Hoards of denarii are common in Britain and the number which have been recorded in detail means that it is now possible to suggest reasonably accurately what a ‘normal’ hoard of a particular date should look like. That being the case, we can then look for variation around that norm and both investigate and speculate what that variation means. A methodology is developed which suggests periods of faster and less rapid coin circulation which has implications for consideration of monetisation. The model also enables us to view where denarii entered circulation; unsurprisingly the army looms large in this picture. The methodology is directly transferable to other provinces and other periods where there are long-lived, relatively stable monetary systems.

The consensus view that all Roman money consisted of coins has been undermined in recent times and should be discarded. The inhabitants of the Roman Empire frequently and on a significant scale made payments by means of credit-money, creating a ‘multiplier effect’, which meant that in high classical times the Roman economy was not constricted, as is often supposed, by an inelastic money-supply. Yet the monetary system was not modern; rather it had its counterparts in such economies as those of seventeenth-and early eighteenth-century Britain.