Without intraperiod tax allocation, NCFO is understated by the tax on gains and overstated by the tax savings on losses related to investing and financing activities.
This way, users are apprised of the tax effects of these gains and losses on reported NCFO. In the example above, transparency is increased by disclosing that income tax payments include $12,000 on a plant disposal gain and that disposal proceeds are reported as an investing inflow.
For example, Sinclair Broadcast Group, Inc., discloses in the management discussion and analysis (MD&A) section of its 2000 Form 10-K that the $69.1 million reported NCFO for 2000 is after deducting $115.1 million of income tax payments on the sale of its radio broadcast assets, which proceeds were reported as investing inflows.
Similarly, in the cash flow statement included in its 2001 Form 10-K, Delmarva Power & Light Company (DPL) reported a $60,749,000 NCFO and a $528,215,000 investing inflow from sales of electric generating plants in 2001.
For 2000, it also reported NCFO of $36,603,000 from continuing operations and an outflow of $327,000 from discontinued operations, for a total NCFO of $36,276,000.
Presumably, Duquesne Light and Amcol allocate income taxes in the cash flow statements because reported NCFO would otherwise be misleading, but neither discloses that fact.
H3: Aggregate NCFI has incremental information content to corporate bondholders conditioned on total accruals, aggregate NCFO, and aggregate NCFF.
H4: Aggregate NCFF has incremental information content to corporate bondholders conditioned on total accruals, aggregate NCFO, and aggregate NCFI.
These definitions for NCFO, NCFI, and NCFF, as well as specific Compustat PC Plus data items, are reported in the Appendix.
Hypothesis one examined the association between NCFO and annual holding period returns on corporate bonds, given total accruals, NCFI, and NCFF.
Hypothesis two examined the association between bond returns and total accruals, given NCFO, NCFI, and NCFF.
Variance inflation factors exceeded the rule-of-thumb cutoff of ten for both experimental variables NCFO and TACC, indicating potential problems from multicollinearity (Neter et al., 1990).