James Macksoud, Stanford University
In recent decades, attempts to quantify the expenditures of the Roman Imperial state have reached a consensus figure of per annum outlays near 1 billion sesterces during the 2nd century CE (Duncan-Jones 1994; Wolters 1999; Scheidel 2015). As part of the larger social science turn in ancient history, this work is important for debates concerning the nature, complexity, and size of the Roman economy (Bang 2008; Scheidel & Friesen 2009) as well as for comparative historical work that situates the Roman Empire alongside other premodern states and fiscal regimes (Monson & Scheidel 2015; Ando & Richardson 2017). However, within these studies, the fiscal operations of the empire’s cities, and particularly the scale of their revenues and expenditures, are often minimally addressed. This is unfortunate because without clarity concerning the scale of the fiscal operations of local governments, one cannot wholistically discuss the fiscal regime of the Roman Empire vis-à-vis those of peer polities, nor can one discern the extractive burden of government (the share of income or ‘GDP’ captured by local authorities and the Imperial state) during the early empire.
This paper aims to partially bridge this gap in scholarship by both offing a methodological approach to quantifying the expenditures of local governments as well as providing some provisional estimates for the average and aggregate outlays of cities in the first two centuries CE. To do so, the paper first discusses some previous efforts to estimate the expenditures of the empire’s cities (MacMullen 1974; Zuiderhoek 2009) and the complications resulting from the piecemeal evidence concerning civic outlays which has heretofore prevented robust modeling. The paper then presents the results of an extensive survey of the largely epigraphic material which offers price information concerning, inter alia, construction costs of public buildings, expenses for staging ludi, and wages for public employees (Liebenam 1900; Abbott & Johnson 1926; Duncan-Jones 1962). Although this price data relating to the expenses of local governments remains relatively scarce, this paper argues that this evidence can nevertheless be confidently used as a basis for estimating the expenditures of individual cities and of local authorities in aggregate. This is accomplished by using, on the one hand, architectural energetics (Abrams 1989; DeLaine 1997; Bernard 2018) to evaluate construction costs at those sites without price information and, on the other hand, by drawing upon recently aggregated archaeological data concerning Roman urbanism (Hanson 2016) to better contextualize case studies and local finds in relation to the wider urban network. The paper concludes by presenting case studies estimating annual expenses at Pompeii and Timgad as well as by offering a provisional estimate for the aggregate outlays of local governments in the 1st and 2nd centuries CE. This assessment shows that while the cities did extract a measurable share of Roman ‘GDP’ this figure was only a small fraction of that appropriated by the imperial state. If accepted, this conclusion both upholds existing models of the early Roman Empire as a low tax burden state while also illustrating its perhaps unusually diffuse fiscal regime.