Based on within-sample and out-of-sample tests on 348 listed Australian companies over 1992-2004, we find that direct method cash flow components have greater predictive ability for future cash flows than for aggregate operating cash flows. Unlike prior research, we use actual direct method cash flow data instead of estimated data in our forecasting models. The results are robust to a battery of sensitivity tests, including control for firm size, industry membership, profitability, negative cash flows, and the length of the operating cash cycle. Moreover, our results remain qualitatively similar for forecast horizons up to six years. Our findings directly contribute to the policy debate on whether standard setters should mandate direct method cash flow statements.