Recall that I argued that because the decline in hours worked between the 1890s and 1973 was largest among men earning the lowest wages, most of the decrease in the mean length of the work day in this time period probably came from disproportionate declines in the hours of men in the lowest deciles of the wage distribution. Therefore the mean length of the work day provides a good indicator of inequality in hours worked and testing whether hours legislation affected the distribution of hours worked becomes a test of whether hours legislation affected the average length of the work day. I will therefore examine whether the decline in the length of the work day coincides with the adoption of maximum hours legislation.
As previously noted, most of the decline in the length of the work day probably occurred by 1920. But, prior to the 1930s state legislation restricting maximum hours of work applied only to women and to relatively few men in dangerous industries. Federal legislation applied only to railroad workers. Although hours of work in 1920 were lower in states with hours legislation, hours of work were lower for men as well, even in industries where there were virtually no female employees, suggesting that hours fell where workers favored decreased hours and that the states where workers favored decreased hours passed hours legislation (Goldin 1988). There is some evidence from the Kansas surveys of 1897 and 1899 that hours were low where workers favored hours legislation. Although 73 percent of workers favored hours legislation, 44 percent favored an 8 hour day and 38 percent a ten hour day. Those who worked a longer day were more likely to favor a 10 hour day or longer and those who worked a shorter day an 8 hour day or less.
There may be some role for federal action during World War I in hastening the decline in the length of the work day. But, although the War Labor Policies Board and the War Labor.
Board required the adoption of the 8 hour day for contract work, the work day did not return to its previously high levels after the war.