We have so far analysed the aggregate stock of unemployment and the problems associated with its measurement. How important is this statistic? It certainly conceals as much as it reveals. It provides no clue as to the differential incidence of unemployment amongst different groups of the labour force. Moreover, it tells us nothing about the rate of turnover of the stock. If, however, we are interested in assessing the costs of unemployment and the need for corrective action these are crucial bits of information. For example, in chapter one we explored the relation between the costs of unemployment and its duration. It is clear from that analysis that we should take a different view of, say, a 10% unemployment rate if we know that the stock was made up of a large number of short duration spells (high turnover) than if it consisted of a smaller number of long term spells (low turnover). Equally if the short spells were experienced by a large number of different workers rather than by a small number of individuals experiencing repeated spells of unemployment, we would come to different conclusions about both the cost of total unemployment and the necessity for and the form of the policy action required to correct it. Similarly if all the unemployment is concentrated on, say, a particular region both the need for and the form of corrective action would differ from a situation in which the geographical spread is more even. The degree of concentration of the stock and the turnover within it are, therefore, critical aspects of unemployment that will be explored in this chapter.