Friday, January 31, 2014

Were Nominal Wages Flexible in 1890s and Early 1900s America?

There is some empirical evidence that they were not.

Sundstrom (1990) looks at industrial wage data from the Ohio State Bureau of Labor Statistics (BLS) during the recessions of 1893 and 1908 and finds considerable wage rigidity even at that time.

In 1893, Cincinnati manufacturing employment contracted by over 16% and factories reduced working days by 9.2% between 1892 and 1893, but Sundstrom found that 77.1% of manufacturing workers experienced no change in nominal wages, and what evidence that exists for 1894 and 1895 also indicates that wage cuts were small in these years too (Sundstrom 1990: 312–313).

Given that deflation occurred in 1893 until 1898, this would suggest that many workers’ real wages probably rose.

In the data for the severe recession of 1908, Sundstrom (1990: 314) finds that, throughout Ohio, 6.37% of industrial workers experienced wage increases, while only 7.36% experienced wage cuts. This suggests that most workers experienced no nominal wage cuts.

Overall, Sundstrom (1990: 314) finds that nominal average hourly earnings only fell by about 1% between 1907 and 1908.

Sundstrom (1990: 310) concludes that as “early as the 1890s, Ohio employers were much more likely to respond to downward demand fluctuations by reducing employment, days worked, and hours than by reducing wage rates.”

To put this into a broader historical context, we can turn to Hanes (1993). Hanes concludes that 19th century American nominal wages had already become relatively inflexible by the 1890s (Hanes 1993: 733–734).

His explanation for this is as follows:
“The years between the … [sc. American Civil War] and World War I saw little or no increase in the fraction of workers belonging to unions. Countercyclical macro policy did not exist; wage contracts were extremely rare, and unenforcible. On the other hand, the period was one of enormous change in the structure of product and labor markets, associated with the spread of large-scale industrial production. Production workers in the new industrial establishments, whether or not they were formally unionized, were likely to strike against nominal wage cuts in downturns. I argue that firms learned to avoid, or at least delay, nominal wage cuts in downturns. I present evidence that firms in industries that had suffered especially large numbers of strikes in the 1880’s were less likely to cut nominal wages in the depression of 1893.” (Hanes 1993: 733–734).
The crucial point here is that this development cannot be blamed on trade unions because they were weak in the late 19th century in terms of numbers and membership, they faced hostility from the courts and government, and even when they existed they could not generally create binding legal employment contracts with employers, because the courts did not recognise them (Hanes 1993: 750–751).

Hanes (1993: 751) contends that strikes were a widespread phenomenon amongst non-unionised workers, and that it was the spread of large-scale manufacturing firms with numerous workers that was itself that main factor that caused labour strife and the increasing wage rigidity accepted by private firms.

What is the lesson? It is that people, generally speaking, have shown a strong propensity to oppose nominal wages cuts since at least the late 19th century, and private businesses themselves – even with the incredible amount of violence against labour in those days – were coming to shun nominal wage cuts to avoid troublesome labour difficulties.

Nor can trade unions or governments be blamed for this phenomenon.

Sundstrom, William A. 1990. “Was There a Golden Age of Flexible Wages? Evidence from Ohio Manufacturing, 1892–1910,” The Journal of Economic History 50.2: 309–320.

Hanes, Christopher. 1993. “The Development of Nominal Wage Rigidity in the Late 19th Century,” The American Economic Review 83.4: 732–756.


  1. Off topic. But this might interest you...