Essays on Price and Time in Trade and Household Production

This dissertation consists of three chapters that estimate the elasticities regarding price and time in trade and household production. Chapters 1 and 2 estimate price elasticities. Chapter 1 estimates the one-factor-one-price elasticity of substitution (OOES)—or how the percentage change in the quantity of one good responds to the percentage change in the price (of itself or another good)—in an international trade context. Chapter 2 estimates the two-factor-one-price elasticity of substitution (TOES)—or the difference of percentage changes between two quantities with respect to the percentage change in the price of one good—in the context of household food production. Chapter 3 estimates the elasticity of export quantity and value with respect to delays in the time it takes to load or unload products at US ports. Chapter 1 estimates the price elasticities in agricultural trade. Armington elasticities, the elasticity of substitution between goods from different countries, are key parameters in agricultural trade policy evaluation and welfare calculation. We estimate Armington elasticities for a selected basket of 38 agricultural commodities in 5 categories by compiling a sample of 118 countries' production and trade flows. Following and extending Feenstra et al. (2018), we estimate both the micro-elasticity of substitution between foreign sources of imports and the macro-elasticity of substitution between home and imported products at the commodity level. The median of the micro- and macro-elasticities are 6.4 and 5.0, respectively. Meat products have the lowest micro- and macro-elasticities, with the micro-elasticities ranging from 4.2 (pork) to 5.0 (poultry) and the macro-elasticities ranging from 2.9 (pork) to 4.5 (beef). Crops products have the widest range of Armington elasticities, with micro-elasticities ranging from 2.5 (pigeon peas) to 90.3 (peanuts), and macro-elasticities ranging from 1.2 (pigeon peas) to 20.1 (peanuts). In line with the literature, we find that 75 percent of the agricultural commodities have numerically smaller macro-elasticities than micro-elasticities, even though only 6 of them (pork, poultry, corn, peanuts, apples, and peppers) are statistically smaller at the 5 percent level. We explore the robustness of our estimates by slicing the sample into separate periods and importing countries. Finally, we discuss the policy implications of our estimates on predicting trade due to tariff changes and understanding welfare gains from agricultural trade. Chapter 2 estimates the goods-time elasticity of substitution (EOS), the responsiveness of the difference between money and time in household production for change of opportunity cost of time (OCT). This chapter bridges the gap between literature that directly and indirectly estimates the goods-time EOS in household production. Inspired by the studies in environmental economics, we argue the opportunity cost of time in household production not only depends on wage but life-cycle dynamics and household demographics as well. We proceed with the estimation by two strategies: direct estimation of the household production, and the demand-supply approach borrowed from Feenstra's (1994) research on trade elasticities. Both strategies report the estimates are much larger than unit and closer to previous indirect estimates. We show our results are robust when applied to Aguiar and Hurst's (2007) sample, in which they employed the indirect estimation. The larger goods-time EOS indicates policies aiding households with money for groceries like the Supplemental Nutrition Assistance Program (SNAP) are more sufficient, since money for certain groceries can more easily substitute for time in making meals. Chapter 3 explores the elasticity of trade with respect to port congestion time. U.S. ports have struggled with significant supply chain congestion during the past two years. Anecdotal evidence shows the increasing port congestion brought substantial losses to U.S. exports, particularly agricultural shipments. However, previous studies are limited by the availability of explicit data on congestion times for unloading. This study first quantifies the association between port congestion days and U.S. agricultural exports, using monthly export data of top U.S. ports and their monthly average container and bulk shipments delays. We find one extra day delay of container shipments decreases U.S. agricultural monthly exports by 5 percent in quantity or 2 percent in value on average. That amounts to $63 million in monthly loss of export value on average, and Western U.S. ports are responsible for 69 percent of this total. The effect is most pronounced for the Western U.S. exports of bulk commodities, where congestion results in a 9 percent loss in quantity or 8 percent loss in value. For Eastern U.S., the most salient effect is on consumer commodities, with a loss of 3 percent in quantity and 3 percent in value. For the Gulf region, the largest effect is on bulk commodities, with a loss of 4 percent in quantity and 5 percent in value. The impacts of congestion on bulk shipments are both statistically and economically insignificant. However, we find some evidence that exporters substitute bulk cargoes with containers when bulk shipment delays at ports increase. The substitution of container shipments with bulk shipments, however, is unlikely.