Abstract
Trade theory relies heavily on the classic iceberg trade cost assumption, but empirical studies have generally rejected it. These studies, however, overlook variations in product weight. Utilizing detailed trade data, I show that unit weight varies widely even within narrowly defined categories, and increases systematically with unit value. I develop and estimate a model of international transportation that accommodates these patterns. Two remarkable results emerge: First, I find strong empirical support for the iceberg assumption. Specifically, a 10% increase in the price of an item increases the shipping cost by 9.5%. More than 80% of this effect is driven by product-weight. Second, even when facing iceberg (ad-valorem) shipping costs, countries tend to export heavier and higher-priced items to faraway locations. This pattern cannot be explained with the classic Alchian-Allen conjecture. I show that, instead, it is driven by specialization across markups.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have