Online Appendix for Approximating the Cost-of-Living Index for a Storable Good

The time-series patterns of price and quantity shown in Figure 1 of the main paper are consistent with consumer stockpiling of canned tuna and canned soup: when product prices fall for a short time, quantity spikes to a level much higher than the baseline level that is observed at regular prices. In this section I present some additional evidence that patterns of consumer purchases are consistent with stockpiling behavior. To do this I perform a regression analysis following Hendel and Nevo (2006) using individual household purchases. I regress the total quantity a household purchases during a given week on a measure of the household’s inventory, the price, a dummy variable for whether a sale occurs, the interaction of that dummy variable with price, and feature and display variables. Household inventory and consumption are both unobserved, so I compute a proxy for inventory under the assumption that a household’s consumption rate is constant over time. I estimate a household’s consumption rate by dividing the total quantity consumed by the number of weeks the household is observed. One can construct an estimate of the household’s inventory during a given week as the sum of quantity purchased prior to that time, minus total consumption, plus initial inventories. Initial inventories are also unobserved, so a household fixed effect is included to control for them. Note that in the Nielsen data, household store visits where no purchases were made were only tracked from the middle of 1986 onwards. Due to this limitation the regressions only use the final year of data for the canned tuna category, and the final 2 years of data for canned soup. Regression results for both categories are shown in Table A1. In both categories, the impact of inventory is negative and significant, indicating that when a household’s inventory increases, the quantity purchased decreases, consistent with stockpiling behavior. Turning to the three price interaction variables, in the canned tuna category less is purchased for higher prices, and when a product goes on sale, more is purchased. Furthermore, when a product is on sale the quantity purchased is higher, and the The Nielsen data tracks household purchases for the entire 3 years of the data, but it is not known if a household visits a store but makes no purchase prior to the middle of 1986.