In this paper I suggest a unified explanation for two puzzles in the inventory literature: first, estimates of inventory speeds of adjustment in aggregate data are very small relative to the apparent rapid reaction of stocks to unanticipated variations in sales. Second, estimates of inventory speeds of adjustment in firm-level data are significantly higher than in aggregate data. The paper develops a multisector model where inventories are held to avoid stockouts, and price markups vary along the business cycle. The omission of countercyclical markup variations from inventory targets introduces a downward bias in estimates of adjustment speeds obtained from partial adjustment models. When the cyclicality of markups differs across sectors, this downward bias is shown to be more severe with aggregate rather than firm-level data. Similar results apply not only to inventories, but also to labor and prices. Montercarlo simulations of a calibrated version of the model suggest that these biases are quantitatively significant.