Heuristic admissibility and A* optimality

You are running A* on a grid-based occupancy map where the robot may move in the 4 cardinal directions only (no diagonals), with uniform edge cost 1 per step. A teammate, hoping to speed up the search, replaces the Manhattan-distance heuristic with the straight-line Euclidean distance to the goal, scaled by a factor of 1.5 (i.e. $h(n) = 1.5 \cdot \lVert n - goal \rVert_2$). Which statement best describes the effect on A*'s behavior?