The role of social values, accounting values and institutions in determining accounting conservatism.
| Verfasser | Salter, Stephen B. |
| Posten | RESEARCH ARTICLE |
Abstract:
* The extant literature on cross cultural differences in conservatism finds that a mix of institutional and economic variables influences differences in the level of conservatism across countries. Using the modified Gray model Doupnik and Tsakumis (J Acc Lit 3(1): 1-20, 2004) we examine whether societal values are a neglected explanatory factor for differences in levels of accounting conservatism across countries.
* Both unconditional and conditional conservatism are measured. Social values are measured using both a composite value for conservatism Hope et al. (J Acc Public Policy 27(5): 353-357, 2008) and individual values from Hofstede and Hofstede in Cultures and organizations: Software of the mind (2nd ed., New York, McGraw-Hill, 2005). A test sample of 89,481 firm-years from 1989 to 2006, compressed to 341 (or 342) country-year observations from 22 countries provides the data for OLS regressions.
* We find that after controlling for previously documented legal and financial institutional variables, both conditional and unconditional accounting conservatism are greater in countries with more conservative societal and accounting values. The Hofstede value of femininity (the importance of caring for others) is also significant for both measures of conservatism.
* These findings provide a more complete model for further testing of accounting conservatism and other topics using societal values. They also provide support for the modified Gray model using archival data. Finally, while both dimensions of conservatism matter and while they share some common explanators, both have unique additional influencing factors in culture and institutions.
Keywords: Conservatism * Societal values * Accounting values * Institutions
Introduction
Conservatism's influence on accounting practice has been both long and significant. Penndorf (1930), for example, provides evidence that historical records from early fifteenth century trading partnerships show that accounting in medieval Europe was conservative. As Basu (1997, p. 9) notes, conservatism "practices predate income and property taxes, shareholder litigation, and accounting regulation (1), and hence one cannot rely on institutional forces such as taxes to explain the origins of accounting conservatism."
There is universal agreement in the primarily U.S. capital markets (archival) literature that accounting conservatism exists (see Ball et al. 2008). In addition, the general broad concept of conservatism has been bifurcated into two sub concepts: (1) conditional conservatism, and (2) unconditional conservatism. Conditional conservatism is defined by Basu (1997, p.4) as "capturing accountants' tendency to require a higher degree of verification for recognizing good news than bad news in financial statements ... earnings reflect bad news more quickly than good news". Ball and Shivakumar (2005, p. 89) codify the much earlier (2) concept of "unconditional conservatism" as "an accounting bias toward reporting low earnings and book values of stockholders equity."
There is some disagreement as to whether either sub-component differs cross-nationally (Giner and Rees 2001 vs. Beuselinck et al. 2007). The majority of authors accept that cross-national differences in accounting conservatism exist, but there are a number of proposed reasons for these differences which may be classified as: (1) economic explanators, such as the importance of debt or equity financing in a country (Ball et al. 2008; Raonic et al. 2004), and (2) institutional explanators, such as legal system and creditor rights (Giner and Rees 2001; Garcia Lara and Mora 2004; Gassen et al. 2006; Bushman and Piotroski 2006).
It is interesting, given that the decisions to be conservative are not made by automatons but rather by managers or board members, that societal values have not been suggested as an explanator of accounting conservatism. Conservatism choices are essentially management choices and yet global studies ignore the beliefs or value systems of those managers and their society. As Williamson (2000) summarizes, while institutions certainly affect firm decisions and performance, the same institutions are man-made and cannot be separated from cultural values of a society. In the extant literature, societal values have been found to affect the outcomes of accounting phenomena such as level and form of corporate governance (Daniel et al. 2012), level of earnings management (Han et al. 2010), earnings quality (Kanagaretnam et al. 2011), volume and nature of accounting disclosure (Hope 2003), extent of expense recognition (Doupnik and Richter 2004; Doupnik and Riccio 2006), and application of IFRS rules (Ding et al. 2005).
This paper adds the impact of societal values (national culture) as a neglected explanatory factor which can be used to help explain differences in levels of accounting conservatism across countries. As a model of how societal values affect conservatism, we use the modified Gray model (MGM) (Doupnik and Tsakumis 2004, p.41, Fig.5). The MGM describes the factors that influence the manner in which accountants apply financial reporting rules, as an integrative framework. The MGM proposes that accountants' applications of financial reporting rules (APP) (e.g., conservatism) are determined both by institutional consequences (IC) and accounting values (AV). Both IC and AV, in turn, are determined by societal values (SV). After controlling for both legal and financial institutional variables found in prior literature to affect the level of conservatism, we test the model, focusing on the influence of SV on the final level of accounting conservatism.
Our sample consists of 89,481 firm-year observations with available net income, stock price, and societal values data from the years 1989 to 2006, compressed to 341 (or 342) country-year observations (3) from 22 countries. These markets comprise 85 % of global stock market capitalization.
The results indicate that the Gray (1988) composite AV of conservatism is a significant explanator of both conditional and unconditional accounting conservatism. Using the Hofstede SV directly as hypothesized, unconditional conservatism is best explained by the uncertainty avoidance and femininity. The overall level of creditor rights and an English legal system supplement that explanation. Also, as hypothesized, conditional conservatism is best explained by the SVs of collectivism and femininity. The level of corruption and the economic explanations of debt and the breadth of stock market holdings add additional explanatory power to the social values. We believe our findings provide a more complete model for further testing of accounting conservatism and other topics using societal values. They also provide support for the modified Gray model using archival data.
This paper proceeds with an in-depth review of prior research followed by hypotheses, sample and methodology, results, and conclusions.
Prior Research
Examining Conditional and Unconditional Conservatism Using Archival Data
As discussed above, two related yet distinct sub-concepts of accounting conservatism have been studied. Unconditional conservatism is the older of the two sub-concepts and was, until Basu (1997), the dominant concept of accounting conservatism. Unconditional conservatism is defined by Ball and Shivakumar (2005, p. 89) as "an accounting bias toward reporting low book values of stockholder equity (and hence, if clean surplus accounting is being followed, low average net incomes)." Ball et al. (2008, p. 194) note that "the unconditional definition of conservatism has been employed in much prior literature, including the international accounting literature (e.g., Gray 1980)."
Basu (1997) added an additional sub-concept of conditional conservatism. Conditional conservatism is the speed with which companies recognize bad news as opposed to good news. (Basu 1997, p.4) describes it in detail thus:
... capturing accountants' tendency to require a higher degree of verification for recognizing good news than bad news in financial statements. Under my interpretation of conservatism, earnings reflect bad news more quickly than good news. For instance, unrealized losses are typically recognized earlier than unrealized gains. This asymmetry in recognition leads to systematic differences between bad news and good news periods in the timeliness and persistence of earnings. Beginning with Gray (1980), many authors began to examine if cross-national differences in accounting conservatism might exist and, if so, why. Pre-Basu (1997) literature, such as Gray (1980), Nobes (1998), Hailer (2003), and the European Federation of Accountants (1997), focuses on unconditional conservatism and argues that creditor protection is the key reason for its existence. Yet almost all of these studies are based on limited empirical data and focus on the experience of the European Union. Later archival studies, such as Ball (2004) and Ball and Shivakumar (2005), question this creditor protection premise and argue that gains in contracting efficiency by providing better creditor protection can arise only from conditional conservatism (early recognition of losses). Unconditional conservatism, they point out, once it is known to exist, will be compensated for by investors and hence brings no new information to market participants.
A second stream of research uses an institutional perspective to explain national differences in unconditional conservatism. Giner and Rees (2001), Garcia Lara and Mora (2004), and Gassen et al. (2006) focus on institutional systems, primarily national legal systems (such as common law vs. code law). Non-European studies posit and test a variety of institutional and economic explanations for cross-national differences in unconditional conservatism. Finally, Ball et al. (2008), using a sample of 22 national average scores for unconditional conservatism, find no relationship between...
Um weiterzulesen
FORDERN SIE IHR PROBEABO ANCOPYRIGHT GALE, Cengage Learning. All rights reserved.
