The Time-Series Properties of Quarterly Cash Flows: Working Paper Series--09-12

Considerable advancements in the structural modeling of annual cash flow prediction models have been accomplished in recent years [Dechow et al. (1998) and Barth et al. (2001), among others].Yet, the modeling of quarterly cash flow data has not been as forthcoming due to: (1) the unavailability of sufficiently long time-series data bases of quarterly cash flows reported in accordance with SFAS No. 95 and (2) the presence of seasonality. We provide new empirical findings supportive of the Brown-Rozeff ARIMA model as a candidate statistically-based expectation model for multi-period ahead projections of quarterly cash flows. The Brown-Rozeff ARIMA model provides one-thru-twenty step-ahead projections of quarterly cash flows that are significantly more accurate than those generated by a quarterly time-series, disaggregated-accrual regression model originally popularized by Lorek and Willinger (1996). Although both quarterly earnings and quarterly cash flow from operations are modeled by the same ARIMA structure, we find that the autoregressive and seasonal moving-average parameters of the quarterly earnings model are significantly larger than those of the cash-flow prediction model. This finding is consistent with Beaver (1970) who argues that short-term and long-term accruals induce incremental amounts of serial correlation in the quarterly earnings time series vis-a-vis the time series of quarterly cash flows. These findings are of interest to standard-setting bodies seeking to understand the linkages between accruals and cash flows, analysts who wish to derive multi-step ahead cash flow predictions, and accounting researchers attempting to adopt a statistical proxy for the market's expectation of quarterly cash flows