Applying Hsiao＇s version of the Granger causality method, this article makes the first attempt to examine causality between (1) energy and growth and (2) energy and employment, by applying the recently developed methods of cointegration and Hsiao＇s version of the Granger causality to transformed U.S. data for the period 1900-1945. Phillips-Perron tests reveal that the original series of energy data are not stationary, and therefore a first differencing is performed to induce stationarity. The study finds interestingly that there is causality unidiretionally running from energy to growth without feedback. It is also found that there are no causal link between energy and employment The results of this study ave consistent with some past studies using postwar-period data but contrary to some other past studies that covered the period after World War II.