![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
In this paper we exploit the choice allowed by International Financial Reporting Standards (IFRS) regarding the presentation of interest payments on the cash flow statement to answer two related questions: First, whether the classification choice is explained by firm reporting incentives and second, whether it is value relevant. Using a UK sample, we find that firms reporting losses, with a greater proportion of their debt stemming from public sources, with CFO-based covenants and greater increases in leverage in the year of adoption are less likely to report interest payments in cash flows from operating activities (CFOA). Results also suggest that the incentive to meet or beat analyst CFO forecasts decreases, but strong corporate governance increases the probability of including interest payments in CFOA. Based on the assumption that the decision not to classify interest payments in CFOA captures lower disclosure quality or poor future expected performance, we posit that these firms should also exhibit lower valuations. Results obtained after correcting for self-selection bias confirm this assertion. We conclude that managers’ decision not to classify interest payments in CFOA is consistent with the opportunistic use of the choice allowed by IFRS.
Acknowledgements
We would like to thank C. Louca, G. Nishiotis, and E. Neofytou for useful comments. Marios Markitsis and Kyriaki Fragkou provided excellent research assistance. Remaining errors are the responsibility of the authors.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. The classification choice of interest paid is rather sticky and is, at least to some extent, related to IASB requirements. According to IAS 7.31, ‘interest and dividends received and paid may be classified as operating, investing, or financing cash flows, provided that they are classified consistently from period to period’.
2. Refer to Ball (Citation2006), Soderstrom and Sun (Citation2007), Pope and McLeay (Citation2011), Brown (Citation2011), Brown and Tarca (Citation2012), and Brüggemann et al. (Citation2013) for a thorough review of the IFRS related literature.
3. Gordon et al. (Citation2013) also examine the presentation choices related to the statement of cash flows for a sample of firms from 13 European countries. Other than the fact that we focus our attention to UK firms only for the reasons explained above, our paper differs from theirs in another two important ways. First, in addition to examining financial distress as an incentive to include interest paid in CFFA we also examine whether this choice is affected by corporate governance characteristics that can significantly reduce the tendency to inflate CFO. Prior studies have shown that effective corporate governance mechanisms are related to increased disclosures and higher quality earnings (Karamanou and Vafeas Citation2005) suggesting that they could also affect a firm’s propensity to inflate CFO. Second, we examine how the market perceives this presentation choice by relating it to firm value. Even though it is important to first examine the incentives behind any financial statement presentation choices, whether these choices have capital market consequences is equally important, especially when assessing the effectiveness of new regulations.
4. Standard headings in FRS 1 are: Net cash from operating activities, Dividends from associates, Returns on investments and servicing of finance, Taxation, Capital expenditure, Acquisitions and disposals, Equity dividends paid, Management of liquid resources, and Financing.
7. http://simplestudies.com/what-are-debt-covenants.html; http://quickbooks.intuit.com/r/cash-flow/understanding-loan-covenants/
8. We are thankful to two anonymous reviewers for a number of suggestions that have significantly improved the development of the classification choice model.
9. Even though theory suggests that high audit fees may compromise auditor independence such adverse effect is generally not supported by empirical research (Craswell et al. Citation2002).
10. Our results are unchanged if the Altman Z-score is replaced by an indicator variable based on the cutoff point of 2.675 as commonly used in the literature, or even when we use a more conservative threshold of 1.81.
11. Throughout this study, we refer to the fiscal year prior to the switch as year 2004 and the year of the switch as 2005, even though for firms with fiscal years ending in any month other than December the first year of (prior to) the switch actually occurs in 2006 (2005).
12. There is a wide concern in the accounting literature with regards to the selection of best instrumental variables. We have tried to justify theoretically and empirically the selection of our instruments, however, we acknowledge that the exclusion restriction is always an important issue for the validity of the tests and inferences.
13. In untabulated tests we rerun our analysis by removing all interacted variables from the first stage model. These alternative specifications do not affect our main inferences.
14. After excluding financial firms from the initial sample, we have missing data for a total of 67 non-financial firms. For 40 of these firms the annual report is not available and for the remaining 27 the statement of cash flows does not include interest paid.
15. Given that some variables are included in both the classification choice and valuation models their correlations are shown in both panels for completeness. Some minor differences exist between the two panels due to the slightly smaller sample size of the valuation model.
16. In untabulated results we find that the choice to classify interest received in CFOA does not explain the classification choice of interest paid. We also find that the amount of interest received is significantly lower than the amount of interest paid. This implies that the impact of interest received on the firm’s cash flow is minimal and should not be expected to affect the reporting incentives associated with the classification of interest payments.
17. DMEET is set to 0 if analysts do not make CFO forecasts. If we drop this assumption the number of observations is reduced substantially but results remain qualitatively the same.
18. The model’s partial R2 of 14.85% and the Wald Chi-square of 26.8344 provide further evidence that the instruments used are not weak.
19. The Heckman bias correction, λ, is significant at the 10% level suggesting that endogeneity is less of a concern in this specification.
20. Linking changes in classification choice to changes in firm value would provide further support for our results. However, as the IASB notes, presentation choices on the SCF should be consistently applied. Nevertheless, we test this assertion by examining the classification category of interest paid on the SCF for 30 firms selected randomly from our initial sample. Unsurprisingly, their 2009 annual reports indicate that all 30 firms continue to report interest paid in the same section as they did back in 2005. This precludes the identification of a big enough sample to perform such analysis.