'Small-World' Trade Networks Associated with Economic Growth, New Study Finds

Photo Submitted

Raja Kali and Javier Reyes, University of Arkansas

FAYETTEVILLE, Ark. – New research by University of Arkansas economists decodes the mystery of what transforms a network of interconnections among firms into a single organism that functions as an economic powerhouse.

Raja Kali and Javier Reyes, both associate professors in the Sam M. Walton College of Business, examined the relationship between product-clusters in international trade and their connection to accelerated economic growth at the country level. They found that the way in which a country’s exported products are connected to each other and to the other products in the global-trade network – rather than mere participation in global trade – determines whether or how much a country will achieve accelerated economic growth.

“The connectedness between firms and industries could stem from a number of different sources,” Kali said. “It could be the result of production patterns that share similar inputs, infrastructure or managerial techniques. It could be that an input/output relation between industries exists, or it may be that products from different industries simply share similar technologies.”

With help from doctoral student Joshua McGee and undergraduate student Stuart Shirrel, Kali and Reyes mapped the relationship between all products in the global trade network and specific products exported by individual countries and found that those countries whose product-specialization patterns resembled “small worlds” experienced a greater likelihood of an acceleration in economic growth. A small world is a network that possesses high average density of links among exported products and short average distance between their export products and other potential products in global trade. For example, South Korea, which is considered a growth-acceleration country, has exports concentrated in telecommunications, data processing, electrical machinery and automobiles, products that yield both high clustering among products and short distances to other products in global trade.

The researchers’ finding is a step toward decoding the mystery of why some countries experience accelerated economic growth and how that growth is related to trade. The research may help countries, especially poorer ones, target or prioritize sectors of their economies in relation to broader global patterns of trade.

“The vast majority of economic growth accelerations are unrelated to standard determinants such as political change and economic reform,” Kali said. “Most instances of economic reform do not accelerate growth. So does this mean that growth accelerations are idiosyncratic and just a matter of luck? Our work indicates this isn’t the case. We suggest that countries that experienced episodes of growth acceleration had patterns of trade that overlapped with those of the broader network of global trade. We see this in so-called ‘spillovers,’ or clustering of products, and the potential of these countries to develop and trade new products.”

Applying a network approach to the study of international trade, the research team first charted patterns of relatedness among all products in global trade from 1965 to 2000. These patterns of relatedness, which the researchers called “product space,” may be thought of as a network. The researchers then mapped “product specialization,” or patterns of trade by individual countries, and superimposed these country-level “sub” networks onto the larger product-space network.

These overlays enabled the researchers to measure spillovers, or the clustering of products, as well as the distance of a country’s product-specialization pattern from that of the overall global trade network. The distance measure indicated how difficult it was for a given country to move from its current product specialization to new products. In other words, countries with shorter distances (i.e., more closely related industrial structures) could move more easily from one product specialization to another or could exploit economies of scope more effectively and thus facilitate economic growth. The high clustering and short distances between country-specific patterns of trade and the overall global trade network epitomized the small-world phenomenon.

“Countries that experienced growth acceleration had an intersection between their product specialization pattern and product space, which created favorable, small-world-like conditions,” Reyes said. “But why should this small-world network facilitate rapid economic growth? Because clustering of products enables economies of scale and scope. The short-path length in the network allows leaps across product space to new products, and the extent of scale and scope of economies determine cost reductions, freeing up resources for investment.”

Copies of the study are available at http://wcob.uark.edu/rkali/GrowthNetworks.pdf and upon request from the authors.