Among unmarried, childless minimum integration income recipients, those who receive housing benefit go back to work more often than the others. This finding may seem surprising. Housing benefit recipients have a higher total income. Moreover, housing benefit is highly redistributive and the gains obtained from going back to work are often reduced by the parallel decrease in the benefit. This could dissuade jobseekers from taking a job, especially when it is part time. Yet there is a dual explanation for this finding. It is firstly due to the incentive mechanism, whereby minimum integration income is combined with earned income in the short term, and the fact that housing benefit does not cover all the rent, since part of it ­ the remainder to be paid ­ is to be paid by the recipient. Standard economic theory predicts that this person will try to work more to pay the remainder. This is the factor, and not the housing benefit amount, that proves to be the determining economic variable. Secondly, the benefit is not reduced as soon as the individual finds a new job. The time lag before this happens can be as long as eighteen months. Solely the effects on the initial months back at work are analysed here. There is not enough data to answer the question regarding the benefit's longer-term effect on work. Yet it has to be said that this purely economic explanation ignores the essential objective of housing benefit: to help its recipient attain or maintain residential autonomy, which is a factor for economic and social integration. This sample is actually found to have very high residential mobility, which cannot be understood accurately enough from the data available. A joint analysis of residential and work paths would shed more light on the link between housing benefit and work.