November 12, 2003

The Pre-Civil War Standard of Living

What was going on before the Civil War with respect to the American standard of living?

"Development, Health, Nutrition, and Mortality: The Case of the 'Antebellum Puzzle' in the United States" by Michael R. Haines, Lee A. Craig, Thomas Weiss (NBER Working Paper No. h0130, Issued in).

Abstract: The 'Antebellum Puzzle' describes the situation of declining stature and rising mortality in the three decades prior to the American Civil War (1861-65). It is labeled a puzzle, since this period was one of rapid economic growth and development in the United States. Much of the debate regarding this puzzle has centered on whether the American diet, both in terms of protein and caloric intake in the mid-nineteenth century. But the mortality environment also appears to have worsened (or at least failed to improve), a situation associated with rapid urbanization, commercialization, transport improvement, and increased geographic mobility. The disease environment was being nationalized and internationalized. This paper analyzes the relationship between local agricultural surpluses, nutritional status, mortality conditions, and adult heights. Employing a sample of the muster records of Union Army recruits (1861-65) as well as data from the published population and agricultural censuses of 1840 and mortality data from the 1850 census of population, it tests the hypothesis that adult height is positively correlated with local production of nutrients in early childhood and negatively correlated with local mortality conditions, urbanization, proximity to transport, and population mobility. Results indicate that, although the United States was experiencing robust Smithian' economic growth induced by transport improvements and widening markets nation was also suffering from serious negative externalities which affected the health and longevity of the population.

David Levine writes that he is doing similar analyses for Indonesia, Mexico, and China:

...heights fell during the first two generations of industrialization in the UK & the US (although not most other nations). At the same time, the US height decline during the generations of industrialization also occurred in the south -- where industrial structure was fairly static. It seems sensible to break up the national data by region and see if the height decline was concentrated where factories were located: in New England first, then to NY and PA, then to the Midwest, and finally to the south. The first part of this study can be performed with the Civil War dataset. I propose to use double-difference method comparing height changes in regions that did and did not industrialize.

Intuitively, if industrialization is responsible for height declines, then heights of natives should be similar in 1810 MA, 1810 ME or PA (without them) and 1840 ME (all with few factories), but shorter in 1840 MA (with factories). If there are fixed differences across states or across decades, then the best estimate of how factories affect height is the double difference of regional means: (1840 MA - 1810 MA) - (1840 ME - 1810 ME). Additional studies can examine adult health on war survivors, but that is more complex....

A key part of the study is linking in data on manufacturing employment by state & decade to get the timing right and to incorporate more states. It is also important to study urbanization, a separate process from industrialization, but related.... I am doing similar studies with recent data from Indonesia, Mexico, and China...

Anybody still looking for an economic history paper topic? There's lots to do in addition to what Haines, Weiss, and Craig did.

Union Army Dataset:

Posted by DeLong at November 12, 2003 05:01 PM | TrackBack


There's been lots of speculation on this very topic for years, and most of it comes out about where Haines, Weiss, and Craig do. I always found it interesting stuff. Rick Steckel from OSU has also been doing work in the area for quite some time.

Posted by: VJ on November 14, 2003 12:46 AM


Post a comment