The Rise and Fall of World Trade, 1870-1939

Abstract

The ratio of world trade to output was a mere 2% in 1800, but it then rose to 10% in 1870 to 17% in 1900 and 21% in 1913. It then fell back to 14% in 1929 and only 9% in 1938. The period 1870-1913 thus marks the birth of the first great era of trade globalization, the period 1914-39 its death. What caused the trade boom and bust? The textbook interpretations offer a variety of narratives, but few precise answers. We use an augmented gravity model of trade to examine the gold standard, tariffs, and transport costs as determinants of trade. In the nineteenth century the gold standard was much more important than tariff policy, and just as important as transport costs as a trade-creating force. In the 1920s, the slowdown in trade was driven by a rise in transport costs, though trade barriers other than tariffs might have been important. In the 1930s, the final collapse of the gold standard, persistently high transport costs, and the expansion of other barriers drove trade volumes even lower.

Reassessing the Evolution of World Trade, 1870-1949

“…Here again, the blue line depicts our estimate that denominates the existing nominal trade data with our estimated nominal GDP series, while the green line corresponds to the one obtained if the real GDP data of Maddison (2001) As Figure 5 documents, our estimated world trade share series, the blue line, is characterized by a secular rise during the …rst globalization era (1870-1913) and a sharp subsequent decline during the interwar period . Thus, qualitatively our series appears in line with the narrative o¤ered by Estevadeordal, Frantz, and Taylor (2003) as well as Jacks, Meissner, and Novy (2011) regarding the pre-World-War-II evolution of world trade. 26 Yet, quantitatively our estimated trade shares are much larger than the incorrectly calculated one, depicted by the green line, that one would obtain by dividing the nominal trade data with real GDP.…”

“…In the historical context of the …rst globalization era and the interwar period the two main factors in ‡uencing the strength of these barriers were the extent of tari¤ protection and the participation in the gold standard. In this respect we follow the work of Estevadeordal, Frantz, and Taylor (2003), Jacks, Meissner, and Novy (2011) and others who have analyzed bilateral trade ‡ows during these periods based on a gravity framework. Speci…cally, we estimate the following regression speci…cation:…”

“…In this respect we follow closely the work of Eichengreen and Irwin (1995), Irwin and Terviö (2002), Estevadeordal, Frantz, and Taylor (2003), Lopez-Cordova and Meissner (2003), Jacks, Meissner, and Novy (2010) and others who have employed gravity models to analyze historical bilateral trade ‡ows. Yet, in contrast to most existing contributions, we estimate our gravity speci…cation combining non-PPP-adjusted bilateral trade data with our estimated non-PPP-adjusted income data instead of PPP-adjusted ones.…”

“…To assess the quality of our estimated nominal income and implied trade share data, we 1 See for example the work of O' Rourke and Williamson (1999), Estevadeordal, Frantz, and Taylor (2003), Jacks, Meissner, and Novy (2010), Schularick and Solomou (2011) and the references therein.…”

“…3 Following the language of the international comparison literature, we will often use the term "real" to refer to PPP-adjusted measures and the term "nominal" to non-PPP-adjusted ones. 4 Examples of cases were nominal trade data were combined with real income data can be found in Estevadeordal, Frantz, and Taylor (2003), Lopez-Cordova and Meissner (2003) and Jacks, Meissner, and Novy (2011).…”