The concept that changes in the quantity of money in an economy affect only nominal variables, as a change in units, and are irrelevant to the behaviour of rational economic agents. The neutrality of money was first identified by David Hume (1711–1776) in the 18th century and developed later into the quantity theory of money. Hume's prediction that in the long run nominal prices grow proportionally to the money supply has been confirmed empirically using data on different countries in different time periods. The ‘neutrality theorems’ derive this result in the framework of dynamic general equilibrium theory. The non-neutrality of money in the short run and the potential of monetary policy, in particular, monetary expansion, to stimulate real economic activity were the focus of Keynesian economics after the depression of the 1930s. In the 1970s the difference between the effects of anticipated and unanticipated money shocks and the role of rational expectations were emphasized and tested on both time series and cross-country data. The main finding was that anticipated changes in the growth rate of money supply are not associated with changes in employment and production and only have an inflation tax effect. On the other hand, unanticipated monetary expansions were found to be associated with economic booms, and contractions with depressions.