OCF classification choices
Footnote 26 The number of firms is reduced from 798 to 538 because the following are excluded: firms that changed their classification choice during the period; firms from Denmark, Finland, and Sweden (where classification choices for interest paid and interest received exhibit little or no variation); and firms missing data to compute all independent variables.
Results of the regression using differences in OCF as the dependent variable are presented in the left columns of Table 6, panel B. A higher value of the differences in OCF variable, OCF_Reported less OCF_Pro forma_USGAAP, signifies a greater OCF-enhancing impact of classification choices that differ from the hypothetical benchmark. As expected, we find that Distress_Hi (an indicator variable signifying greater likelihood of financial distress) is positively and significantly related to OCF_Reported less OCF_Pro forma_USGAAP. This finding suggests that financially distressed firms make more OCF-increasing classification choices. Equity Issues is also positive and significant, suggesting that firms that access equity markets more frequently opt to make classification choices to report higher OCF. Leverage_Hi is also significantly positive, indicating that firms with greater leverage are more likely to make classification choices to show higher OCF. Profitability is significantly negative indicating that less profitable firms are more likely to make OCF-enhancing classification choices. In terms of economic significance, our results indicate that one standard deviation in the variables Equity Issues and Profitability corresponds to a change in the dependent variable (OCF_Reported_less OCF_Pro forma_USGAAP) of $10.0 million and $7.9 million, respectively (corresponding to approximately 0.1% of average assets). Similarly, one standard deviation in the dichotomous variables Distress_Hi and Leverage High corresponds to a change in OCF_Reported_less OCF_Pro forma_USGAAP of $11.3 million and $8.3 million, respectively.
Finally, size is negative and significant. Neither Analysts Cash Flow Forecast, Industry Homogeneity, Cross-listed in US nor any of the industry indicator variables (not tabulated) are significant. Country indicator variables are all negative and significant with p-values below 0.01. Footnote 27
The right columns of Table 6, panel B, present the results of estimating the logistic regression, where the classification choice to report interest paid in financing is the dependent variable. Similar to the results of the OLS regression, Leverage_Hi is positively and significantly associated with the choice to classify interest paid in financing. This result implies that highly leveraged firms are 44.2% more likely to make an OCF-increasing classification choice of interest paid as financing. In addition, Cross-listed in US is negative and significant, indicating that firms with cross-listings are more likely to follow the classification choices permitted for US firms. Specifically, US cross-listed firms in our sample classify interest paid under financing activities with 55.8% lower likelihood than non-US cross-listed firms. Footnote 28
Neither Distress_Hi, Equity Issues, Profitability, Analysts Cash Flow Forecast, Industry Homogeneity, Size nor any of the industry indicator variables (not tabulated) are statistically significant. Country indicator variables are all negative and significant with p-values lower than 0.01. The finding that country predicts classification choice while industry does not could reflect firms’ view of their relevant peer group. Despite political and accounting-standard union, payday loans in Maynardville country membership dominates as a predictor of accounting choice within allowable alternatives.
Changes in OCF classification choices
In our sample, 99 firms, or 12%, reclassify interest paid, interest received, or dividends received within the statement of cash flows during the sample period. Table 7, panel A, shows that the 99 changers represent all countries except Portugal. The most changers were in the United Kingdom (24) and Germany (17), and the highest percentages of firms making a change (26%) were in Norway (11 of 43 firms) and Spain (15 of 58 firms). The majority of firms (58%) increase OCF in the year of the change, increasing OCF by 1.20% (0.78%) at the mean (median). Companies in all industries, except chemicals, made changes, with the greatest number in services (14) (not tabulated). Among the reclassifications affecting OCF, the greatest number move interest paid out of OCF. As shown in Table 7, panel B, the majority of these firms (49) changed OCF through reclassifying interest out of OCF.