This paper which just appeared in the journal Preventative Medicine studied the effects of fat taxes, and several types of "messaging" at reducing sweetened soda consumption. The results reveal exactly why we should be skeptical about the obesity-reducing effects of these policies. One should also keep in mind that this study was conducted in several Hospitals - places that employ and house people who one would think might be especially susceptible to attempts to encourage healthier eating.
Here are some key exerpts from the abstract:
This prospective interrupted time-series quasi-experiment included three sites in Philadelphia, PA, Evanston, IL, and Detroit, MI. Each site received five interventions: (1) a 10% price discount on zero-calorie beverages; (2) the 10% discount plus discount messaging; (3) messaging comparing calorie information of sugared beverages with zero-calorie beverages; (4) messaging comparing exercise equivalent information; and (5) messaging comparing both calorie and exercise equivalent information.
The overall analysis failed to demonstrate a consistent effect across all interventions Two treatments had statistically significant effects: the discount plus discount messaging, with an increase in purchases of zero calorie beverages; and the calorie messaging intervention, with an increase in purchases of sugar-sweetened beverages. Individual site analysis results were similar.
The effects of price discounts and calorie messaging in different forms on beverage purchases were inconsistent and frequently small.
Incidentally, the messaging+discount intervention that "worked" was the following:
a 10% price discount on all zero-calorie bottled beverages plus messaging that explained the reason for the discount. Messaging consisted of colorful marketing posters, flyers, and signs displayed prominently in the cafeteria These promoted the 10% price discount with the message, “Lighten up for less – 10% off all zero-calorie bottled beverages and wate
I fully agree with the study's conclusions:
This research augments previous work finding weak, null or even contradictory effects of calorie labeling and price discounts. Our results point to the need for further studies examining the effectiveness of these interventions and their potential moderators. The reality of varying effects in different settings and different populations need be analyzed carefully before contemplating policy interventions such as calorie labeling or sugar taxation in order to avoid ineffective interventions and unintended consequences.