Selective capital controls tax only some components of capital flows. One of the rationales for such controls is that they increase the scope for an independent monetary policy, without taxing foreign direct investment and other long term flows. The first part of this paper offers a new framework to evaluate how selective capital controls might increase monetary autonomy, which considers two types of capital flows that coexist: the taxed and exempt flows. It is found that under free floating selective controls increase monetary autonomy, in the sense of allowing the authorities to set the path of the nominal exchange rate. But under predetermined exchange rate rules, the contribution of selective controls to monetary autonomy depends of the ability to reduce total inflows, which is an empirical matter. The second part describes the Chilean unremunerated reserve requirement (URR), a selective control introduced in June 1991 on a permanent basis, in a setting of predetermined exchange rates. This control collected substantial revenue, proving that it was relevant. An econometric model with data for 1987–1996 finds that substitution from the exempt short-term flows compensated reductions in taxed short-term flows, so the Chilean URR did not discourage total net short-term credit inflows to the private sector. This implies that the Chilean URR failed to contribute to monetary autonomy.