financial reporting / international
IFRS: Beyond the Standards
By George T. Tsakumis, David R. Campbell Sr. and Timothy S. Doupnik
Since the European Union’s 2002 regulation mandating IFRS for EU public companies and the execution of the Norwalk Agreement by FASB and the International Accounting Standards Board (IASB), momentum has been building for global standards convergence. Currently, more than 100 countries have adopted IFRS, and a number of other economically important countries, including Japan and the United States, have programs in place to converge their national standards with IFRS. IASB Chairman Sir David Tweedie has said that by December 2011, U.S. GAAP and IFRS “should be pretty much the same.”
At that point, about 150 countries would be using very similar accounting standards, though some countries have adopted versions of IFRS that vary from IFRS as published by the IASB. In 2007 the SEC extended the question beyond mere convergence by accepting the English language version of IFRS by foreign issuers without reconciliation. And in November the SEC released a proposed road map that could require a phased adoption of IFRS by U.S. issuers beginning in 2014, dependent in part on whether seven milestones are achieved. In the road map, which has a comment period ending Feb. 19, it is noted that, “The Commission has long expressed its support for a single set of high-quality global accounting standards as an important means of enhancing comparability.”
This article points out that even among countries that have adopted the same version of IFRS, recent accounting research suggests that two factors—national culture and language translation—could undermine the rigorous interpretation and application of IFRS and lead to a lack of comparability across countries. The objective of this article is to highlight two significant hurdles that impede the consistent interpretation and application of converged standards: the influence of national culture on the interpretation of standards and the difficulty of translating standards into other languages.
Research suggests that cultural differences cause accountants in different countries to interpret and apply accounting standards differently. This research reveals that two accounting values directly influenced by national culture are conservatism and secrecy, which affect the measurement and disclosure of financial information in financial reports and have the greatest potential to affect cross-border financial statement comparability. The underlying framework helping to explain these findings is based upon one of the largest crosscultural surveys ever conducted. Social psychology researcher Geert Hofstede (Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations, 2nd Edition) collected data on cultural values from approximately 116,000 employees of a multinational company located in 50 countries and three regions around the world. He identified four cultural dimensions that reflected core values and helped explain general similarities and differences in cultures. These dimensions were:
■ Uncertainty avoidance—how comfortable individuals in a society feel with uncertainty and ambiguity;
■ Individualism—a society’s preference for a loosely knit social fabric or a more interdependent, tightly knit social fabric;
■ Achievement orientation—how much values such as performance and visible achievement are emphasized; and
■ Power distance—how much hierarchy and unequal power distribution are accepted in a culture.
Hofstede’s cultural framework has been used extensively in management and other disciplines to examine the influence of national culture on organizational and individual performance. This framework can be used in an accounting context to explain the SEC’s concern that “proper application encompasses not only faithful adherence to the requirements of the standards, but also understandable standards such that across the spectrum of issuers those requirements are consistently understood and applied” (Concepts Release 33-8831, August 2007, page 30).
From an accounting perspective, high conservatism implies a tendency to defer the recognition of assets and items that increase net income. Within Hofstede’s framework, higher levels of conservatism are most closely linked with countries that have higher uncertainty avoidance and lower individualism and achievement orientation. High secrecy implies a tendency to restrict the disclosure of relevant information to outside parties. Higher levels of secrecy within a culture are associated with higher uncertainty avoidance and power distance and with lower individualism and achievement orientation. In summary, research reveals that the cultural values existing in a country influence a country’s accounting values (accountants’ levels of conservatism and secrecy), which influence how financial reporting standards are applied:
Cultural Values g Accounting Values g Application of Financial Reporting Standards
National culture is most likely to influence the application of financial reporting standards where judgment is required. This is of concern since most of IFRS is principles-based. This requires substantial judgment on the part of the accountant. Recent research has found that national culture influences both the interpretation and application of accounting standards across countries.
For example, a study by Timothy S. Doupnik, a co-author of this article, and Martin Richter (“Interpretation of Uncertainty Expressions: A Cross-National Study,” Accounting, Organizations and Society, Jan. 2003) found that German accountants exhibit more of a conservative bias than their U.S. counterparts in their interpretation of the word “probable” in establishing the threshold for the recognition of a construction contract loss. In this context, “probable” is used as a threshold for the recognition of an item that decreases income. German accountants assigned a lower numeric threshold (66% likelihood of occurrence) to the term “probable” than their U.S. counterparts (74% likelihood of occurrence), exhibiting greater levels of conservatism in recognizing a construction contract loss (IAS 11).
Another study (“The Influence of Culture on Accountants’ Application of Financial Reporting Rules,” Abacus, March 2007) by George T. Tsakumis (lead author of this article) found that when presented with similar economic facts and financial reporting guidelines, accountants in the United States and Greece made different contingency recognition and disclosure decisions in their application of IAS 37, Provisions, Contingent Liabilities and Contingent Assets. U.S. accountants were more conservative than Greek accountants in their recognition of a lawsuit in the financial statements as either a contingent asset or a contingent liability. Only 33% of U.S. accountants responded that they would be likely to recognize the lawsuit as a contingent asset, while 65% of Greek accountants indicated that they would recognize the lawsuit as a contingent asset. In the same situation, 72% of U.S. accountants indicated that they would recognize the lawsuit as a liability, while only 59% of the Greek accountants elected to do so. From a disclosure perspective, Greek accountants were much more reluctant to disclose the existence of a lawsuit to outside parties than U.S. accountants. While 84% of U.S. accountants indicated that they would disclose the lawsuit in the notes to the financial statements, only 56% of the Greek accountants indicated a disclosure preference.
Other research (“The Influence of Conservatism and Secrecy on the Interpretation of Verbal Probability Expressions in the Anglo and Latin Cultural Areas,” by Doupnik and Edson Luiz Riccio, The International Journal of Accounting, Vol. 41, Issue 3, 2006) also has found Brazilian accountants to be much less likely than U.S. accountants to disclose contingencies in the notes to the financial statements. In both studies, Greek and Brazilian accountants exhibited more secrecy than U.S. accountants. Another study (“The Impact of National Influence on Accounting Estimates: Implications for International Accounting Standard- Setters,” by Joseph J. Schultz and Thomas J. Lopez, The International Journal of Accounting, Vol. 36, Issue 3, 2001) found that French and German accountants recommended recording warranty estimates higher (more cautious or conservative) than their American counterparts.
These findings suggest that wherever professional judgment is required, national culture plays a significant role in how accountants interpret and apply IFRS. Culture is a pervasive environmental factor that can lead to inconsistent interpretation and application of converged financial reporting standards. This is troublesome because different judgments could lead to significant differences in financial statements. These differences could severely impact the comparability of financial statements across countries.
Translation of IFRS into various languages poses another threat to comparability. The official working language of the IASB, and the language in which IFRS is published, is English. For non- English speaking accountants to have access to IFRS, they are translated into other languages. The International Accounting Standards Committee Foundation (IASCF) created an official translation process in 1997, and IFRS was first officially translated into German. By the end of 2006, IFRS had been translated into nearly 40 languages. Translations primarily have been into European languages, but Chinese, Japanese and Arabic translations also have been made.
The IASCF coordinates the translation of IFRS. Its translation process is well-organized and quite rigorous. However, despite the care taken by translators and the oversight provided by translation review committees, translation problems exist. In some cases, words and phrases used in English-language accounting standards cannot be translated into other languages without some distortion of meaning. Several research studies have shown this to be true.
In one study (“Interlinguistic Comparison of International Accounting Standards: The Case of Uncertainty Expressions,” by Ronald A. Davidson and Heidi H. Chrisman, The International Journal of Accounting, Vol. 28, Issue 1, 1993), Canadian researchers examined accounting students’ interpretations of probability expressions (probable, not likely, and reasonable assurance) used in Canadian accounting standards to establish recognition and disclosure thresholds. They found Anglophone students’ interpretations of some English-language expressions to be significantly different from the interpretations made by Francophone students of the French-language translation. In the study, cited earlier by Doupnik and Richter, German accountants fluent in English assigned values to both the English original and the German translation of probability expressions used in IFRS. For several expressions, the original and translation were interpreted differently.
A particularly troublesome term to translate is remote, which is used to establish a threshold for the disclosure of contingent liabilities in IAS 37, Provisions, Contingent Liabilities and Contingent Assets, and in IAS 31, Interests in Joint Ventures. IAS 37 (para. 28) indicates that a contingent liability is disclosed “unless the possibility of an outflow of resources embodying economic benefits is remote,” and IAS 31 (para. 45) requires separate disclosure of specific types of contingent liabilities “unless the probability of loss is remote.” Given the similar purpose and context in which remote is used in these two standards, it would appear that the IASC intended the term to convey the same meaning in both cases.
Webster’s Online Dictionary (www.websters-online-dictionary.com) provides five definitions for remote, four of which indicate that it is an expression of distance in space or time (remote location, remote past). The fifth definition is “very unlikely.” The IASC clearly was using this last meaning when selecting remote as the threshold for disclosing contingent liabilities. It is at the end of a continuum that begins with probable, moves to possible, and ends with remote. The fact that remote has several different meanings in English suggests that translation might create some problem.
Translating remote into Spanish appears to cause little or no difficulty. The cognate word remoto (or remota) is used in both IAS 31 and IAS 37. The online translation dictionary available at www.freedict.com indicates that remote can be translated into Spanish as vasto, remoto or lejano. On the other hand, remoto can only be translated into English as remote. Remoto also is used in the Italian and Portuguese translations of IFRS.
Translating remote into French is not quite as easy. The word faible is used in the French translations of both IAS 31 and IAS 37. The Freedict online dictionary indicates that remote can be translated into French as isolé, distant or vaste but does not list faible as an option. Similarly, the French word faible is translated into English as faint, light or weak but not remote. The French translator of IFRS selected a French word that is intended to convey the same meaning as remote, in the sense of very unlikely, even though it is not a direct translation. However, there must have been some uncertainty on the part of the translator as to whether faible adequately captures the essence of remote, because the adjective trés (very) is added to form the expression trés faible in IAS 31. Thus, remote is literally translated as weak (faible) in IAS 37 and very weak (trés faible) in IAS 31. Preparers of financial statements using the French translation of IFRS are likely to interpret IAS 37 as establishing a stronger prescription against nondisclosure than IAS 31. Given that the IASC used remote in both IAS 31 and IAS 37, there is no evidence that the standard writer intended for the nondisclosure threshold to be higher in one standard than in the other; yet the French version gives that impression.
Translation of remote into German is even more problematic, as evidenced by the fact that the translation is inconsistent between the two international standards. In IAS 31, remote is translated as unwahrscheinlich (improbable), and in IAS 37, it is translated as äußerst gering (extremely remote). In a strictly statistical sense, improbable (unwahrscheinlich) implies a probability of anything less than 50%. Extremely remote (äußerst gering), on the other hand, suggests a much smaller probability. As a result, the German translation of IAS 31 establishes a much more stringent threshold for nondisclosure of contingent liabilities than the German translation of IAS 37; this difference does not appear to have been the intent of the IASC. Moreover, neither translation appears to be the equivalent of what is intended in the original English-language standard. It is interesting to note that the German translator felt the need to add the modifier extremely to remote in IAS 37, whereas the French translator added the modifier very in IAS 31. Thus, the French and German translations are inconsistent in the manner in which modifications have been made. Remote is but one expression that appears to be difficult to translate into German. Research by Doupnik and Richter identified other inconsistencies in the way that English probability expressions have been translated into German. Expected has been translated either as erwartet or voraussichtlich; likely as either voraussichtlich or wahrscheinlich, and probable as either wahrscheinlich or hinreichend wahrscheinlich (see Doupnik and Richter).
In that research, a group of German accountants fluent in English were asked to interpret both the English-language original and German translation of approximately 20 probability expressions used in IFRS to establish recognition or disclosure thresholds. Significant differences in the interpretation of the English expression and its German translation were found in eight cases. For example, the German accountants associated a probability of 89.5%, on average, with the term assurance, whereas its German translation (Gewißheit) was interpreted, on average, to mean 95.8%. At the other end of the probability continuum, the German accountants associated an average probability of 22.5% with the term remote but a probability of only 9.2% with its German translation (äußerst gering). Given a particular set of facts and circumstances, a German accountant applying the English-language version of an international standard might make different recognition or disclosure decisions than if he or she were using the German translation.
These are just a few examples of the translation challenge. Undoubtedly, other words and concepts used in IFRS are difficult to translate from English into other languages, despite the rigorous quality of the IASCF’s efforts.
RECOMMENDATIONS AND CONCLUSIONS
Given the potential problems resulting from culture and translation, how do we go about ensuring the consistent interpretation and application of common standards across countries? Hofstede suggests that a society’s value system changes very slowly, so waiting for national cultures to change and for accountants worldwide to have similar accounting values is not a feasible solution. In our judgment, it will be important for multinational corporations and global audit firms to strengthen their cultural awareness training. This could benefit multinational corporations and their auditors in two key ways: It will make them aware of potential biases held by their international staff and by colleagues in their international offices, and it will help professionals recognize their own country’s cultural accounting tendencies and better understand how these values affect their own interpretations and judgments.
In terms of addressing the translation barriers highlighted in this article, we believe the IASCF needs to enhance the translation process to ensure that inconsistencies are avoided. In addition, the IASB could reduce the potential for distorted translations by exercising greater care in avoiding the use of ambiguous English words and phrases that require translators to make assumptions regarding the intended meaning. A comprehensive review of existing standards could replace ambiguous language.
For example, small probability could replace remote in IAS 31 and IAS 37 without changing the basic intent of the standard. Replacing remote with a specific level of probability, for example, 10% or less, would eliminate any linguistic ambiguity entirely. However, doing so would replace a vague principle with a clear-cut rule, and this is inconsistent with the philosophy behind IFRS; therefore, it is unlikely to be acceptable to the IASB’s constituents.
Finally, the IASCF may wish to consider using the established methodology of back-translation. For example, if a standard is to be translated from English into Italian, a native speaker with an accounting background first translates the standard into Italian. Then a second native speaker translates the standard back into English. This process provides translators a chance to identify and resolve any discrepancies between the original and translated versions.
Finally, the education and preparation of future professionals will play a key role in overcoming the effect that national culture can have on the comparable application of IFRS across countries. Exposure to converged standards and their proper application must be emphasized early in the education of accountants worldwide. In addition, standard setters and employers must play an active role in educating current and future accountants about the challenges (and possible solutions) that lie ahead. This can be accomplished through various initiatives including campaigns to increase the profession’s awareness of specific interpretation and application challenges and cross-border personnel exchanges between accounting firms. v
n Research suggests that cultural differences cause accountants in different countries to interpret and apply accounting standards differently. This research reveals that two accounting values directly influenced by national culture are conservatism and secrecy, which affect the measurement and disclosure of financial information in financial reports and have the greatest potential to affect cross-border financial statement comparability.
n Translation of IFRS into various languages poses another threat to comparability. The official working language of the IASB, and the language in which IFRS is published, is English. For non-English speaking accountants to have access to IFRS, it is translated into other languages.
n It will be important for multinational corporations and global audit firms to strengthen cultural awareness training. This could benefit multinational corporations and their auditors by making them aware of potential biases held by their international staff and by colleagues in their international offices, and by helping professionals recognize their own country’s cultural accounting tendencies and better understand how these values affect their own interpretations and judgments.
n The IASCF must enhance the translation process to ensure that inconsistencies are avoided. In addition, the IASB could reduce the potential for distorted translation by exercising greater care in avoidinG the use of ambiguous English words and phrases that require translators to make assumptions about the intended meaning.
George T. Tsakumis, Ph.D., is an assistant professor of accounting at Drexel University in Philadelphia. David R. Campbell Sr., CPA, Ph.D., chairs the Department of Accounting at Drexel University. Timothy S. Doupnik, Ph.D., is a professor and chair of the School of Accounting at the University of South Carolina. Their e-mail addresses are firstname.lastname@example.org, email@example.com and firstname.lastname@example.org, respectively.
International Versus U.S. Accounting: What in the World is the Difference?, a CPA self-study course (#731665)
For more information or to make a purchase, go to www.cpa2biz.com or call the Institute at 888-777-7077.
Web siteIFRS Resources, www.ifrs.com
SEC road mapRoadmap for the Potential Use of Financial Statements Prepared in Accordance with International Financial Reporting Standards by U.S. Issuers, www.sec.gov/spotlight/ifrsroadmap.htm