Purpose: The purpose of this paper is to consider whether the shift from training to learning is related to employment categories using a categorisation popularised by Robert Reich. Design/methodology/approach: Collation and analysis of existing CIPD research information and assessment of labour statistics. Findings: An examination of the national data on training reveals an uncertain picture. There is a problem surrounding data capture. In short, what is measured is training, particularly off-the-job training. If the definition of learning as a process is accepted, there are both conceptual and practical problems involved in measuring the move to learning in different sectors. Practical implications: The testing of this proposition and analysis of its consequences is of critical importance. It will affect public policy--what interventions promote learning as opposed to training? It should assist in determining the research agenda. Finally it central to the role of HRD practitioners in organisations: how do they develop and implement strategies for learning as opposed to delivering training? Originality/value: As far as the authors are aware no-one else has explored this topic using this approach. It is a speculative investigation at an early stage but the implications are considerable. (Contains 1 table.)