Firms whose names are lexicographically earlier in the Chinese-character rank export more to countries that have greater language proximities to Chinese; same for English-romanization names

Lexicographic biases in international trade. Hua Cheng, Cui Hu, Ben G. Li. Journal of International Economics, May 30 2020, 103346. https://doi.org/10.1016/j.jinteco.2020.103346

Highlights
• This study identifies a new type of bias in international trade patterns.
• We use an empirical strategy that exploits the interplay between firm-name lexicographic variations and destination-country language proximity variations.
• The Chinese language is outside the Indo-European system, a linguistic feature that strengthens our identification.
• Quantitative, lexicographic biases are approximately one to three percent of export volume.
• In a nutshell, Exporter A exports more than Exporter Z to countries that speak languages that are more proximate to English.

Abstract: The names of traders should not matter if information is symmetric across traders. By examining export data from Chinese customs, we find persistent lexicographic biases in firm-level export records. Firms whose names are lexicographically earlier in the Chinese-character rank export more to countries that have greater language proximities to Chinese, while firms whose names are lexicographically earlier in the English-romanization rank export more to countries that have greater language proximities to English. The lexicographic biases signify linguistic visibility as a source of comparative advantage in international trade.

Keywords: Alphabetic biasLanguageGravity modelsBehavioral economics