3
1. Introduction
There is ongoing debate as to whether International Financial Reporting Standards IFRS adoption necessarily improves accounting quality through a uniform framework for
accounting and reporting across countries, or if due to specific country characteristics, disparity remains an issue Ball, 2006; Ball, Robin, Wu, 2003; Lehman, 2005; Nobes,
2006; Weetman, 2006. In this context, Houqe, Monem, Tareq, and van Zijl 2016 positively argue that mandatory adoption of IFRS improves earnings quality in adopting countries as
inter alia IFRS are designed to reduce managerial discretion to manage or smooth earnings and see Ball, 2006; Shah, Liang, Akbar, 2013. In contrast
, ‘the inherent flexibility in principles-
based standards could provide greater opportunity for firms to manage earnings’ Barth, Landsman, Lang, 2008, p. 468.
One specific area of contention within IFRS in this regard is the treatment of research and development hereafter RD costs. IAS 38 Intangible Assets clearly sets out a number of
conditions under which development costs are to be capitalised. Arguably, by imposing restrictive conditions, the discretion involved in RD capitalisation is reduced Markarian,
Pozza, Prencipe, 2008; Matolcsy Wyatt, 2006. Thus, one would expect that, after the mandatory adoption of IFRS, only development expenditures from those RD projects,
which are highly likely to be successful, are capitalised. However, as the application of these conditions requires managers to exercise their judgment over proprietary and subjective
information, RD capitalisation under IAS 38 still remains open to managerial discretion and hence the reliability of such information can be questioned Cazavan-Jeny, Jeanjean, Joos,
2011; Dinh, Kang, Schultze, 2015. Because of the judgement involved, it follows that the potential discretionary treatment of
development costs under IAS 38 may be associated with two outcomes. First, a genuine
4 signalling effect conveying the future earnings
’ power of the business associated with the level of capitalisation
. In these cases, ‘…investors perceive the capitalisation of RD to be related to successful RD projects
’ Shah et al., 2013, p. 168. Second, consistent with earnings management, a managerial opportunism manifested through a positive impact on
current earnings and possible earnings benchmark beating Dinh et al., 2015, when costs are capitalised and not expensed Ahmed Falk, 2006; Cazavan-Jeny Jeanjean, 2006;
Cazavan-Jeny et al., 2011; Ciftci, 2010; Dinh et al., 2015. In these cases, managers provide a noisy signal with regards to the future benefits associated with the RD costs capitalised.
In this study, we examine the potential influence of corruption as a country characteristic which may be associated with the accounting treatment of RD, specifically development
costs, under IFRS. Corruption is commonly defined as the ‘abuse of entrusted power for
private gain ’ Transparency International, 2009, p. 7. Where corruption is prevalent,
business cheating and scandals are merely one manifestation of a much broader and more insidious acceptance of corruption within society Zuckerman, 2006, cited in Ashforth, Gioia,
Robinson, and Trevino 2008. Effectively, in such settings, corruption is collectively ‘normalized Ashforth Anand, 2003 because it leads to a gradual erosion of moral agency
over time Ashforth et al., 2008; Brief, Buttram, Dukerich, 2001; Fleming Zyglidopoulos, 2008. Thus, as put forward by Ashforth and Anand 2003, p. 8,
‘once a corrupt decision or act produces a positive outcome and is included in organizational
memory, it is likely to be used again in the future ’ and this is not perceived as unacceptable.
Hence, in such environments, based on the rational expectation of a profit-maximising agent Kimbro, 2002, some managers use this non-market mechanism to improve their well-being
through exercising excess discretion Hamir, 1999 to achieve income maximisation , over the well-being of many other stakeholders.
5 Drawing upon this reasoning, prior empirical research has indeed shown that quality of
accounting is susceptible to the level of country-level corruption Doupnik, 2008; Fan, Guan, Li, Yang, 2014; Han, Kang, Salter, Yoo, 2008; Nabar Boonlert-U-Thai, 2007; Picur,
2004. This is evident via a negative relationship between levels of corruption and information transparency DiRienzo et al., 2007, presence of big four auditors and perceived
levels of accounting and audit quality amongst business people Malagueño, Albrecht, Ainge, Stephens, 2010, compliance with mandatory disclosures e.g., about goodwill as examined
by Mazzi, Tsalavoutas, Slack, 2017 and positive relationship between corruption and earnings opacity
1
Picur, 2004; Riahi-Belkaoui, 2004. On reflection of this evidence, our first hypothesis is that managerial discretion and
accounting choice associated with the amount of development costs capitalised, is positively influenced by the level of country corruption. We argue that, in an attempt to project a
positive imag e for their firms’ current and future performance, managers in countries with
high levels of corruption, capitalise larger amounts of development costs, in comparison to those countries with low levels of corruption. As explained earlier, capitalising and not
expensing development costs results in better current-year earnings i.e., earnings management effect and raises expectations for future earnings due to high amounts of assets
capitalised i.e., a noisy signalling effect. On that basis, our second hypothesis is that development costs capitalised have a lower association with cumulative future earnings in the
long-run for firms in countries with higher corruption levels i.e., those associated with over- capitalisation. This is because the capitalised development costs would not deliver as high as
signalled future economic benefits.
1
Earning opacity, like earnings management, can be defined as ‘the alteration or design of firms reported
economic performance by insiders to either ‘mislead some stakeholders’ or to ‘influence contractual outcomes
Healey and Wahlen, 1999 ’’ Riahi-Belkaoui, 2004, p. 75. Hence it is considered a measure of low
accounting quality Picur, 2004.
6 Building on these two hypotheses, we next consider whether levels of firm
internationalisation moderate the influence of corruption on their capitalisation of development costs and also the relationship between capitalised costs and their contribution
to future earnings. As firms become international through, for instance, exporting, cross- listing or owning overseas subsidiaries, the influence of domestic corruption may diminish as
the firm becomes more exposed to international norms Reid, 1983. Indeed, prior literature shows that internationalisation positively influences management attitudes towards
stewardship and accountability Murtha, Lenway, Bagozzi, 1998; Segaro, Larimo, Jones, 2014, compared to more domestically-orientated firms Nadkarni Perez, 2007 and
serves to lessen the effect of domestic corruption Sandholtz Gray, 2003. Thus, our third hypothesis is that the more international a firm is, the lower will be the influence of domestic
corruption on the magnitude of development costs capitalised. From this, our fourth hypothesis follows. More specifically, the influence of domestic corruption on the
relationship between capitalised costs and future earnings will also be weaker for those firms that are more international compared to more domestically-orientated firms where the
influence of local corruption will not be so diminished. To test our hypotheses, we employ a sample of over 3,300 firm-year observations, across
20 countries, which were mandated to adopt IFRS in 2005. Our analyses cover the periods 2006 to 2015 and the findings show the following. First, after controlling for various firm and
other country characteristics, there is indeed a positive relation between country level corruption and the amount of development costs capitalised. Further, we find that the higher
the levels of corruption, the lower the association of the capitalised development costs to future profitability, compared to countries with lower levels of corruption. This finding is
consistent with our hypothesis that in countries with higher corruption levels an earnings
7 management and a noisy signalling effect prevail, in relation to the managerial discretion on
the development costs capitalised. Additionally, in more international companies, the amount of capitalised development costs is less influenced by domestic corruption compared to those
companies that are more domestically orientated. Finally, the contribution of capitalised development costs to future profitability is negatively influenced by the level of domestic
corruption in less international firms. By examining the potential effect of corruption on the levels of capitalised development
costs and their respective contribution to firm level earnings, we contribute to the literature as follows. Firstly, in the only other IFRS study which examines the determinants of RD
capitalisation to our knowledge i.e. Dinh et al., 2015 neither corruption nor internationalisation as country and firm level contextual factors respectively were considered.
Unlike our research which employs a global dataset, their study is restricted to Germany only, and thus their findings are necessarily country specific and do not shed light on the potential
disparity of accounting and managerial choice associated with IAS 38 in a global context. By drawing on corruption as a possible influence on accounting choice, this study provides one
hitherto unexamined aspect of that analysis. In doing so, this study responds to Shah et al. 2013 p. 168 who call for
‘analysis regarding capitalising RD expenditures [which] helps managers to conduct earnings management
’. Secondly, this study addresses the call by Houqe Monem, 2016, p.3 who note that
‘literature linking corruption with accounting is sparse
’. Our analyses provide empirical findings in support of Ball 2006, Nobes 2006, Weetman 2006, Zeff 2007 suggesting that country specific characteristics result in an
uneven application of IFRS worldwide, and arguably, with adverse effects to comparability. The remainder of the paper is organised as follows. Section 2, discusses the relevant
literature and the hypotheses development. Section 3 describes the sample selection process
8 and the methods employed. Section 4 presents and discusses the empirical findings. Section
5 illustrates additional tests and Section 6 concludes the paper by outlining the contributions arising from this research. It also discusses limitations and avenues for further research.
2. Background, literature review and hypotheses development
Kategori