VoxEU Column Economic history International trade Productivity and Innovation

Temporary protection and technology adoption: Evidence from the Napoleonic blockade

The effect of trade protection of infant industries in developing countries on their long-term growth has been widely debated. This column provides evidence on this topic using a novel dataset from the Napoleonic blockade against British trade. The author analyses the effect of this temporary trade protection on the cotton spinning as an infant industry, employing within-country variation in the trade protection. In the short run, better protected regions increased their production capacity in the infant industry. There is a persistence of this productivity in the long run as well. 

Can temporary trade protection of infant industries in developing countries foster their long-term development? The question has a long tradition in the history of economic thought and has been much debated both in academia and in policy circles. Advocates often point to the historical experience of Germany and the US, two countries which industrialised behind highly protective tariffs in the 19th century. More recently, the East Asian and Chinese growth miracles coincided with targeted protection and subsidisation of certain industries. Sceptics of infant industry policy, however, point to an equally long list of episodes across Latin America and Africa where decades of similar policy intervention failed to deliver on the promised gains.

Theory tells us that, under certain conditions, an industry which is not competitive at free trade prices in a given country can become competitive in the future if given temporary trade protection from technological leaders. To date, the literature has attempted to assess the empirical relevance of the economic mechanism supposedly driving infant industry predictions by quantitatively assessing the importance of infant industry protection as implemented by the policymaker (Baldwin and Krugman 1986 and 1988, Head 1994, Irwin 2005).

New research on trade protection and technology adoption

In a recent paper (Juhasz 2014), I exploit quasi-random within-country variation in temporary trade protection of an infant industry of crucial importance for 19th century development. Using a novel dataset collected and complied from hundreds of hand-written archival sources and thousands of shipping bulletins, I estimate the effect of temporary trade protection on the development of mechanised cotton spinning across different regions of the French empire during and after the Napoleonic blockade (1806-1813) against British trade.

The approach taken in this paper improves on the existing literature in two important respects.

  • First, the presence of within-country variation in temporary trade protection allows me to compare regions of the French empire before and after the Napoleonic blockade which were more or less affected by the rupture to trade with Great Britain.

In the absence of variation in trade protection, existing studies have had to simulate the counterfactual of no protection. As such, this study constitutes the first credibly identified evidence consistent with the economic mechanism underpinning the infant industry argument.

  • Second, the fact that trade protection was driven not by the actions of a policymaker, but by exogenous events shaped outside of the French empire, means that I am able to focus on the economic mechanism at work without worrying about the confounding political economy considerations which are undoubtedly at work when infant industry protection is implemented by the policymaker.

Existing studies analyse instances of infant industry protection as implemented by the policymaker and estimate the joint effect of the economic mechanism and the efficacy of the implemented policy.

The Napoleonic blockade was implemented in the unusual way of attempting to stop British goods from entering continental Europe. Ports across Europe were closed to ships carrying British goods and the military was active in enforcing the blockade along the coastline of the French empire. My empirical strategy exploits the fact that while the blockade was generally effective along the coastline of the French empire, it was unevenly successful outside of France for geopolitical reasons outside of Napoleon’s control (see Figure 1). This meant that instead of stopping British goods from entering continental Europe, as Napoleon intended, trade was instead displaced from direct to more costly indirect trading routes which entered the French empire at its overland borders via third countries. Variation in trade protection within France was driven by the fact that, depending on where the given region was located within the empire, the costs of trading with Britain (as measured by the effective distance to Great Britain) increased to a greater or smaller extent. In particular, trade protection generally increased markedly for regions in the northern parts of the empire, while the increase was much smaller in southern regions.

Figure 1. Intensity of shipping with the British at the port level before (1802) and during (1809) the Napoleonic blockade. Intensity of shipping with the British is measured as the number of ships sailing between Britain and the given port in a given year.

Cotton spinning as an infant industry

In my analysis, I focus on mechanised cotton spinning as it has many features which make it an ideal candidate for an infant industry. Mechanisation in cotton spinning was a new technology in which the French were not competitive prior to the Napoleonic blockade. The machines were invented in Britain in the closing decades of the 18th century. This new production method had a large impact on productivity, with prices decreasing tenfold in the space of a decade. Learning spillovers from operating the technology have been widely documented for mechanised cotton spinning (David 1970 and Mokyr 2009). This is an important feature, as this is one mechanism via which theoretical models lead to infant industry predictions. Despite the French having similar conditions in the cotton industry prior to mechanisation, the new technology was not adopted on a wide scale before the onset of the Napoleonic Wars. According to one estimate, there were 18,000 spinning jennies in use in Britain in 1789, while the French only had 900 (Aspin and Chapman 1964). Consistent with the lag in technology, French spinners were producing at twice the price of their British counterparts on the eve of the blockade.

Empirical findings

I find that temporary trade protection had effects consistent with an infant industry mechanism both in the short run, during the period of temporary protection, and in the long run, once trade routes reverted back to their pre-blockade levels.

I estimate that during the blockade, regions which became better protected from British competition (in the northern regions of the French empire) increased production capacity in mechanised cotton spinning to a much larger extent than regions which remained more open to trade with the British (regions in the south) as is evident from Figure 2.
Turning to the long-term effects of temporary trade protection, I find persistence in the location of cotton spinning within France, and spinning firms located in regions with higher post-blockade spinning capacity were more productive 30 years later.
Finally, consistent with evolving comparative advantage in cottons, I find that exports of cotton goods from France increased markedly after the end of the blockade as Figure 3 shows.

About 15 years after the blockade ended, 7.5% of French exports were in cotton goods, suggesting that the sector was relatively important in the French economy.

Figure 2. Spinning capacity (spindles per capita) in each department in 1803 (the year the Napoleonic wars began) and 1812 (towards the end of the Napoleonic wars)

Figure 3. Exports of cotton manufactures from France (millions of 1820 francs)

Relevance for developing countries

What can developing countries hoping to move into high growth sectors learn from this episode? One appealing aspect of early 19th century France is the extent to which it is general to the development experience of many countries. My analysis focuses on France at a point in time in which a large fraction of its labour force moves out of agriculture and into unskilled, labour-intensive textile manufacturing. The generality of the context suggests that similar market imperfections may inhibit developing countries today from exploiting their underlying comparative advantage. Interestingly, differences between Britain and France in the early 19th century were tiny compared to disparities between rich and poor countries today. On the one hand, this suggests that similar mechanisms may be even more pronounced in cases with larger disparities. On the other hand, one of the reasons I find such large effects of a short episode of trade protection is that France had the necessary underlying conditions which made adoption of the new technology profitable once competition from Britain was dampened. In other words, France was ready to adopt this new technology. One reason for so many episodes of failed infant industry protection may be that policy was promoting industries in which the necessary underlying conditions were not present. This suggests that in the absence of the necessary underlying conditions, infant industry protection as implemented by the policymaker can turn out to be a very blunt instrument.

Concluding remarks

The findings from this paper do not lead directly to policy prescriptions for at least two reasons. As infant industry protection of mechanised cotton spinning in France was not implemented directly by policy, it was not susceptible to government failure in the same way as infant industry protection implemented by the policymaker. Second, even if a sector can be successfully encouraged by policy, it will not necessarily be good for the economy overall. Protection of an industry, be it a subsidy, a tariff or another policy tool always comes with a price tag. Determining whether the short-run costs of protection outweigh the long-term benefits is a further difficult question for policymakers. The findings of this paper, however, strongly suggest that more research is needed both on when and how infant industries should be promoted.

References

Aspin, C and S D Chapman (1964), James Hargreaves and the Spinning Jenny, Preston: Helmshore Local History Society.

David, P A (1970), “Learning by doing and tariff protection: A reconsideration of the case of the ante-bellum United States cotton textile industry”, The Journal of Economic History, 30(03), 521-601.

Baldwin, R and P R Krugman (1986), “Market access and international competition: a simulation study of 16K random access memories”, NBER working paper No. w1936

Baldwin, R and P Krugman (1988), “Industrial policy and international competition in wide-bodied jet aircraft”, in Trade policy issues and empirical analysis (pp. 45-78) University of Chicago Press

Head, K (1994), “Infant industry protection in the steel rail industry”, Journal of International Economics, 37(3), 141-165.

Irwin, D A (2005), “The welfare cost of autarky: evidence from the Jeffersonian trade embargo, 1807–09”, Review of International Economics, 13(4), 631-645.

Juhasz, R (2014), “Temporary Protection and Technology Adoption: Evidence from the Napoleonic Blockade”, CEP Discussion Paper No. 1322

Mokyr, J (2009), The Enlightened Economy: An Economic History of Britain 1700-1850, Yale University Press.

3,150 Reads