1 Typescript forthcoming in: P. Erdkamp & K. Verboven (eds.), Structure and performance in the Roman economy: models, methods and case studies, Brussels & Leuven: Collection Latomus 2015 Structural Determinants of Economic Performance in the Roman world and early-modern Europe A Comparative Approach Paul Erdkamp Vrije Universiteit Brussel Introduction: evidence and models Although archaeologists and historians have extracted a veritable mass of data on the economy of the Roman world from a wide range of sources, it is an understatement to say that the study of the Roman economy is hampered by a lack of data on some of the core issues in economic analysis, such as population, division of labour, and trends in wages and prices. Symptomatic of our ignorance are the extremely wide margins in the debate on the population size of such well-documented regions as Italy or Egypt. Market prices of grain in the entire Roman Empire over a period of centuries can be counted on the fingers of two hands, while division of labour has to be studied by counting the number of occupations mentioned on gravestones. When historians working on the economy of early modern Europe complain about the difficulties of their data on population sizes and densities, mortality and fertility, wages and living standards, they may expect little sympathy from ancient historians, whose response can merely ďe if oŶlLJ…. Archaeologists offer many proxy-data that partly help to circumvent this shortage, varying from shipwrecks to olive presses and pottery sherds, 1 but sufficient clarity about the economic context of these data is often lacking. A basic problem in evaluating this kind of evidence is that each of these proxy-data can only be assessed relatively, not absolutely. In other words, each phenomenon is plotted on its own map, with its own, unknown scale, which hampers our attempt to integrate these individual figures into one map of Roman economic development. In his chapter, Jeroen Poblome discusses the problems in interpreting the scales and processes in the Roman economy on the basis of archaeological data, stressing the need for interpretative models. The problems of the textual evidence hardly require elaboration: the written evidence on 1 See also Jongman, in this volume.