Although there has been much writing about ecosystem services in the last decade, there has been insufficient which clearly elucidates their value. Rather, much of the writing has been classificatory rather than analytical (e.g. the Millennium Ecosystem Assessment, Total Economic Value (TEV), and environmental accounting) and has tended to complicate rather than enlighten. Further, these classification systems are not necessarily internally consistent, nor consistent with each other. There is no systematic analysis of what economic values to report within these frameworks although the SEEA would account in value added terms, consistent with a national accounting framework, and TEV focuses on economic surpluses but is often inter-temporally muddled. This paper explores the necessary modelling that is required to properly identify the value of ecosystem services, and to distinguish them from "supporting" biophysical processes that only have indirect and derivative anthropogenic values. The paper also explores the limitations of environmental accounting - comparable to national accounting - as policy-relevant analysis.