The effect of discretionary accruals on annual report readability: An empirical study on BIST

Discretionary accruals inherently have flexible characteristic. This flexibility enables discretionary accruals to be used for earnings manipulation by firm managements. However, earnings management is such behaviour that managers would prefer to obfuscate. In fact, managements may obfuscate bad news by narrating them more complex style while they put forward to disclose good news (Moffitt and Burns, 2009, p. 400). Thus, it can be expected that earnings management may affect readability of financial disclosures. Annual reports are among the most important documents where financial information is disclosed. There are several studies that examine the effect of discretionary accruals on annual report readability. The results of studies by Ajina et al. (2016), Lo et al. (2017) and Cheng et al. (2018) suggest that firms that have managed their earnings have less readable annual reports. However, these studies are conducted in France, USA and China respectively and any study that examines such relationship in Turkey could not be reached. The aim of this study is to examine the effect of discretionary accruals on annual report readability in Turkey. In this study, sample used in the analysis consists of 54 BIST firms' firm-year observations between 2012-2021 periods. Discretionary accruals are estimated both Modified Jones (1995) model and Jones (1991) model. Fog (1952) index and FleschKincaid (1975) index are used for readability measurement. In consequence of panel regression analysis, it is found a significant and positive relationship between discretionary accruals and annual report readability. Consistent with existing literature, annual report readability can be considered as an indicator of earnings management.