The Great Trade Collapse: An Evaluation of Competing Stories
One sentence summary: Retail inventories have contributed the most to the great trade collapse and the corresponding recovery, followed by protectionist policies, intermediate-input trade, and trade finance.
The corresponding working paper is available here.
Abstract
The reduction in international trade has been more than the reduction in economic activity during the 2008 financial crisis, against the one-to-one relationship between them implied by standard trade models. This so-called the great trade collapse (GTC) has been investigated extensively in the literature resulting in alternative competing stories as potential explanations. By introducing and estimating a DSGE model using eighteen quarterly series from the U.S., including those that represent the competing stories, this paper evaluates the contribution of each story to GTC. The results show that retail inventories have contributed the most to the collapse and the corresponding recovery, followed by protectionist policies, intermediate-input trade, and trade finance. Productivity and demand shocks have played negligible roles.
Non-technical Summary
The reduction in international trade has been more than the reduction in economic activity during the 2008 financial crisis. This observation has been accepted as extraordinary, because its magnitude has been far larger than in previous downturns; accordingly, it has been called as the Great Trade Collapse (GTC, henceforth).
Since the relation between trade and economic activity is one to one in standard trade models (mostly implied by constant elasticity of substitution preferences in gravity-type studies), this collapse in trade has attracted attention in the recent literature, and its causes have been investigated extensively not only because the decline in trade flows relative to overall economic activity is surprisingly high but also because it has important implications for optimal policy response. Accordingly, alternative explanations have been achieved, including the dynamics of inventories, intermediate-input trade, compositional differences between traded goods and GDP, trade finance/credit, declining aggregate demand, or higher trade costs due to protectionist policies. Since most of these papers have competing stories, they have sometimes found conflicting results with each other as well. However, what if there were multiple stories contributing to GTC at the same time? If yes, what was the contribution of each story? In other words, are these stories complements of or substitutes to each other? Based on the literature introduced so far, answering these questions requires a structural estimation of a dynamic trade model with ingredients such as intermediate-input trade, inventories, protectionist policies, trade finance, and financial interactions between countries.
Accordingly, this paper introduces a dynamic stochastic general equilibrium (DSGE) trade model to create a bridge between the literatures of international trade and macroeconomics through investigating trade patterns in a dynamic framework that borrows the stories explaining GTC from the literature introduced above. The model considers individuals, manufacturers and retailers, where the latter two hold inventories of finished goods. There is a monetary authority who decides for the policy rate, although the interest rate faced by individuals (due to intertemporal choices) and manufacturers/retailers (due to financial needs, including trade finance) is subject to the country-specific risk premium. To consider compositional effects, the model distinguishes between traded versus nontraded goods, home versus foreign goods, and durable versus nondurable goods.
The model is estimated by state-of-the-art Bayesian techniques using eighteen series of quarterly data from the U.S., including durable and nondurable imports, durable and nondurable production, services versus overall consumption, prices, inventories, duties, risk premium, and wages. The estimated model is further used to decompose durable and nondurable imports into their components, representing the competing stories of intermediate-input trade, retail inventories, protectionist policies, trade finance, retail productivity shocks, and consumer demand shocks. When overall U.S. imports are considered, the results show that retail inventories have contributed the most to GTC and the corresponding recovery, followed by protectionist policies, intermediate-input trade, and trade finance. The compositional effects within imports are significant: while retail inventories are mostly responsible for the changes in durable imports, intermediate-input trade is responsible for the changes in nondurable imports. In all cases, productivity and demand shocks have played negligible roles.
The reduction in international trade has been more than the reduction in economic activity during the 2008 financial crisis. This observation has been accepted as extraordinary, because its magnitude has been far larger than in previous downturns; accordingly, it has been called as the Great Trade Collapse (GTC, henceforth).
Since the relation between trade and economic activity is one to one in standard trade models (mostly implied by constant elasticity of substitution preferences in gravity-type studies), this collapse in trade has attracted attention in the recent literature, and its causes have been investigated extensively not only because the decline in trade flows relative to overall economic activity is surprisingly high but also because it has important implications for optimal policy response. Accordingly, alternative explanations have been achieved, including the dynamics of inventories, intermediate-input trade, compositional differences between traded goods and GDP, trade finance/credit, declining aggregate demand, or higher trade costs due to protectionist policies. Since most of these papers have competing stories, they have sometimes found conflicting results with each other as well. However, what if there were multiple stories contributing to GTC at the same time? If yes, what was the contribution of each story? In other words, are these stories complements of or substitutes to each other? Based on the literature introduced so far, answering these questions requires a structural estimation of a dynamic trade model with ingredients such as intermediate-input trade, inventories, protectionist policies, trade finance, and financial interactions between countries.
Accordingly, this paper introduces a dynamic stochastic general equilibrium (DSGE) trade model to create a bridge between the literatures of international trade and macroeconomics through investigating trade patterns in a dynamic framework that borrows the stories explaining GTC from the literature introduced above. The model considers individuals, manufacturers and retailers, where the latter two hold inventories of finished goods. There is a monetary authority who decides for the policy rate, although the interest rate faced by individuals (due to intertemporal choices) and manufacturers/retailers (due to financial needs, including trade finance) is subject to the country-specific risk premium. To consider compositional effects, the model distinguishes between traded versus nontraded goods, home versus foreign goods, and durable versus nondurable goods.
The corresponding paper by Hakan Yilmazkuday has been accepted for publication at Macroeconomic Dynamics.
The corresponding working paper is available here.
The corresponding working paper is available here.