In a VAR model of the US, the response of the relative price of durables to a monetary contraction is either flat or mildly positive. It significantly falls only if narrowly defined as the ratio between new house and nondurables prices. These findings survive three identification strategies and across subsamples. Then, they are rationalized via the estimation of a two-sector New-Keynesian model. Here, the degree of overall durables price stickiness is not dramatically lower than that of nondurables. Such macroeconometric results are close to recent microeconometric evidence. Moreover, they suggest that monetary policy is not very distortive of sectoral allocations.