Nearly half of all tenants receive housing benefit in France. Economic theory predicts that giving tenants individual benefits can increase aggregate demand for housing and hence rents, at least in the short run. The long-run level of prices depends on the elasticity of supply. In the absence of a comprehensive model of the rental market and rent growth, it is hard to measure the benefit's specific effect on rents. We present three phenomena based on the natural experiment formed by the «completion» of benefits from 1992 to 1994 (i.e. their extension to new categories of tenants). At aggregate level, rents for housing whose tenants received benefits grew more sharply after the extension than rents for housing whose tenants received no benefits. The benefit helped the households to find better housing. However, this effect persists, albeit to a lesser extent, when rent growth is measured for equal qualities of housing. This means that there was also a pure increase in rents. Lastly, a panel analysis of rent growth rates shows that these rates are highest where there is a transition from a «non-assisted» status to an «assisted» status. This is checked for the entire period observed, regardless of any benefit completion. We find an effect whereby it is as if the landlord had taken advantage of the new benefit and recuperated part of it in the form of rent. However, the study does not allow for an accurate measurement of the effect of the benefit on the general level of rents.