Women of an uncertain age: quantifying human capital accumulation in rural Ireland in the nineteenth century†

Abstract

Geary and Stark find that Ireland's post‐Famine per capita GDP converged with British levels, and that this convergence was largely due to total factor productivity growth rather than mass emigration. In this article, new long‐run measurements of human capital accumulation in Ireland are devised in order to facilitate a better assessment of sources of this productivity growth, including the relative contribution of men and women. This is done by exploiting the frequency at which age data heap at round ages, widely interpreted as an indicator of a population's basic numeracy skills. Because Földvári, van Leeuwen, and van Leeuwen‐Li find that gender‐specific trends in this measure derived from census returns are biased by who is reporting and recording the age information, any computed numeracy trends are corrected using data from prison and workhouse registers, sources in which women ostensibly self‐reported their age. The findings show that rural Irish women born early in the nineteenth century had substantially lower levels of human capital than uncorrected census data would otherwise suggest. These results are large in magnitude and thus economically significant. The speed at which women converged is consistent with Geary and Stark's interpretation of Irish economic history; Ireland probably graduated to Europe's club of advanced economies thanks in part to rapid advances in female human capital.

Gender Gaps in Education

“…Italy actually showed a peculiar pattern in the GPI in enrollments, mostly due to the wide regional divide which characterized the country in many dimensions, including human capital formation: between 1861 and 1911, while higher and increasing gender equality, both in enrollments and literacy, was documented in the North, very low and stagnating levels were found instead in the Center-South of the country, a pathdependent pattern mostly inherited from preunification states and their differing educational policies prior to 1861 (Bozzano and Cappelli 2019;Cappelli and Vasta 2019;Ciccarelli and Weisdorf 2019). By contrast, in Ireland, gender differences in attendance were even reversed by the 1880s and those in enrollments followed suit by 1901 (Blum et al 2017). On the other side of the Atlantic, in the USA, mass schooling expanded since the early 1800s in a balanced way for males and females, both in terms of enrollment ratios and years of schooling.…”

Scarring and Selection in the Great Irish Famine

“…A set of control variables is based on the type of offence for which an individual was imprisoned, following the example of Blum et al (2017). Offences are distinguished into seven sub-categories, adapting a contemporary criminal classification: offences against the person, violent and non-violent offences against property, begging and prostitution, drunkenness, rioting, treason and a category for offences related to indecency.…”

“…For exemplary studies on this widely-observed phenomenon, seeMartínez-Carrión and Moreno-Lázaro (2007) on Spain; Riggs and Cuff (2013) on Scotland; Baten (2009) on Bavaria; and Zehetmayer (2017) on the US.15 See Appendix D for population-wide comparisons. Literacy levels were relatively high given the existence of state-funded education dating from the 1830s(Blum et al 2017).16 We categorise recidivists by the first crime for which they were imprisoned. 17 Similarly, accounting for the effect of shrinking among individuals aged 50 and above -estimated byFernihough and McGovern (2015) at 0.09% per year -allows us to extend average height trends further back in time.…”

“…15See Appendix D for population-wide comparisons. Literacy levels were relatively high given the existence of state-funded education dating from the 1830s (Blum et al 2017). …”

Age heaping and numeracy: looking behind the curtain

“…Stolz, Baten, and Botelho, ‘Growth effects’; Blum, Colvin, McAtackney, and McLaughlin, ‘Uncertain age’; Baten and Sirbiladze, ‘Elitegruppe’.…”