My co-author, Scott Stern, visited this week and came up with an interesting activity. He took a random book from my bookshelf, opened it to a page and looked at what it said. Turns out that can be both amusing and interesting. I may do the same exercise on a regular basis. 

The book he picked was Irving Fisher’s The Making of Index Numbers 1922 (3rd Edition, 1967). Why did I have this book? I like old books. Scott opened up to page 321 in a chapter entitled “Speed of Calculation.” There Fisher describes how he produced certain tables:

Hirtherto we have ignored the very practical question of speed and ease of calculation. Table 44 gives the results of time studies for calculating the index numbers of prices by various formulae. The table is constructed on the assumption of 36 prices and quantities supplied to the computer. He is furnished with a computing machine and logarithmic tables. The time requires to construct index numbers for either prices or quantities for the years 1914-1918 by Formula 51 (fixed base) is taken as unity. In the case of the particular computer who gave himself to these time studies, Formula 51 required 56 minutes. As he was probably slightly more rapid than the average computer, we may think of the time for 51 as one hour, and of all the other figures in the table as, therefore, representing hours. In every case the time calculation was that required to calculate the five index numbers, to two decimal places.

The big deal here is that the computer is a person. Clearly such a job is obsolete today but the title remains. Pure substitution of labour for capital.