward" themselves with junk if there was no feel good feeling associated with bringing their own bag ?
Another point is that while the limited facts about the study shared in the interview indicate a "correlation" between purchase of junk food and bringing their own bags, I am not so sure that "causation" has been clearly established - bringing own bag led to purchase of more junk food.
It could be possible that the basket of purchases for the individuals in question skewed towards junk foods even when they were not bringing their own bags in. When the historical purchase behavior is studied then only a more definitive conclusion that the bag being brought in is influencing the purchase decision can be drawn.
Right now the facts as presented in the HBR article are equivocal and not conclusive enough to draw the inferences implied in the article.
I asked these questions to Uma and here's what she had to say:
I think the interviewer for the HBR article just cherry picked aspects of the study and so the correlation vs. causation question is not adequately answered in the piece.