The labour share of income in national product has shown a declining trend in many advanced economies over the past 30 years. However, many economists still hold the view that the wage share remains almost constant in the long run. The notion of the relative stability of the wage share in the long run is considers to be a stylized fact or even sometimes called a law of economics. This paper attempts to show how the alleged stability of the labour share of income became known as one of the great magnitudes in economics. It also shows how this law made its way into the three major theories of macroeconomic income distribution, i.e. neoclassical, post-Keynesian, and Kaleckian distribution theory. Since the data show strong fluctuation of aggregate income shares over the long run, the conclusion is reached that the major macroeconomic theories of growth and distribution are built around an invalid or at least highly questionable assumption about the real world.