In recent years, the U.S. has seemed to achieve the best of all possible worlds: robust economic growth, very low unemployment, and low inflation. Many would attribute this performance to fewer supply side constraints, as the U.S. has moved away from stifling regulations and other impediments to trade. Indeed, our lower unemployment rates - especially when compared with the very high unemployment rates suffered in European countries - would appear to be due to freer labor markets and to a less generous social safety net that saps private initiative. However, in this paper we show that while it is true that the U.S. has enjoyed a higher, and growing, employment rate than that of all of our major competitors, per capita GDP growth since 1970 actually lags behind that of all other major countries. The reason, of course, is our dismal rate of productivity growth. Indeed, we show that when one decomposes per capita GDP growth into its component parts - growth of employment rates and growth of output per employee - the U.S. experience is quite different from that of the other countries. In some sense, countries "choose" high employment paths or low employment paths, but regardless of that choice, economic growth does not appear to be much affected. We argue that this is because countries have not faced significant supply constraints; rather, per capita GDP growth has been largely demand constrained. This is why policies that would remove supply constraints do not really lead to more rapid economic growth. The policy conclusion is that Keynesian "demand side" policies are preferable to "supply side" policies if one is to increase growth rates.