Hopkins (1978) and then Finley (1980 and following) had famously characterized the Roman world as a slave society and therefore a non-market economy. This view has now been firmly dismissed by the last thirty years of research on ancient economies.

There are now strong arguments, from many different scholars, that Rome facilitated a fully-operational labor market. Slaves were an element of that labor force, but not the driving element, and there was always (as far as we can detect) an available and functioning pool of hired labor in many different aspects of economic life, including the mines. Slaves were not the de facto preference for any task, and in fact in many instances were detrimental as far as cost and availability. Sudden infusions of new slaves (like during the 50s BCE, via Caesar's activities in Gaul) must have caused temporary shifts in the overall labor market, but overall, through multiple centuries, the labor economy of the empire remained relatively stable despite the erratic "production" of slave labor pools (if I can term it that).

Most free laborers were farmers, and the status of free laborer between free and slave was often fluid. If a tenant farmer "got behind" on his debt due to a succession of bad harvests, he might go into slavery voluntarily for a period to relieve that debt. We have several examples of such contracts, and also examples of a myriad of other free labor contracts (including mining workers). One from 194 CE shows a gold miner working for several months at a rate of 70 denarii per day (plus board), and subject to prorated days in case of flooding.

For a detailed discussion of the Roman labor economy, see Chapter 6 "The Labor Market" in Peter Temin, The Roman Market Economy (Princeton 2013).