Labor supply in the past, present, and future: a balanced-growth perspective

We argue that a stable utility function of consumption and hours worked for which income effects are slightly stronger than substitution effects can rationalize the long-run data for the main macroeconomic quantities. In these long-run data, in the U.S. as well as in other countries, as productivity grows at a steady rate, hours worked fall slowly and at an approximately constant rate. We narrow down the set of preferences consistent with balanced growth under constant (negative) hours growth. The resulting class amounts to a slight enlargement of the well-known “balanced-growth preferences†that dominate the macro literature and are based on requiring constant hours worked. Thus, hours falling at a constant rate is not inconsistent with the remaining balanced-growth facts but merely requires a slight broadening of the preference class considered. The broadening of the preference class introduces some well-known cases not previously thought to be consistent with balanced growth. From our perspective, we interpret the recent decades of stationary hours worked in the U.S. as a temporary departure from a long-run pattern, and to the extent productivity will keep growing, we predict that hours will fall further.