Picture a global supply chain. The one that puts together the Amazon Kindle, for example: The flex circuit conductors are made in China, the wireless card is made in South Korea, and the tablet is assembled in Taiwan. The system works because each location specializes in something, whether it is relatively cheap labor, a cluster of machinery, or technical skills. But unlike a product made in a single factory, the Kindle’s components must cross borders.
The ease of crossing those borders – including through seaports or airports – is crucial to the production network. And, as it happens, fluidity is more important to trade in components than trade in final products. This makes sense, logically – it is easy to see how a whole holiday season’s worth of Kindles could be held up if the flex circuit conductors or wireless cards don’t get to Taiwan on time.
Now there is new evidence that, as production networks become more common globally, trade logistics are increasingly vital. In a working paper published through the International Trade Department, we show that networked trade in parts and components is more sensitive to the importing country’s logistics performance than trade in final goods. Supply chains go where the logistics are smooth.
Our research also finds that this dynamic is particularly important in the Asia-Pacific region, where international production networks have flourished in recent years, and in trade between poor countries. These findings have important implications for development. They help us identify areas where government policy could aid in attracting sections of a supply chain, increasing foreign investment and trade.
Consumer electronics, such as the Kindle or Apple’s iPod, contain hundreds of components that are made in several different locations. Automobiles often get put together across international borders. Fast border-processing, well-run ports and good roads help the companies in these industries and others move components efficiently and cheaply and reduces the amount of inventory they must hold.
In our research, we use a gravity model to learn how components trade and logistics performance are linked. The trade data – including data on parts and components trade – comes from the WITS/UN COMTRADE database. To measure logistics performance, we use the World Bank’s Logistics Performance Index (LPI), which is based on a worldwide survey of logistics professionals. The 155-country index is a weighted average of the survey responses to a number of questions about the countries’ logistics “friendliness,” including the efficiency of the clearance process and the quality of infrastructure available.
We find that good logistics in the importing country are associated with higher trade values. In gravity model regressions we find that both importer and exporter logistics-performance variables are positive and significant in explaining trade volume.
When we look at interaction terms between logistics performance and trade in parts and components, however, the importer influence stands out. Using several methods – including a Poisson estimator, two other pseudo-maximum likelihood estimators and an OLS estimator – we find that only the importer-interaction term is positive and significant. The magnitude of the Poisson estimator, in particular, suggests that the semi-elasticity of trade with respect to importer logistics performance is about 45 percent stronger for parts and components than for final goods. One interpretation of this result is that the quality of logistics in the receiving country – import processing and infrastructure – is important to supply-chain companies when they decide where to locate. The association between logistics in the exporting country and components trade is less clear.
We find that this result is prounounced in the East Asia and Pacific region, home to a wide range of international production networks. Using the Poisson model, we find that the import-side interaction term is positive and significant in three regressions: one with only EAP exporters and all importers, a second with only EAP importers and all exporters and a third with only EAP importers and exporters (intra-regional trade). Though the sample size is small, we find strong evidence that logistics in the importing country are important to trade and particularly important to trade in parts.
Finally, we separated countries by income, and looked at the effect of logistics on trade. The reasoning was this: many supply chains operate across poor countries, while final product often goes to wealthy countries. Specifically we were interested to know how logistics affected components trade between low- and middle-income (“South”) countries compared with trade between those countries and wealthy (“North”) countries. We found that flows between the lower-income countries were particularly sensitive to import logistics. Specifically, a comparison between an econometric model of South-South trade and one of South-North trade shows that South-South flows of parts and components are particularly sensitive to logistics performance on the import side and that South-South flows are more sensitive to logistics than South-North flows.
All of this evidence suggests that countries – and especially developing countries – should pay attention to logistics performance if they want to attract investment by companies operating international supply chains. Improving efficiency at border-crossings or in port operations could be key to allowing a country take part in this increasingly important source of trade growth. A logical place for policymakers to start is by applying diagnostic exercises that use the LPI to identify bottlenecks in their countries’ logistics performance.