Britain was the first country to industrialise, and it was also the first country to contend with foreign industrialisation – in the US and Continental Europe. In the 21st century, as the locus of manufacturing activity expands, policymakers in both developed and newly industrialised countries can look to the late 19th century to understand what determines the pattern of comparative advantage (disadvantage) within a country’s manufacturing sector.
The presence and extent of a comparative advantage can be measured or – in the field of economic history – estimated. The earliest systematic estimates of the manufacturing comparative advantages of Britain are for several years from 1899-1950 (Crafts 1989). By then, the relative industrial decline of Britain was well underway. In a new paper, I produce estimates of the manufacturing comparative advantages of Britain for an even earlier period – the late Victorian era (Varian 2016). These estimates span 17 manufacturing industries in each of three years – 1880, 1890, and 1900.
An already (de)specialised workshop
My estimates are a modified version of revealed comparative advantage (RCA) indicators (Balassa 1965), whereby a country’s (e.g. Britain’s) share of world exports of industry x is normalised for the country’s share of total world exports. World exports of manufacturing industry x are approximated by the exports of Britain, Belgium, France, Germany, and the US, which are combined and then rescaled to account for the manufacturing exports of other smaller industrial countries.
As the late 19th century well-preceded the establishment of the Standard Industrial Trade Classification (SITC), the classifications of industries differs widely among the trade statistics of these five countries. Consistent classifications of industries are achieved by reconstructing the industries using data from the finest levels of disaggregation reported in the trade statistics.
Table 1 RCA indicators of British manufacturing industries, 1880-1900
The RCA indicators are presented in Table 1. An indicator greater than 1 signifies a comparative advantage, and an indicator less than 1 signifies a comparative disadvantage. Britain’s iconic strongholds of cotton textiles and iron and steel are supported by the RCA indicators. However, even as early as 1880, Britain was at a distinct comparative disadvantage in a number of manufacturing industries – clocks and watches; glass; leather and manufactures thereof; silk manufactures; and spirits – though Britain went on to realise a comparative advantage in this last industry. Altogether, the late Victorian ‘workshop of the world’ was already ‘de-specialising’ away from certain industries.
These RCA indicators offer the first quantitative, sector-wide profile of the relative performance of British manufacturing industries extending back into the 19th century. They supplement a vast qualitative literature on the subject, which was consolidated and assessed in Broadberry (1997), though it should be stressed that he was less concerned with comparative advantage than with comparative labour productivity (vis-à-vis the US and Germany).
Factor determinants of comparative advantages
What were the factor determinants of Britain’s manufacturing comparative advantages in the late Victorian era? This question is answered using a four-factor Heckscher-Ohlin model that relies on symmetrised forms of my RCA indicators (Laursen 2015). The four factors are capital, unskilled labour, human capital and material inputs. Proxies for factor intensities per industry are constructed using data from British government publications, chiefly the Census of production (1907). Great care is taken to match the industry definitions of the proxies with the industry definitions of the RCA indicators.
The results unsurprisingly indicate that Britain’s manufacturing comparative advantages were relatively capital-intensive.
In contrast with previous scholarship (Crafts and Thomas 1986), I find that Britain’s manufacturing comparative advantages were relatively non-intensive in unskilled labour. Crafts and Thomas (1986) find a positive relationship between British gross exports and unskilled labour intensity per industry. Unlike Crafts and Thomas, my analysis relies on indicators of comparative advantage that are normalised for the country-share of world exports, which includes the exports of the lower-wage manufacturing economies of Continental Europe (Williamson 1995). Not only did Britain’s manufacturing comparative advantages economise on unskilled labour, they did so to an increasing extent throughout the late 19th century.
I also find that no clear relationship emerges between human capital intensity and Britain’s manufacturing comparative advantages. There is some evidence that the comparative advantages were relatively human capital-intensive, but this finding is very much dependent on the particular proxy used.
Despite the limited natural resource endowments of Britain, there is no evidence that its manufacturing comparative advantages were in those industries with a lower factor proportion of material inputs. There are three possible explanations for this somewhat surprising result:
- First, Britain was favourably endowed with one very important material input – coal. The material neutrality of the comparative advantages may be driven by a generally high share of energy in material inputs.
- Second, Victorian Britain espoused a policy of free trade, which extended to raw materials and intermediate goods. Unlike in other industrial countries, where a protectionist backlash had taken hold, the British manufacturing sector could obtain material inputs at close to the world price.
- A third potential explanation for the material neutrality of Britain’s manufacturing comparative advantages comes from the recent gravity literature, which argues for a strong empire effect on trade (Mitchener and Weidenmier 2008) and trade costs (Jacks et al. 2010). Recourse to a resource-rich empire would have mitigated the effects of Britain’s relatively unfavourable (non-energy) resource endowments on its manufacturing sector.
The policy implication of this research takes the form of a perhaps obvious, but nonetheless important reminder of the considerable diversity within the manufacturing sector. Even the ‘workshop of the world’, so readily associated with the strength of its manufacturing sector, was at a measurable comparative disadvantage in a number of manufacturing industries during the late Victorian era, and these disadvantages can be explained at least partly by the Heckscher-Ohlin model.
In the globalising world of the 21st century, discussions of the manufacturing sector as a whole are seldom meaningful. The industrialisation and deindustrialisation of countries, when discussed in broad terms, ignores the considerable variation and changes that are likely to be occurring at the level of individual industries. In this respect, the first era of globalisation offers an important lesson for policymakers in this second era of globalisation.
Balassa, B (1965) “Trade liberalisation and ‘revealed’ comparative advantage”, Manchester School, 33: 99-123.
Broadberry, S N (1997) The productivity race: British manufacturing in international perspective, 1850-1990, Cambridge: Cambridge University Press.
Crafts, N F R (1989) “Revealed comparative advantage in manufacturing, 1899-1950”, Journal of European Economic History, 18: 127-137.
Crafts, N F R and M Thomas (1986) “Comparative advantage in UK manufacturing trade, 1910-35”, Economic Journal, 96: 629-645.
Jacks, D S, C M Meissner and D Novy (2010) “Trade costs in the first wave of globalization”, Explorations in Economic History, 47: 127-141.
Laursen, K (2015) “Revealed comparative advantage and the alternatives as measures of international specialization”, Eurasian Business Review, 5: 99-115.
Mitchener, K J and M Weidenmier (2008) “Trade and empire”, Economic Journal, 118: 1805-1834.
Varian, B D (2016) “The revealed comparative advantages of late-Victorian Britain”, LSE Economic History Working Paper 239.
Williamson, J G (1995) “The evolution of global labor markets since 1830: Background evidence and hypotheses”, Explorations in Economic History, 32: 141-196.