The consequences of a demand constraint (low willingness to pay) are examined in a model of merger by spatial price discriminators. The imposition of a demand constraint reduces the extent of inefficiency associated with merger and also eliminates the resolution of the merger paradox obtained in the earlier, unconstrained case. Moreover, with the introduction of a demand constraint, a tax on transport cost can actually improve efficiency, which is never the case in the absence of the demand constraint. Indeed, the optimal tax often eliminates all spatial price competition by creating local monopolies. (JEL D43, L41) *Affiliations of the authors are respectively Department