As part of our project, we study the effect of migration and remittances on wages and living standards, specifically during two periods of large scale migration. The first is migration from Europe to the lands of new settlement, mainly the U.S., in the period roughly from 1850 to 1914. The second is the more recent emigration from developing countries. In this first post, we look at migration from Europe to the lands of new settlement; a subsequent post will explore current migration from developing countries.
Migration from Europe to the New World was about 46 million in the 19th century before 1913 while population increased from 192 million in 1800 to 423 million in 1900. So, annual migration was about 10 percent of the population over the century and even higher after 1880 (Taylor and Williamson, 1994). The extreme case was emigration from Ireland. Four million people migrated, contributing to a population decline from 8.2 million in 1841 (a level not since exceeded) to 4.4 million in 1911 (O’Rourke, 1995). Almost a quarter of the population emigrated at the rate of about 13 per thousand per annum.
Migration contributed substantially to the increase in Irish incomes between 1850 and 1914 (O’Rourke, 1995). Real agricultural wages doubled between 1860 and 1913, growing at 1.6% a year, and grew even faster, 1.9% a year, between 1860 and 1895. Unskilled building wages increased at 2.2% a year (Boyer, Hatton and O’Rourke, 1994)[1] which contrasts with the increase in per capita output of 1.6% a year between 1830 and 1913 (Kennedy, Giblin and McHugh, 1988). Relative dispersion among European and the new lands declined between 1870 and 1910 by 28 % for the real wage, 13% for GDP per capita and 24% for GDP per worker. In the absence of migration, wage, productivity and income per capita levels would have been higher in the new world and lower in the old world. For instance, without migration Irish wages would have been 31% lower, Italian wages 23% and Swedish 10%, while Argentine wages would have been 36% higher, Australian by 22%, Canadian by 25% and the US by 12%.[2] The simulations suggest that more than all the actual reduced dispersion, 168%, was accounted for by migration.[3] The migration particularly helped to raise relative wages in the poorer European countries. For example, real wages in Denmark increased from 52% to 85% of those in Britain, in Ireland from 71% to 88%, Italy from 38% to 40%, Sweden from 41% to 82% and Norway from 41% to 65% (Williamson, 1995).[4] These migrants were young male adults, single and unskilled (Hatton and Williamson, 1992) and some, such as those from Norway, were workers with lower productivity and poorer economic prospects (Abramitzky, Boustan and Eriksson, 2010).[5] They were then an immediate addition to the labour market in the host country.
What of changes in living conditions? There is considerable controversy about this. The increase in the mean height of recruits to the army in the second half of the 19th century is taken as an indicator of improved living standards (Flood, Wachter and Gregory, 1990). However, more direct measures dispute this conclusion, “for the last six decades of the nineteenth century, infant mortality for England and Wales fluctuated around a level of about 155 deaths per thousand births" (Huck, 1995). Mortality rates in the developed countries in the late nineteenth century were much higher than are these rates in today’s developing countries (Crafts, 1997). Crafts calculates the human development index (HDI) for the 16 richest countries for which Maddison had provided per capita incomes. He finds that the value of the HDI in today’s developing countries is much higher than in that of the rich countries in the 19th century, and this is mainly because of much lower mortality rates.
[1] They grew even faster if adjusted for lower probability of unemployment.
[2] The relatively poor were more likely to migrate; but not very often the poorest, at least initially, because of the cost of migration (Abramitzky, Boustan and Eriksson, 2010). But later remittances from those who had already migrated helped poorer migrants. Wages rates grew by 1.3% a year in the UK and so for all these countries wage rates grew faster than 1.3% a year and this was more than the growth of per capita output which was 0.8%, 1.2% and 1.3% for Italy Norway and Sweden respectively.
[3] Analysis shows that the CV is almost cut in half over the three decades, 1870-1900 (Williamson, 1995). Most of this was (60%) due to reduction of gap between the new and old worlds. Very little change in CV in the old world or the new world.
[4] Convergence to real wages in the U.S. must have been even more dramatic as relative real wages fell from being 98% higher than those in Britain in 1855 to only 54% in 1913. Furthermore, the proportion of the poor fell and of those living in lower quality housing declined from 63% in 1861 to 29% after half a century (Boyer, Hatton and O’Rourke, 1994, quoting O’Grada, 1988)
[5] This was likely to be the case for other countries also.