Do UK firms manage earnings to meet dividend thresholds?

Abstract

This paper examines earnings management by dividend-paying firms in cases where pre-managed earnings would fall below the expected dividend, and by non-dividend paying firms aiming to avoid reporting losses. We find that within the UK market the likelihood of upward earnings management is significantly greater in the former case than the latter, though both are drivers for earnings management. Large firms are less likely to upwardly manage earnings to reach dividend thresholds, consistent with prior UK evidence on the ability of the largest firms to avoid restrictive debt covenants. We also find that earnings management is more clearly observable through examining working capital discretionary accruals than through examining total discretionary accruals.

Keywords:

• accruals
• debt covenants
• dividend payments
• dividend thresholds
• earnings management
• loss-avoidance

Acknowledgements

The authors would like to acknowledge the significant contribution of the two anonymous referees and the journal editor to the development of this paper. We would also like to thank all staff members who have contributed comments and suggestions during the development of Abdallah's doctoral research, especially John Capstaff (Strathclyde) and Tony Appleyard (Newcastle).

Notes

See Oswald and Young (2004) for a critique of the dataset used by Rau and Vermaelen (2002).

DDN use both models within their study but present most of their results for the cross-sectional Jones model. We prefer to focus on Dechow et al. (1995) because it takes account of changes in receivables, which are ignored in the original specification of the Jones model. However, results are virtually identical as indicated by our sensitivity testing (see Table 6).

DDN investigate a number of alternative proxies for expected dividend but report (p. 15) that their results are materially unchanged.

These data are presently no longer available on The Guardian's website but can be obtained from the authors on application.

Coefficients for our control variables are often insignificant or inconsistent with regard to signs across different models. This mirrors the results reported in DDN's Tables 3 and 4.

In section 4.3 we show that although the slope of DEFICIT is much smaller for non-payers, it is positive and statistically different from zero at the 0.05 probability level.

This finding is not surprising since Benito and Young (2001, p. 21) note that non-paying UK firms tend to be smaller and are more likely to exhibit poor current earnings numbers.

The dummy variables LARGE and LOWGEAR are used on the slopes of the DEFICIT variable to examine whether being a large firm or a low-geared firm influence the relationship between earnings management and deficits. However, these dummies are not included as intercept dummies since our set of control variables already contains measures for firm size (SIZE) and gearing (GEARING).

The dummy variable PAYER equals (1 – NONPAYER).