Trend breaks in money growth and the money-output relation in the US

Stock and Watson (1989) demonstrate that contradictory money-output causali ty results can be resolved by appealing to the trend properties of money. T hey find that money growth exhibit a time trend, and detrended money growth causes output. Imcorporating recent experience suggests that the trend has vanished, in which case empirical results support the conclusion that mone y does not cause output. In this paper, we argue that detrended money growt h is a capable measure of policy but that trend breaks occurred in 1984 and 1991. Using detrended money growth that incorporates trend breaks, we find that money does indeed cause output.