VoxEU Column Economic history International trade

Globalisation everywhere, except in the growth numbers: Pessimism reaffirmed?

Understanding the relationship between trade and growth is still at the core of the economics profession. This column seeks to identify the pathways by which globalisation affects economic growth looking at the case of Belgium in the decades preceding the First World War. It argues that the collapse in fixed export costs promoted the entry of uncompetitive firms into export markets and as the trade component of GDP rose, the share of high performing firms contracted, slowing growth.

A fundamental proposition of economics is the positive relationship between trade and growth. But in a recent Vox contribution, Constantinescu et al. (2015) find the relationship, while positive, to be unstable over time. An explanation consistent with this type of findings is that the benefits of trade are more difficult to realise at higher degrees of openness – a case of squeezing more juice out of old lemons (Rodrik 2011). At issue are the dynamics underlying the changing relationship between trade and growth. Under what circumstances might we be pessimistic about future trends?

We bring a historical perspective to the question using a new granular trade data set for Belgium in the decades preceding the First World War (Huberman et al. 2015). As is well known, the late nineteenth-century was the heyday of globalisation, a precursor to its modern variant. The standard account of the trade boom starts with the transportation revolution, which in Belgium can be dated to the opening of the Scheldt and subsequent investments in Antwerp, Europe’s busiest port. Typical of other European countries, Belgium participated in the expanding treaty network of bilateral agreements combining lower tariffs with the most-favoured-nation clause. These arrangements foreshadowed today’s rule-based trading system (Irwin 2002). Owing to its small size, and committed to trade liberalisation, Belgium resisted the turn to protectionism that embraced large countries, like France, Germany, and Italy, in reaction to the grain invasion emanating from the New World and Russia. Because the country was rich in capital and labour, and doted with an extensive rail network, workers could move seamlessly into new export sectors in manufacturing. Between 1870 and the eve of the war, the degree of openness of the Belgian economy climbed from about 30 to 60%.

The trouble is that, while globalisation was ubiquitous, economic growth was lacklustre. In Figure 1, we contrast the expansion in the number of goods delivered and destinations (independent states and colonies) served, both of which more than doubled between 1870 and 1910, and the growth in output per capita which recorded an increase of less than 50%. The gap between the growth in the number of items shipped and the rate of increase in output per capita widened over the period. This is curious. The expansion in goods is believed to be a metaphor for new innovation in products and in processes (Baldwin and Robert-Nicoud 2008). More generally, growth in trade of this magnitude ought to be reflected in the shifting of resources to large and productive export-oriented establishments, the so-called positive selection effects of trade liberalisation (Melitz 2003). But in Belgium, the decline in old staples, like cotton-textiles, was glacial, the export surge of new industries, steel and chemicals, being postponed until the1920s.

Figure 1. Products, destinations, openness, and economic growth

The puzzle reconsidered

Our resolution of the trade-growth paradox draws on recent contributions that study ‘heterogeneous’ firms, product differentiation, and international trade (Bernard et al. 2007). To begin, we posit that a decline in fixed costs was a leading factor in the trade boom, alongside the decline in variable (transport) costs. In Belgium’s case, the decline in fixed costs was related to the establishment of an extensive diplomatic network that served over 70 countries by 1910. Large countries had a natural advantage of selling goods abroad, but the Belgian state’s investments in foreign representation levelled the playing field. The network provided extensive business intelligence on distant markets, thereby lowering beachhead costs and enabling the entry of new products, many of which were previously uncompetitive in international markets.

To gain leverage on the importance of the effect of the decline in fixed costs on new goods, Figure 2 decomposes the change in the value of exports that can be attributed to the introduction of new goods, the extensive margin, and to continuing or incumbent goods, the intensive side (Bernard et al. 2009).  For the entire period, the contribution of new goods (58%) exceeded that of continuing goods (45%), much of the growth occurring in the first half of the period when new goods were being developed. The importance of new products anticipates the growth in the extensive margin caused by the decline in fixed costs associated with WTO membership (Dutt, Milhov, and Van Zandt 2013).

Figure 2. Contribution of extensive and intensive margins

Next, we examined the effect of different firm types and product differentiation on the trade boom. As in the model of Chaney (2008), we find that the trade boom was greater in sectors which exhibited a high degree of product differentiation because markets for these goods were less competitive. In addition, the trade boom was bigger in sectors in which firm types were more homogeneous or similar because a small change in trade costs would enable many more new firms to enter trade, especially in markets for differentiated goods.

The positive impact of trade on productivity was not guaranteed. For instance, Belgian was renowned for its streetcars, exporting transportation equipment to far flung destinations. In this sector, a handful of leading firms produced differentiated goods, and productivity increased in step with trade. But in other sectors, like textiles, a high level of firm uniformity in terms of technique and processes muted the effect of trade on productivity. Many of the new firms entering export activity were in fact small and, even as they shipped differentiated goods to markets abroad, productivity growth in the sector was negligible. Such dynamics are evidence of negative selection effects of increased integration. The rub was that the share of the old staples in total output trumped that of the new industries, and average industry productivity stagnated as a result. Contrary to expectations, globalisation had the effect of sustaining the non-performing old economy rather than re-orienting resources toward the new more dynamic sectors.

Concluding remarks: Technology vs geography

Our research has sought to identify the pathways by which globalisation affects economic growth. While it is hazardous to project on the basis of a case study, we leave a bittersweet message. The collapse in fixed costs enabled a small country like Belgium to participate in the global trade boom. The investment in a diplomatic trade network extended and deepened the country’s reach in global markets. But the flipside gives less ground for optimism. In certain markets, the collapse in trade costs promoted the entry of formerly uncompetitive firms. A type of contest ensued between the positive selection effects of shifting resources and the negative effects due to declining trade costs – a race between technology and geography. The balance between the two changed as globalisation intensified. Globalisation proceeded unabated to deliver new goods to new destinations, but even as the trade component of GDP rose, the share of high performing firms contracted and less competitive businesses filled the void.


Baldwin, R E and F Robert-Nicoud (2008), “Trade and Growth With Heterogeneous Firms”, Journal of International Economics, 74: 21-34.

Bernard, A B, J B Jensen, S J Redding, and P K Schott (2007), “Firms in International Trade”, Journal of Economic Perspectives, 21: 105-30.

Bernard, A B, J B Jensen, S J Redding, and P K Schott (2009), “The Margins of US Trade”, The American Economic Review, 99: 487-93.

Chaney, T (2008), “Distorted Gravity: The Intensive and Extensive Margins of International Trade”, The American Economic Review, 98: 1707-21.

Constantinescu, C, A Mattoo, and M Ruta (2015), “Explaining the global trade slowdown”, VoxEU.org, 18 January.

Dutt, P, I Mihov, and T Van Zandt (2013), “The Effect of WTO on the Extensive and Intensive Margins of Trade”, INSEAD, working paper, 2013/38/EPS. 

Huberman, M, C M Meissner, and K Oosterlinck (2015), “Technology and Geography in the Second Industrial Revolution: New Evidence from the Margins of Trade”, NBER Working Paper No. 20851.

Irwin, D (2002), “Long-Run Trends in World Trade and Income”, World Trade Review, 1: 89-100.

Melitz, M (2003), “The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity”, Econometrica, 71: 1695-1725.

Rodrik, D (2011), The Globalization Paradox: Democracy and the Future of the World Economy, New York: W.W. Norton.

2,415 Reads