In the past, as the ten-yearly cornerstone of the official series of annual employment estimates, the Population census in its traditional form held a key place in the system for monitoring employment in France. The introduction in 2004 of a new census system based on partial annual surveys rotating on a 5-year cycle prompted a rethink of this role. It may be less central, but it is nevertheless still no less essential. Although the census now has the disadvantage of measuring employment smoothed over 5 years, and is therefore trend-based, it is still an essential complement to the Annual Business Survey and the Labour Force Survey. Although its contribution is less valuable than that of the Labour Force Survey for analysing people's type of activity and describing jobs, the Census does provide a very detailed breakdown of its results, which is beyond the scope of the survey. Its place is justified by the fact that the notion of employment used in the census is similar to that of the ILO, measured by the Labour Force Survey and useful for international comparisons. The Census may overestimate employment levels slightly through being less precise than the survey in its questions; this is due to the presence of some fairly distinctive populations, for whom employment is more of a secondary or incidental situation of activity at certain stages in the life-cycle: students, young mothers, retired people. Despite this slight difference in level (around 1.1%), the employment trend measured by the census is very similar to that provided by the Labour Force Survey. At local level, the predominance of administrative sources (Annual Business Survey) in the short-term economic situation does not exhaust the usefulness of the Census resources, which still have more valuable information to describe people in jobs and their households. It is also less subject to problems of pinpointing the location of jobs, an issue that is not yet fully resolved in the administrative sources.