The impact of mechanisation on wages and employment
VoxEU Column Productivity and Innovation Economic history Poverty and Income Inequality

The impact of mechanisation on wages and employment: Evidence from the diffusion of steam power

Fears about the effects of mechanisation on societies are not new; technology has always generated cultural anxiety throughout history. This column considers one of the most significant waves of mechanisation in history – the rise and spread of steam power in 19th century France – to examine the influence of mechanisation on labour outcomes. Rather than cutting jobs and wages, the authors find that that steam-adopting industries ended up employing up to 94% more workers than their non-steam-adopting counterparts and paid wages that were up to 5% higher on average.

Does growth and technological change enhance or reduce inequality? One key reflection of the impact of growth and technological change on inequality is the evolution of the skill premium, which measures the change of wage inequality between skilled labour and unskilled labour.  As discussed by Acemoglu (2002), both the skill premium and the relative supply of skilled labour increased in the US post-1970s.

Figure 1 Post-war US skill premium and relative supply of skills

Figure 1

Source: Acemoglu (2002).

Acemolgu (2002) attributes this joint increase in the skill premium and the relative supply of skilled labour to skill-biased technological change (SBTC). He states that the increase in the relative supply of skilled labour induces SBTC, which then leads to a rise in the relative demand for the skilled labour. The increase in the relative demand for skilled labour brought about by SBTC is so large that it exceeds the relative supply of skills. As a result, the skill premium goes up. Examples of SBTC are the invention of computers. Industries that are equipped with computers have excessive demand for people with skills in operating computers. This widens the wage gap between those with the skills and those without. In this way, we can see that modern technological change, while bringing tremendous growth, increases inequality.

New research

While the impact that modern technological change has on the skill premium has been well explained by the SBTC hypothesis, there has been no study of the evolution of the skill premium in the very long historical run (c.1300 to 1914). Van Zanden (2009) estimates the skill premium in Western Europe using data on the wages of skilled craftsman and unskilled labourers produced by Allen (2001). Following the estimation made by van Zanden (2009), in a recent paper (Luo 2017), I use the data on wages of skilled craftsman and unskilled labourers in Allen (2001) and, similarly to van Zanden (2009), I show that the skill premium in Western Europe declined from 140% in 1300 to 58% in 1600, and stayed at the level of 58% from 1600 to 1914 (Figure 2).

Figure 2 Skill Premium in Western Europe from 1300 to 1914

Figure 2

Data source: Wage of Craftsman and Labourers Estimated by Allen (2001)

From 1300 to 1914, Western Europe witnessed colossal technological change, growth, and development. This brought about the fundamental socioeconomic transition from a pre-modern society, with agriculture as the main industry of production, to the modern industrialised society. The skill premium declined to a low and stable level as the economy approached the modern era. This indicates that growth and technological change in the past reduced inequality. The aim of the research I present here is to study how growth and technological change in this very long historical period reduced the skill premium to a low and stable level.

The model and findings

I develop a growth model that characterises the growth and technological change in this long historical period.  My model ‘unifies’ the economic growth in the pre-modern era and the modern era, as well as the transition in between in a way that is similar to the canonical unified growth models developed by Oded Galor and co-authors. Similar to Galor and Moav (2004), my model shows that the growth of the capital-human capital ratio (the abundance of physical capital relative to human capital) and human capital investment are the two fundamental driving forces of economic growth and transition from pre-modern society to modern society. And my model generates a ‘take-off’ from pre-modernity to modernity similar to that found by Galor and Weil (2000).

Within this unified growth framework, I show that the growth of both the capital-human capital ratio and human capital investment not only drives economic growth, but also affects wage inequality. And they have competing effects on the skill premium – increasing the capital-human capital ratio has a positive effect on the skill premium, whereas increasing human capital investment has a negative effect on it. Which effect dominates the other depends on the level of the capital-human capital ratio. My study provides several findings.

  • First, when the capital-human capital ratio is low, the negative effect of the capital-human capital ratio dominates.
  • Second, as the capital-human capital ratio grows higher, the two effects cancel out.
  • Third, when the capital-human capital ratio becomes sufficiently high, the positive effect of the human capital investment dominates at the beginning. But as the capital-human capital ratio continues to grow, its negative effect on the skill premium dominates again.

The growth of the capital-human capital ratio is driven by technological change. As the abundance of physical capital relative to human capital grows from an initially low level to a sufficiently high level, the three findings above show that the capital-human capital ratio and the human capital investment interact in three different ways to determine the skill premium.

I can then partition the period from 1300 to 1914 into three regimes and see the evolution of the skill premium:

  • In the first regime (1300-1600), the capital-human capital ratio is low and technological progress is slow (it is powered by inventions that come up intermittently) but is not at a standstill. The negative effect of the ratio dominates.  And as the ratio grows as a result of technological progress, the skill premium goes down (from 140% to approximately 58%).
  • In the second regime (1600-1800), technological progress is as slow as before, yet the capital-human capital ratio becomes higher and triggers mild growth of human capital investment. As the ratio continues grow, powered by technological progress, the human capital investment is growing in tandem with and their competing effects cancel out. This results in a stationary level of the skill premium (at 58%).
  • In the third regime (1800-1914), the capital-human capital ratio becomes sufficiently high. This results in an increase in the amount of human capital investment. The increase in human capital investment also leads the positive effect of human capital investment to be initially dominant over the negative effect of ratio, incurring an upward jump of the skill premium. On the other hand, the human capital investment becomes so large that it starts to augment technological progress. The more investment made to acquire human capital, the higher the rate of the technological progress. This leads to sustainably growing technology. The acceleration of technological progress results in faster growth of the ratio. And as time goes by, the negative impact of its growth becomes dominant again. The skill premium then decreases and converges back to the level in the previous regime (i.e. 58%) in the long run.

According to these three findings, the skill premium, after experiencing an initial decline, converges to a low and stable level afterwards. These findings unveil the insightful linkage between growth and inequality in the very long run – the more advanced the regime an economy enters, the lower the wage inequality. Growth and technological change, while stimulating economic development and the transition from a pre-modern era to modern era, reduced the wage inequality between different working groups. This implies that people from all working groups, both the skilled and the unskilled, benefit from economic growth.

Policy implications

This finding has important policy implications. It shows that physical capital accumulation (i.e. growth of the capital-human capital ratio) is the key to reducing poverty and facilitating technological progress. As physical capital becomes sufficiently abundant relative to human capital, individuals have a stronger preference to invest in human capital, which triggers technological progress that benefits people in all working groups. This shows that policies that enhance physical capital accumulation can increase individuals’ incentive to invest in human capital. At the same time, good institutions are needed to guarantee that each individual will see increases in their capital stock. In combination with infrastructure development, this is essential to reducing poverty and inequality in contemporary developing and underdeveloped countries.


Acemoglu, D (2002), “Directed technical change”, The Review of Economic Studies 69(4): 781-809.

Allen, R (2001), “The Great Divergence in European Wages and Prices from the Middle Ages to the First World War”, Explorations in Economic History 38(4): 411-447.

Galor, O and D Weil (2000), “Population, Technology and Growth: from Malthusian Stagnation to the Demographic Transition and Beyond”, American Economic Review 90(4): 806-828

Galor, O, and M Omer (2004), “From Physical to Human Capital Accumulation: Inequality and the Process of Development”, The Review of Economic Studies 71: 1001-1026

Luo, R (2017), “Skill Premium and Technological Change in the very Long Run: 1300-1914”, Discussion Papers in Economics 17/09, University of Leicester.

van Zanden, J L (2009), “The Skill Premium and the Great Divergence”, European Review of Economic History 13: 121-153.