We report on a natural field experiment on quantity discounts involving more than 14 million consumers. Implementing price reductions ranging from 9-70% for large purchases, we found remarkably little impact on revenue, either positively or negatively. There was virtually no increase in the quantity of customers making a purchase; all the observed changes occurred for customers who already were buyers. We found evidence that infrequent purchasers are more responsive to discounts than frequent purchasers. There was some evidence of habit formation when prices returned to pre-experiment levels. There also was some evidence that consumers contemplating small purchases are discouraged by the presence of extreme quantity discounts for large purchases.
Keywords: King; natural experiment; price discrimination; quantity discount; virtual.