In a well-known paper, Christopher Sims [10] presented a regression method for testing direction of causation in the sense of Granger [6] between two economic time series. He applied his test method to quarterly U.S. nominal GNP and money stock data of the 19471969 period and obtained results which are consistent with the hypothesis that money causes income without any feedback from income to money. Sims's paper stimulated other economists to test the direction of causation between money and aggregate income in the United States as well as in other countries [1; 3; 4; 12]. Either Sims's exact regression procedure or a modified version of his procedure was used in most of these studies, while a residual cross-correlation test procedure was used in at least one such study. In general, the results of these studies are not in line with Sims's finding for the United States in that they do not support the hypothesis of unidirectional causality running from money to income. Our purpose in undertaking this study was to test for the direction of causation between money and aggregate income in all the developed market economies for which we could obtain quarterly data series of more than ten years in length. There were six such countries: Australia, Canada, Germany, Japan, the United Kingdom, and the United States. We also wanted to obtain some empirical evidence on whether different test procedures applied to the same data might yield conflicting results. Consequently, we used three test procedures: Sims's exact regression procedure, a modified version of Sims's regression procedure, and a residual cross-correlation test method developed by Haugh [7].