The purpose of carrier systems is to save money by carrying more traffic on less infrastructure. 19th century telephone systems, operating at baseband, could only carry one telephone call on each wire, hence routes with heavy traffic needed many wires.
In the 1920s, frequency-division multiplexing could carry several circuits on the same balanced wires, and by the 1930s L-carrier and similar systems carried hundreds of calls at a time on coaxial cables.
Capacity of these systems increased in the middle of the century, while in the 1950s researchers began to take seriously the possibility of saving money on the terminal equipment by using time-division multiplexing. This work led to T-carrier and similar digital systems for local use.
Due to the shorter repeater spacings required by digital systems, long-distance still used FDM until the late 1970s when optical fiber was improved to the point that digital connections became the cheapest ones for all distances, short and long. By the end of the century, analog connections between and within telephone exchanges became rare.
Western Electric (1969) Fundamentals of Telephone Communication Systems, p.16.2 ↩