Nick Rowe has a nice explanation of why output in the U.S. is generally demand determined, and why this isn't necessarily true in all countries. See here.
Just one peripheral comment. Nick says that it took "half a century for monopolistic competition to make the transition to macro." That is true, for a long time perfect competition was assumed in these models for analytical convenience. But I've always thought that Keynes had monopolistically competitive markets in mind when he wrote the General Theory, and that this came about due to the influence of Joan Robinson (her theory of monopolistic competition appeared in 1933), something that was lost in the macroeconomics literature until the revival of these models in the early 1980s and their incorporation into the New Keynesian framework.