Discussion Paper

No. 2017-30 | June 09, 2017
A replication of ‘Education and catch-up in the Industrial Revolution’ (American Economic Journal: Macroeconomics, 2011)

Abstract

Although European economic history provides essentially no support for the view that education of the general population has a positive causal effect on economic growth, a recent paper by Becker, Hornung and Woessmann (Education and catch-up in the Industrial Revolution, 2011) claims that such education had a significant impact on Prussian industrialisation. The author shows that the instrumental variable they use to identify the causal effect of education is correlated with variables that influenced industrialisation but were omitted from their regression models. Once this specification error is corrected, the evidence shows that education of the general population had, if anything, a negative causal impact on industrialisation in Prussia.

Data Set

JEL Classification:

I25, N13, N63, O14

Assessment

  • Downloads: 277

Links

Cite As

Jeremy Edwards (2017). A replication of ‘Education and catch-up in the Industrial Revolution’ (American Economic Journal: Macroeconomics, 2011). Economics Discussion Papers, No 2017-30, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2017-30


Comments and Questions


Anonymous - Referee report 1
July 10, 2017 - 10:00

The original paper, Becker, Hornung and Woessermann's "Education and Catch-Up in the Industrial Revolution" (BHW) made a striking claim: that human capital played a casual role in the industrial revolution. While this claim appears as an assumption at the heart of much growth modelling, there is little or no historical ...[more]

... evidence to support the idea.

The replication challenges this result by noting that BHW have used an IV strategy that does not, in fact, allow them to estimate the causal impact of education on industrialization.

BHW have two different cross sections. In 1849 their measure of human capital is years of schooling; in 1882, it is the literacy rate as of 1871. In both cases they use the education level as of 1816 (which they argue is "pre-industrial") as the instrument for education in 1849 and 1871. The replication argues convincingly that BHW omitted several important right-hand side variables, and shows that all of these are correlated with the instrument. Adding these variables to the model yields a very different result: if education levels have an effect on industrialization, the effect is negative, not positive.
Many replications focus on finding some alternative specification that yields a different result, thus showing that the original results are fragile. That is not the case here. The replication author clearly knows a great deal about Prussian economic history, and uses that knowledge to think about the way the instrument works and what might be missing. The most important omitted variables in the new models are regional effects, which most economic histories of Prussia stress. Much of Prussia in 1849 had only been Prussian since 1815. Where and how these provinces came to Prussia mattered for their internal development as well as the integration into a larger Prussian economy.

It is worth noting, as does the replication, that part of BHW lacked any plausibility and thus represents a referee/editor error in the original publication. The first-stage F-stat for the instrument in the 1849 model exceeds 5500. This with a dataset that has 334 observations. As specified, the instrument is not so much an instrument as the endogenous regressor with some measurement error.


Jeremy Edwards - Response to referee report 1
August 16, 2017 - 10:44

see attached file


Anonymous - Referee report 2
August 09, 2017 - 09:29

see attached file


Jeremy Edwards - Response to referee report 2
August 16, 2017 - 10:44

see attached file


Anonymous - Referee report 3
September 07, 2017 - 10:28

The manuscript offers an important criticism of a well-known notion that there is a relationship between education and industrialisation in Prussia. It carefully presents the arguments of the criticised paper, offers a rebuttal which is fair, constructive, and opens new avenues for future research.