The Effect of Cash Flow Presentation Method on Investors' Forecast of Future Cash Flows

The decision usefulness of the direct versus indirect presentation method of a cash flow statement has been a long-standing issue in both practice and accounting research. By capitalizing on comparative advantages of experimental methods, we provide insights into how investors process the information contained in different presentation methods to make cash flow forecasts, especially in the context of various types of nonrecurring items. We predict and find that, when nonrecurring accrued expenses are present, the indirect method leads to lower forecast errors by activating investors’ underlying knowledge structures of operating cash flows in terms of an accrual (versus cash) basis. We also find that, in the presence of nonrecurring cash or nonrecurring accrued revenues, there is no difference in forecast errors between the indirect and direct methods. Moreover, we find that the combination of the direct and indirect methods (the direct-plus-indirect method) leads to lower forecast errors than the direct method, but it does not provide an incremental benefit for forecast accuracy beyond the indirect method. This paper was accepted by Brian Bushee, accounting.