Chong, KW 2012, 'Prediction of operating cash flows using accrual-based and cash-based accounting information among Malaysian industrial corporations', DBA thesis, Southern Cross University, Lismore, NSW.
Copyright KW Chong 2012
This study investigated the comparative abilities of accrual-based and cash-based accounting information for future operating cash flows prediction. It aimed to explain the extent earnings, accruals, EBITDA and operating cash flows are able to predict listed Malaysian industrial products manufacturing corporations’ operating cash flows. Ordinary least squares method was used to develop fifteen multiple linear regression models in examining up to three years of prediction horizon. Current operating cash flows, as proxy for future operating cash flows, are regressed on past one, two and three years of earnings, operating cash flows, operating cash flows combined with aggregated/disaggregated accruals and EBITDA as predictors. The sample was collected from annual financial statements of 97 listed corporations from 1999 to 2009 (11 years) and categorised into two subsamples; the within sample (1999 to 2008) to develop the models and the holdout sample (2009) to validate the robustness of the models.
Results from the regression analyses revealed that all the accounting measures examined have significant predictive abilities but differing powers. Earnings are superior to operating cash flows but EBITDA are stronger than both earnings and operating cash flows while the operating cash flows have the weakest predictive powers. The model comprising of both operating cash flows and disaggregated accrual components have the strongest predictive abilities. This suggests that disaggregating earnings into operating cash flows and accruals and further decomposing the aggregated accruals into the individual accrual components enhance the predictive abilities. These findings revealed that accruals have incremental explanatory powers beyond that contained in operating cash flows and disaggregated accrual components provided incremental explanatory powers exceeding that provided by aggregated accruals. Aggregated earnings masked the contributions of each accrual component. Disaggregating earnings and further decomposing the aggregated accruals unmask the contributions and effects of each accrual component. As individual predictors, both earnings and accruals have the significant abilities to predict future operating cash flows for only one year-ahead (short term), EBITDA can predict up to two years-ahead (medium term) and operating cash flows can predict up to three years-ahead (long term).