Social science vs. hard science

Jim Manzi has an interesting discussion about the limits of social science, particularly economics, relative to hard sciences:

When it comes to deciding what policy actions to take, we should listen carefully to what economists and other social scientists say, but we should treat their assertions differently than we do predictions arrived at via experiments. We should subject them to useful cross-examination by specialists in other fields, reflect on how to weigh technical and non-technical opinions, ponder human motivation, and all the rest.

He's right, of course, but I think he doesn't make a clear enough distinction between economists' predictions of the direction of effects, and the precise magnitude of effects. Often, the magnitude is not that relevant.  For example, as reported by Greg Mankiw, 93 percent of economists agree that "(a) ceiling on rents reduces the quantity and quality of housing available." If we know that rent controls are a bad idea and, therefore, should not be implemented, how important is it to know precisely how bad they are? Just don't use them!