Are IFRS and U.S. GAAP Converging

Journal of International Accounting, Auditing and Taxation 15 (2006) 48–71

Are IFRS and U.S. GAAP converging? Some evidence from People’s Republic of China companies listed on the New York Stock Exchange  John L. Haverty ∗ St. Joseph’s University, 5600 City Ave., Philadelphia, PA 19131, United States

Abstract

This research investigates the comparability and convergence of two sets of accounting standards from 1996 to 2002: United States’ Generally Accepted Accounting Principles (U.S. GAAP) and International Financial Reporting Standards (IFRS). The investigation involves a sample of companies from the People’s Republic of China (PRC) that are listed on the New York Stock Exchange (NYSE). PRC companies traded on the NYSE NYSE genera generally lly prepar preparee IFRS IFRS financi financial al statem statement entss and provide provide a limite limited d reconc reconcili iliati ation on to U.S. U.S. GAAP GAAP, creating a unique quasi-experimental opportunity to examine differences between two sets of accounting numbers produced by two different sets of accounting standards while holding the company constant. Compara parabi bilit lity y is meas measur ured ed by using using Gray Gray’’s inde index x of comp compar arab abil ilit ity y, and and a set set of meas measur ures es are are intr introdu oduce ced d to capt captur uree several dimensions of convergence over time in reported net income, net assets, return on net assets, and earnings per share. The evidence shows lack of comparability, caused largely by the revaluations of property, plant and equipment permitted under IFRS, but not permitted under U.S. GAAP. There is, however, substantial evidence of convergence over time. © 2006 Elsevier Inc. All rights reserved. Keywords: Convergence; People’s Republic of China; International Financial Reporting Standards; IFRS; Accounting

harmonization

1. Introduct Introduction ion

This research investigates the comparability of two sets of accounting standards from 1996 to 2002: United States’ Generally Accepted Accounting Principles (U.S. GAAP) and International 

∗

Data availability: The data used in this study can be obtained from public sources. Tel.: +1 610 660 1656(O)/+1 610 544 9120(R); fax: +1 610 660 1126. E-mail address: [email protected].

1061-9518/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.intaccaudtax.2006.01.004

J.L. Haverty / Journal of International Accounting, Auditing and Taxation 15 (2006) 48–71

49

Financial Financial Reporting Reporting Standards Standards1 (IFRS). IFRS). The resear research ch also also exami examines nes change changess in compar comparabi abilit lity y over over time, i.e., convergence. The stated goal of the International Accounting Standards Board (IASB) and its predecessor organization, the International Accounting Standards Committee (IASC) has been to bring about convergence of National Accounting Standards and International Accounting Standa Standards rds.. In additi addition, on, the United United States States account accounting ing standa standardrd-set settin ting g body body, the Financ Financial ial Account Account-ing Standards Board (FASB) has in recent years held the position that it will formulate its policy with due regard to international considerations. This position has been formalized by the signing of the Norwalk Agreement Agreement (Financial (Financial Accounting Standards Board, 2002) 2002 ) in which both the FASB and the IASB pledged their best efforts to make their existing financial reporting standards fully compatible as soon as is practicable. Given this environment of cooperation, convergence of U.S. GAAP with IFRS is expected and should be observable in the financial statements reported by firms over time. Observation of convergence is a complex task, however, and this research presents a methodology to observe convergence using financial statements prepared by firms reporting under U.S. GAAP and IFRS. The The requ requir irem emen entt of the the Unite United d State Statess Secu Securi riti ties es and and Exch Exchan ange ge Co Comm mmis issi sion on (SEC (SEC)) that that forei foreign gn firms listed on a U.S. Stock Exchange such as the New York Stock Exchange (NYSE) provide a limited reconciliation of their net income and net assets financial statements prepared under IFRS to U.S. GAAP presents an opportunity to test for convergence. This multiple reporting creates a quasi-experimental situation in which the company is held constant, and accounting numbers (e.g., net income and net assets) are calculated according to IFRS and U.S. GAAP for a given year. This investigation involves a sample of People’s Republic of China (PRC) firms listed on the New York Stock Exchange. PRC companies were chosen for this study not only because of the growing worldwide economic significance of the PRC, but also because the PRC provides a particularl particularly y unique testing ground to explore explore comparability comparability and conver convergence gence issues. The PRC and the United States represent vastly different economic systems with significantly different accounting systems. With the ongoing economic restructuring within the PRC, however, as well as the gradual opening of the PRC to the international economic community, differences between the PRC accounting system and the U.S. accounting system are expected to become smaller over time time.. Since Since PRC comp compan anie iess seek seekin ing g fore foreig ign n equi equity ty capi capita tall are are requ requir ired ed by PRC PRC law law to repo report rt unde underr IFRS, and those seeking U.S. equity capital are required by U.S. law to reconcile net income and net net asse assets ts to U.S. U.S. GAAP GAAP, we have have a situ situat atio ion n in whic which h we can can comp compar aree IFRSIFRS-pro produ duce ced d acco account untin ing g measurements to U.S. GAAP-produced accounting measurements for the same company in the same year. The fact that the PRC and the U.S. represent vastly different economic and accounting systems leads us to believe that we should expect an initial lack of comparability between IFRS financial statements produced by PRC firms, and U.S. GAAP financial statements produced by those same firms. Over time, however, we should observe convergence. Given this scenario, we have a unique situation in which to test for comparability and convergence between IFRS and U.S. GAAP. GAAP. This research first examines if the accounting accounting numbers produced produced by IFRS and U.S. GAAP are comparable comparable for PRC compani companies. es. If IFRS IFRS and U.S. U.S. GAAP are comparable, comparable, then a given given acco accoun unti ting ng numb number er (suc (such h as net net inco income me)) calc calcul ulat ated ed unde underr IFRS IFRS shou should ld be fair fairly ly close close to that that same same acco accoun unti ting ng numbe numberr (suc (such h as net net inco income me)) calc calcul ulat ated ed unde underr U.S. U.S. GAAP GAAP. The The rese resear arch ch then then expl explor ores es 1

The International Accounting Standards Committee (IASC) was restructured into the International Accounting Standards Board (IASB) as of April 2001. Standards issued by the IASC were known as International Accounting Standards (IAS). As of April 2001 they are now called International Financial Reporting Standards (IFRS). The current terminology is used in this study.

50


the reasons for any observed lack of comparability. Finally, the research examines convergence. If IFRS and U.S. GAAP are converging, then the differences between a given accounting number (such as net income) calculated according to IFRS and the same accounting number calculated according to U.S. GAAP should be shrinking over time. In order to test convergence, this research introduces a series of metrics that measure convergence over time. This research provides evidence of non-comparability. The majority (10 of 11) of the PRC companies in this study report a net income under IFRS that is materially different (at a 5% materiality threshold) from net income under U.S. GAAP. The most significant cause of the lack of comparability is that revaluation of property, plant and equipment is permitted as an option under IFRS, but U.S. GAAP requires reporting property, plant and equipment on an historical cost basis. The non-comparability is tempered, however, by evidence of convergence over time for the PRC companies that did not indicate comparability. The next section discusses International Financial Reporting Standards and provides some background on accounting harmonization. Then, some background on economic restructuring in the PRC and the PRC accounting system is provided. Specific research questions and the methodology of the study are developed. Results are presented, and the concluding section discusses the implications of the results for investors and researchers. 2. Accounting harmonization, convergence, and International Financial Reporting Standards

One challenging aspect of international business is the fact that no two countries have exactly the same accounting standards or procedures (Gernon & Meek, 2004). Diversity in accounting standards is caused by cultural, economic, historic, legal, and political reasons since the accounting system in a particular nation reflects the unique aspects of that nation. Diversity in standards unfortunately is a considerable barrier to the cross-border flow of capital. In order to properly evaluate an investment in another country, an investor must translate financial statements prepared under a foreign set of accounting standards into financial statements in accord with the accounting standards of the investor’s home country. Despite the forces favoring international diversity of accounting systems, there are also a number of strong forces favoring harmonization of the various national accounting systems. Some of these factors include the explosive growth in cross-border financing, improvements in communication technology, the formation of cross-national economic blocs such as the European Union and NAFTA, and efforts by the United Nations. As a consequence, attempts have been made to encourage accounting harmonization at the international level as well as the local level of various nations. 2.1. The accounting harmonization movement

Internationally, a milestone in the march toward accounting harmonization was the formation of the International Accounting Standards Commission (IASC) in 1973. The IASC was as an independent, private sector body whose objective was to facilitate the cross-border flow of capital by making financial statements more comparable even though they were prepared under various sets of National Accounting Standards. Membership in the IASC included professional accounting bodies of various nations. Immediately after its founding in 1973, the IASC chose the politically expedient strategy of allowing a wide variation in permitted accounting methods. Disclosure of the accounting method was emphasized instead of forcing compliance with a par-


51

ticular model. Major international players such as the United States, Japan and various European nations were not forced to change their domestic standards to be in compliance with an emerging set of International Accounting Standards created by the IASC. This strategy insured at least passive international support for IASC efforts from the world’s major financial powers. As the years progressed, however, the IASC gradually reduced the number of permitted alternative accounting methods. In April 2001, the IASC was restructured and renamed the International Accounting Standards Board (IASB, 2002). Its objectives include: (1) developing a set of high quality, understandable, and enforceable global accounting standards; (2) promoting the use and rigorous application of these standards; and (3) bringing about convergence of National Accounting Standards and International Accounting Standards (International Accounting Standards Board, 2002). Since 1973, the IASB and its predecessor organization, the IASC have issued 41 accounting standards (6 have subsequently been superseded). These standards were formerly known as International Accounting Standards (IAS) and are now called International Financial Reporting Standards. The IASB has no authority to enforce compliance with these standards, but many National Accounting Standard-setting bodies have permitted or encouraged use of IFRS as alternatives or supplements to their own National Accounting Standards. Belgium, France, and Italy, for example, passed laws in 1998 allowing IFRS to be used for domestic financial reporting. Recently, the International Organization of Securities Commissions (IOSCO) has recommended that its members allow multinational issuers to use the 30 IFRS current at that time in their cross-border offerings and listings. In addition, the European Union announced plans to require IFRS for all European Union listed companies no later than 2005. Harmonization efforts also occur at the local level of various nations. The very existence of IFRS and its persuasive promotion on the part of the IASB have influenced the development of various National Accounting Standards. Accounting bodies of various nations, even if they have not adopted IFRS, sometimes model their own national standards or at least modify their own standards with international standards in mind. The United States, for example, has not adopted IFRS for domestic reporting but has stated that it will formulate accounting policy with due regard to international considerations. Despite extensive cooperation between the IASB and the United States Financial Accounting Standards Board, many differences exist between IFRS and U.S. GAAP. Deloitte Touche Tohmatsu (2002) lists 78 specific differences between IFRS and U.S. GAAP. The significance of the differences between IFRS and U.S. GAAP will vary from firm to firm depending on the specific economic circumstances facing each firm. The United States permits foreign companies raising capital in the United States to use either their own domestic accounting standards or IFRS but requires a note reconciling net income and net assets to U.S. GAAP. There is considerable strength to the argument that IFRS and U.S. GAAP are becoming closer over time. In recent years, U.S. accounting standards have supposedly been created and revised with international considerations in mind. IFRS, due to the political realities facing the IASB, must necessarily be created and revised with U.S. GAAP in mind due to the U.S. economy’s prominent position in the world economy. These trends have been formalized with the signing of the Norwalk Agreement (Financial Accounting Standards Board, 2002), in which the U.S. and the IASB pledged to use their best efforts to make their existing financial reporting fully compatible as soon as is practical. In spite of these forces favoring harmonization, there are still major differences between IFRS and U.S. GAAP, particularly in the range of accounting practices permitted by IFRS. Recent research (Street & Gray, 1999) concluded that there were major differences still to be resolved, notably in the determination of net profit/loss for the period,

52


research and development, changes in foreign exchange rates, and business combinations, but noted that these differences were not insurmountable. Evidence of non-comparability ( Harris & Muller, 1999) has been presented using the reconciliation of net income and net assets per IFRS to net income and net assets per U.S. GAAP required by the SEC for foreign registrants using IFRS. Evidence of convergence (Street, Nichols, & Gray, 2000) between IFRS and U.S. GAAP was presented for a sample of non-U.S. companies reporting under IFRS and providing a limited reconciliation from IFRS to U.S. GAAP. 2.2. The concept of accounting harmonization

Accounting harmonization is defined as “. . . a process of increasing the comparability of accounting practices by setting limits on how much they can vary. Harmonized standards are free of logical conflicts, and should improve the comparability of financial information from different countries” (Choi, Frost, & Meek, 2001, p. 291). Accounting harmonization is a process leading to the ultimate goal of increasing comparability of financial information across national borders. Accounting harmonization is a multi-faceted concept, containing at least three components of accounting harmonization (Choi et al., 2001): (1) harmonization of accounting standards, which deals with measurement and disclosure; (2) harmonization of disclosures made by publicly traded companies in connection with securities offerings and stock exchange listings; and (3) harmonization of auditing standards. Even this is a limited list since accounting harmonization is influenced by a myriad of additional factors. For example, if a nation does not have an auditing and standards enforcement infrastructure, then there is no guarantee that accounting numbers (such as net income or net assets) reported under a particular set of standards represent an accurate application of those standards. Countries with very similar accounting standards may not be comparable for reasons well beyond the similarity of the respective accounting standards. It is possible to have a situation of only apparent accounting harmonization if the standards are relatively similar, but if either the culture of compliance or the system of enforcement is inadequate. Development of an operational measure of accounting harmonization is a complex issue. Two sets of accounting standards may be “in harmony” but may apply different sets of rules to the same situation since each national set of accounting standards allows some degree of choice of treatment. In addition, some standards are silent on the treatment of a particular issue. Scholars (Canibano & Mora, 2000) note there are two forms of harmonization: de jure and de facto harmonization. De jure, or formal harmonization, refers to the harmonization of regulations. A recent example of an attempt to measure de jure accounting harmonization is Larson and Kenny (1999), who found some measurable progress toward accounting harmonization in the 1990s but concluded that harmonization is not yet a reality. They noted that compliance with IFRS was not the same as harmonization since IFRS allowed many alternative accounting treatments. De facto, or informal harmonization, refers to the actual accounting practices of corporations. Canibano and Mora (2000) studied de facto harmonization in the European Union and found evidence of de facto accounting harmonization among global players within the European Union. Street et al. (2000) f ound that the differences between IFRS and U.S. GAAP are narrowing for a sample of non-U.S. companies that reconcile IFRS reported net income to net income per U.S. GAAP. De facto and de jure harmonization can each be broken down into two components, the degree of disclosure and measurement criteria. This results in four forms of accounting harmonization: (1) de jure disclosure harmonization, which concerns regulations governing what is disclosed; (2) de jure measurement harmonization, which concerns regulations governing how reported quantities are measured; (3) de facto disclosure harmonization, which concerns what corporations actually


53

disclose; and (4) de facto measurement harmonization, which concerns how corporations actually measure quantities. 2.3. The concept of accounting convergence

Recently, the term “accounting harmonization” has at times been replaced by the term “accounting convergence”. This term has been defined as: The process pursuedby the International Accounting Standards Board (IASB) of eliminating the present differences between National Accounting Standards and the avoidance of future differences to achieve international accounting harmonization ( Hussey & Ong, 2005, p. 229). In this sense, accounting convergence is a process that occurs at the standard-setting level intended to achieve a state of de jure accounting harmonization. It is possible to measure de jure harmonization by focusing on the range of choices provided by various accounting standards. This research, however, measures accounting convergence in a de facto sense by examining the financial reports of companies in a multiple reporting scenario over time. Measuring de facto convergence enables us to measure the actual financial impact of differences in the accounting standards and enables us to examine the actual choices made by firms. If we consider measurement as a system of assigning numbers to qualities of an object, we can consider a set of accounting standards as rules for measuring aspects of the financial condition of an organization. If the two measuring systems are equivalent, they should produce the same set of accounting numbers. For example, if U.S. standards and IFRS were equivalent, then net income reported under U.S. accounting standards and net income reported under IFRS should be the same. If the two sets of accounting numbers are different under each set of accounting standards, then the measuring instruments are not equivalent. If the two sets of accounting standards are becoming harmonized over time, then it is likely that the numbers purporting to measure the same quality (e.g., net income) of an organization would be moving closer over time. This is the meaning of the term convergence in this research, a strictly mathematical measure of increasing closeness over time of two numbers purporting to measure the same thing. Since the purpose of accounting harmonization is to increase comparability of financial reports produced by different nations, mathematical convergence2 of reported accounting numbers would be a necessary manifestation of increasing de facto accounting harmonization. Financial information produced under different accounting, disclosure, and/or auditing systems is comparable if it is “ . . . similar in enough ways that financial statement users can compare it (at least along some dimensions) without needing to be intimately familiar with more than one system” (Choi et al., 2001). It is important to note that convergence and comparability are not the same things. It is possible to compare accounting numbers even if they do not converge over time. For example, the accounting numbers produced by one system might be different from the accounting numbers

2

The FASB (1998) defines convergence to mean national accounting standards moving toward each other with the objective of increasing quality. The FASB also notes that the IASB also uses the term convergence in a slightly different fashion, the movement of national standards toward higher quality IFRS. This research ignores the subtle political difference in these two definitions of convergence. The term convergence will be used in this research in a strictly mathematical sense: two sets of different measures of the same thing that come closer over time. In this sense, convergence is intended to serve as evidence of accounting harmonization.

54


produced by a second system in some consistent and understandable fashion. In such a scenario, the financial statement user might be able to make simple adjustments to facilitate comparison. Accounting numbers produced by two different sets of accounting standards could also converge but not be comparable. For example, one set of accounting standards might facilitate income smoothing more than another set of accounting standards. In this scenario, the reported net income numbers might converge over time, but the set of accounting standards that facilitates income smoothing would be expected to show less variability than the set of accounting standards that discourages income smoothing. Such a situation would make comparison quite difficult. 3. Economic restructuring in the People’s Republic of China

The PRC is used in this study due to its growing importance in the world economy. It also provides an example of a nation whose accounting system is very different from the United States but is supposedly becoming more similar to Western accounting systems due to its desire to become more integrated into the world economy. In order to understand the accounting system of the PRC, it is necessary to view it in the larger context of the monumental economic restructuring that is taking place within that nation. As part of this restructuring, the PRC has developed an equity market to raise much needed capital for reform and expansion. In addition, the accounting system of the PRC is being transformed from a largely Soviet-era model designed to facilitate central planning to a model more compatible with the PRC’s growing international aspirations, culminating in membership in the World Trade Organization. 3.1. Economic restructuring and opening in the PRC

Since 1978, the PRC has been undergoing a process of opening and restructuring. Opening has meant the gradual change in economic philosophy from isolationism to integration with the world economy. Restructuring involves a transformation of the economy to make it more competitive with the other major economic powers, changing it from a centrally planned economic system to something called “a socialist market economy.” It is important to note that the transformation does not mean that the PRC is now a market economy. The socialist market economy is a unique blend of socialism and capitalism; it is a large portion of socialism along with certain changes toward a market economy. The basic idea of the socialist market economy appears to be preservation of the economy’s socialist core, state firms and state banks, while making incremental changes in other areas. These changes include establishing markets, eliminating central planning, and allowing non-state-owned industry to prosper (Steinfeld, 1999). This transformation parallels similar but much more abrupt and traumatic transformations in many Eastern European countries over the same time period. The Chinese government is acutely aware of the social disruptions that took place along with the economic transformations in Eastern Europe. As a result, the PRC has chosen a path that has been much more deliberate and much less traumatic. The cornerstones of the Chinese socialist economic system were and still are the state-owned enterprises (SOE’s). These are sometimes very large businesses financed and controlled by the central or local governments. They bought from and sold to other SOE’s and their debts were largely guaranteed by the state. It is important to note that these SOE’s also had an important social role in addition to their economic role. As large employers, they offered a system of guaranteed employment as well as various social services: the SOE’s operated their own housing units, schools, hospitals, recreation facilities, and retirement facilities.


55

3.2. Accounting in the PRC

As the Chinese economic system has been undergoing transformation, Chinese financial accounting has been transformed as well. The role of accounting information in a centrally planned economy is far different from its role in a socialist market economy. In a centrally planned economy, the users of accounting information are primarily the central planners responsible for making broad macroeconomic decisions as well as decisions on allocating government funds to various state-owned enterprises (SOE’s). To serve this group of users’ needs, the accounting information from the SOE’s needs to emphasize uniformity and rigidity and provide information to the central planners to show that the enterprises are meeting state-mandated production quotas, and efficiency standards. In a market economy, financial accounting serves external users, such as banks and equity investors, who provide capital and need high quality information to assess the loan or investment potential of an individual enterprise. As the users of an accounting system change, the accounting system must also change to provide relevant information to the appropriate users. International lenders and investors need credible information to compare a Chinese investment opportunity with similar investments in their own nation as well as other nations. The Chinese government has recognized these issues (Zhang, 1997) and has begun the development of a set of accounting standards to govern external financial reporting for Chinese companies. The PRC has also chosen a government directed method of setting accounting standards, with wide consultation including domestic and international experts. Rather than importing accounting standards, such as IFRS or the Generally Accepted Accounting Principles of another country such as the United States, they have chosen to study these and other available accounting standards. They have used these existing standards and developed a set of accounting standards for China that incorporates IFRS but reflects the unique aspects of the Chinese environment. The government felt this was the most constructive method noting that China was a large, developing country with wide regional development disparities, and the concept of a socialist market economy was quite new to China (Zhang, 1997). Thus, Chinese accounting standards are a rapidly evolving set of standards based largely on IFRS but with unique, evolving aspects, and heavy government involvement. To date 13 standards have been issued by the PRC. The last three of these standards were effective from 1 January 2001. An additional 17 standards are in exposure draft status. Regulations in the PRC financial markets require reporting under a variety of accounting regimes given the extent to which a company’s stock is traded. Companies that have issued Ashares (shares that trade in the PRConly) must follow regulations issued by the Ministry of Finance (PRC GAAP). Companies that have issued B-shares (shares that trade in the PRC but can be held by foreigners as well as PRC citizens) or N-shares (shares that trade overseas, for example New York, London or Singapore) must follow IFRS. Companies that have issued H-shares (shares that trade in Hong Kong only) may follow Hong Kong GAAP or IFRS. If a PRC company lists shares in the United States, the U.S. Securities and Exchange Commission requires that the company provide a reconciliation of net income and net assets per IFRS to net income and net assets per U.S. GAAP. These reconciliations are parts of a foreign firm’s Form 20-F registration with the SEC, and are sometimes found in a foreign firm’s annual report or in a supplemental report to North American investors. Thus, PRC companies listing on the New York Stock Exchange would likely prepare financial statements under PRC GAAP, IFRS (or sometimes Hong Kong GAAP3 ) and provide reconciliations to U.S. GAAP. 3

Hong Kong was returned to the People’s Republic of China in 1997. It is governed as a Special Administrative Region under a “one country–two systems” policy and retains its former financial systems and institutions. International relations

56


As noted, the historic, political, and economic factors at work in the People’s Republic of China have created an accounting system quite different from that of the United States. The pace and extent of economic change, driven by a desire to attract foreign investment and become integrated into the worldwide economy should result in an accounting system that is becoming more similar to IFRS and to U.S. GAAP. It is expected that this situation would cause IFRS reported financial statements (for PRC companies) not to be comparable to U.S. GAAP reported financial statements due to the vastly different underlying economic and accounting systems. These differences are expected to decline over time, however, as the movement toward convergence gains momentum. 4. Research questions

The questions addressed by this research involve the overall comparability and convergence of two sets of accounting standards, IFRS and U.S. GAAP, in one country, the PRC. The relationship between these two sets of accounting standards is examined through the lens of PRC companies listed on the New York Stock Exchange. The first two research questions address the issue of comparability: Research Question 1: Are IFRS and U.S. GAAP producing accounting numbers for PRC com-

panies listed on the NYSE that are comparable? Research Question 2: What are the reasons for any observed lack of comparability between IFRS and U.S. GAAP for PRC companies listed on the NYSE? If the accounting numbers produced by IFRS are comparable, then convergence is not really an issue. If, however, the numbers produced by each of the two accounting systems are not comparable, then the issue of convergence over time must be addressed. Research Question 3: Are IFRS and U.S. GAAP producing accounting numbers for PRC com-

panies listed on the NYSE that are converging over time? 5. Methodology

These research questions are addressed via a longitudinal study of the PRC firms traded on the NYSE. A firm-by-firm approach is taken in this study since investors look at the market on that basis. An alternative approach would involve averaging across all firms. These PRC firms provide a set of financial statements under IFRS and also provide a Form 20-F reconciliation of net income and net assets from IFRS to U.S. GAAP. Table 1 lists the universe of 14 PRC firms listed on the NYSE as of August 20, 2003, and shows their NYSE symbol, and industry classification. Form 20-F’s were obtained for each of the 14 companies listed in Table 1. Three companies, shown in italics, did not provide IFRS financial statements and were dropped from the study, resulting in a sample of 11 firms from a universe of 14 firms. The firms dropped were Aluminum Corporation of China, China Unicom, and CNOOC, which provided statements using Hong Kong GAAP rather than IFRS. Table 2 lists the 11 remaining PRC companies that constitute the sample

and defense are the responsibility of the central government of the People’s Republic of China. Hong Kong maintains its own accounting standards (Hong Kong GAAP) which were developed when Hong Kong was a British colony, and were heavily influenced by British accounting standards.

Table 1 People’s Republic of China companies listed on the New York Stock Exchange as of August 20, 2003 Company

Aluminum Corporation of China

Symbol

Industry

ACH

China Eastern Airlines Corporation China Petroleum and Chemical Corporation China Southern Airlines Company

ZNH

China Telecom Corporation

CHA

China Unicom CNOOC

Total assets 2002 in RMB millions (per IFRS)

Aluminum production

CEA SNP

Not in sample

Total Revenue 2002 in RMB millions (per IFRS) Not in sample

Net Income 2002 in RMB millions (per IFRS) Not in sample

Net income 2002 in RMB millions (per U.S. GAAP) Not in sample

Airlines operation Petroleum and petrochemicals Commercial airline services Fixed-line tele-communication

32,762 375,881

13,345 340,042

86,369 16,080

247,736 19,515

37,188

18,019

575

474

210,852

75,496

16,864

15,794

CHU CEO

Tele-communications Oil and gas

Not in sample Not in sample




Guangshen Railway Huaneng Power International

GSH HNP

11,258 48,461

2,518 18,474

557 3,921

598 3,895

Jilin Chemical Industrial Company

JCC

13,665

13,138

(1,023)

(685)

PetroChina Company Sinopec Beijing Yanhua Petrochemical Sinopec Shanghai Petrochemical Company Yanzhou Coal Mining Company

PTR BYH

Rail transportation Holding company/power plants Chemical products manufacturing Oil and gas exploration Petrochemical production

483,149 10,260

244,424 9,443

46,910 209

49,837 230

SHI

Petrochemical production

26,086

21,723

916

1,125

YZC

Coal mining

12,924

7,772

1,222

1,326

Total in sample (RMB million)

1,262,486

764,394

172,600

339,845

Total in sample (US$ million)

152,539

92,357

20,854

41,061

J .L . H a v er t y / J o u r n a l o f I n t er n a t i o n a l A c c o u n t i n g , A u d i t i n g a n d T a x a t i o n 1 5 ( 2 0 0 6 ) 4 8 – 7 1

Companies listed in italics reported under Hong Kong Accounting Standards and were excluded from the study.

5 7

5 8

Table 2 Comparability measures for People’s Republic of China companies included in the study Company

China Eastern Airlines Corporation China Petroleum and Chemical Corporation China Southern Airlines Company China Telecom Corporation Guangshen Railway Huaneng Power International Jilin Chemical Industrial Company PetroChina Company Sinopec Beijing Yanhua Petrochemical Sinopec Shanghai Petrochemical Company Yanzhou Coal Mining Company a

Gray’s index of comparability

Comparability determination

1996

1997

1998

1999

2000

2001

2002

Mean

At 5% materiality a

At 10% materialityb

0.90

0.80

0.82

0.90

1.04 0.36 1.07

0.26 0.90 0.22

0.97

0.95

0.78

0.58

0.94 1.00 0.50

1.04 0.86 1.05

0.92 0.82 1.03

0.54 0.59 0.85

0.93 1.00 0.90 0.87 0.73 0.79 0.88

0.24 0.89 1.46 1.00 0.92 1.01 0.56 0.91 0.89 0.84 0.92

1.14 0.93 0.80 0.47 0.93 0.92 1.16 0.90 1.38 0.39 0.79

0.35 0.82 1.21 1.07 0.93 1.01 0.51 0.94 0.91 0.81 0.92

0.68 0.78 0.93 0.84 0.94 0.99 0.71 0.90 0.92 0.73 0.92

Not comparable Not comparable Not comparable Not comparable Not comparable Comparable Not comparable Not comparable Not comparable Not comparable Not comparable

Not comparable Not comparable Not comparable Not comparable Comparable Comparable Not comparable Comparable Comparable Not comparable Comparable

Financial statements prepared under IFRS were deemed comparable to those prepared under U.S. GAAP at a 5% materiality threshold if both the average comparability index and the index for the most recent year in the study were between 0.95 and 1.05. b Financial statements prepared under IFRS were deemed comparable to those prepared under U.S. GAAP at a 5% materiality threshold if both the average comparability index and the index for the most recent year in the study were between 0.90 and 1.10.


5 8

Table 2 Comparability measures for People’s Republic of China companies included in the study Company

Gray’s index of comparability

China Eastern Airlines Corporation China Petroleum and Chemical Corporation China Southern Airlines Company China Telecom Corporation Guangshen Railway Huaneng Power International Jilin Chemical Industrial Company PetroChina Company Sinopec Beijing Yanhua Petrochemical Sinopec Shanghai Petrochemical Company Yanzhou Coal Mining Company

Comparability determination

1996

1997

1998

1999

2000

2001

2002

Mean

At 5% materiality a

At 10% materialityb

0.90

0.80

0.82

0.90

1.04 0.36 1.07

0.26 0.90 0.22

0.97

0.95

0.78

0.58

0.94 1.00 0.50

1.04 0.86 1.05

0.92 0.82 1.03

0.54 0.59 0.85

0.93 1.00 0.90 0.87 0.73 0.79 0.88

0.24 0.89 1.46 1.00 0.92 1.01 0.56 0.91 0.89 0.84 0.92

1.14 0.93 0.80 0.47 0.93 0.92 1.16 0.90 1.38 0.39 0.79

0.35 0.82 1.21 1.07 0.93 1.01 0.51 0.94 0.91 0.81 0.92

0.68 0.78 0.93 0.84 0.94 0.99 0.71 0.90 0.92 0.73 0.92

Not comparable Not comparable Not comparable Not comparable Not comparable Comparable Not comparable Not comparable Not comparable Not comparable Not comparable

Not comparable Not comparable Not comparable Not comparable Comparable Comparable Not comparable Comparable Comparable Not comparable Comparable

a

Financial statements prepared under IFRS were deemed comparable to those prepared under U.S. GAAP at a 5% materiality threshold if both the average comparability index and the index for the most recent year in the study were between 0.95 and 1.05. b Financial statements prepared under IFRS were deemed comparable to those prepared under U.S. GAAP at a 5% materiality threshold if both the average comparability index and the index for the most recent year in the study were between 0.90 and 1.10.


59

that provide financial reports in accordance with IFRS. For each of the 11 firms remaining in the sample, Table 1 also shows some summary statistics, including total assets, total revenue, net income per IFRS, and net income per U.S. GAAP. Researchers (Street, Gray, & Bryant, 1999) have noted a disturbing amount of non-compliance with IFRS by companies that claim to comply with IFRS. The issue is not deemed a problem in this study since the IFRS statements in the study were all audited and attested to by either Pricewaterhouse Coopers, KPMG or Deloitte Touche Tohmatsu. In addition, a 20-F filing is subject to U.S. Securities and Exchange Commission review as well. Annual reports and Forms 20-F for each of the years 1996–2002 were collected where possible, resulting in potentially 7 years of data. It was not possible to obtain 7 years of data for all companies in the sample since a number of the companies were not listed on the NYSE until after 1996. Net income per IFRS and U.S. GAAP, net assets per IFRS and U.S. GAAP, and earnings per share per IFRS and U.S. GAAP were obtained from the Supplementary Information for North American Shareholders or from the Form 20-F. All net income, net assets, and earnings per share numbers are reported in PRC currency, the renminbi (RMB) also sometimes called yuan. An additional measure, return on net assets, was calculated by dividing net income by net assets. 5.1. Comparability

In order to test comparability, an index of comparability was calculated for each year of data in the study. The index of comparability is: (Net incomeUSA

Net incomeIFRS )



59

that provide financial reports in accordance with IFRS. For each of the 11 firms remaining in the sample, Table 1 also shows some summary statistics, including total assets, total revenue, net income per IFRS, and net income per U.S. GAAP. Researchers (Street, Gray, & Bryant, 1999) have noted a disturbing amount of non-compliance with IFRS by companies that claim to comply with IFRS. The issue is not deemed a problem in this study since the IFRS statements in the study were all audited and attested to by either Pricewaterhouse Coopers, KPMG or Deloitte Touche Tohmatsu. In addition, a 20-F filing is subject to U.S. Securities and Exchange Commission review as well. Annual reports and Forms 20-F for each of the years 1996–2002 were collected where possible, resulting in potentially 7 years of data. It was not possible to obtain 7 years of data for all companies in the sample since a number of the companies were not listed on the NYSE until after 1996. Net income per IFRS and U.S. GAAP, net assets per IFRS and U.S. GAAP, and earnings per share per IFRS and U.S. GAAP were obtained from the Supplementary Information for North American Shareholders or from the Form 20-F. All net income, net assets, and earnings per share numbers are reported in PRC currency, the renminbi (RMB) also sometimes called yuan. An additional measure, return on net assets, was calculated by dividing net income by net assets. 5.1. Comparability

In order to test comparability, an index of comparability was calculated for each year of data in the study. The index of comparability is: 1−

(Net incomeUSA − Net incomeIFRS ) |Net incomeUSA |

(1)

This index was first used by Gray (1980) who called it an “index of conservatism.” This index has recently been called an “index of comparability” and has been used to compare accounting numbers reported for the same firm under different accounting regimes (Weetman, Jones, Adams, & Gray, 1998 and Street et al., 2000). Alternatives to Gray’s index of comparability are the H, I, and C indices developed by van der Tas (1988), refined by Archer, Delvaille, & McLeay (1995), and a T index developed recently by Taplin (2004). These indices measure the concentration of choices made by firms among alternative accounting treatments for a particular issue. They do not, however, measure the financial impact of those choices, nor are they useful to study differences obtained when different accounting systems are applied to the same firm. In Eq. (1), if Net incomeUSA is the same as Net income IFRS , then the value of this index will be 1.0. One disadvantage of this index is that it produces extreme values if the denominator is close to zero. The index has the advantage of highlighting any material differences between the two net incomes. An index less than 0.90 means that reported IFRS net income is at least 10% less than U.S. GAAP reported net income. Conversely, an index greater than 1.10 means that IFRS net income is at least 10% greater than U.S. GAAP reported net income. A threshold of 5% is used in this study as an indicator of materiality. Although there is no commonly accepted level of materiality, the 5% threshold is consistent with other researchers (Adams, Weetman, Jones, & Gray (1999) and Street et al., 2000) who report results based on both 5 and 10% levels of materiality. This index was calculated for each company for each year of the study. An average index over all the years in the study was also calculated for each company. Net income per IFRS and net income per U.S. GAAP were deemed comparable at a 5% materiality threshold if both the average comparability index and the index for the most recent year in the study were between

60


0.95 and 1.05. They were deemed comparable at a 10% materiality threshold if both the average comparability index and the index for the most recent year in the study were between 0.90 and 1.10. In order to explore the reasons for any observed lack of comparability, an empirical analysis is performed on the 20-F reconciliations for the most recent year in the study, 2002. Prior research using 20-F reconciliations (Harris & Muller, 1999) has grouped the reconciliation adjustments into six major classifications plus one category for any adjustments that did not fit into the six major categories. These categories, along with a brief explanation of some of the major differences between IFRS and U.S. GAAP are as follows: 1. Goodwill: U.S. GAAP, until very recently, required capitalizing goodwill and amortizing it over a period not to exceed 40 years. For accounting periods beginning after December 15, 2001, there is no amortization, but the goodwill must be reviewed for impairment each year. IFRS requires capitalizing the goodwill and amortizing it over a period not to exceed 20 years, along with an annual test for impairment. IFRS permits the charging of goodwill to owners’ equity in the year of acquisition. 2. Deferred income taxes: U.S. GAAP requires recognition of deferred income taxes on a comprehensive basis for all temporary differences and requires the use of tax rates that reflect future tax rates and laws. IFRS allow managers not to recognize deferred assets/liabilities if the book/tax difference is not expected to reverse in the foreseeable future. IFRS also allow managers to choose whether or not to adjust deferred amounts for changes in tax rates and laws. 3. Foreign exchange adjustments: IFRS provide much more flexibility than U.S. accounting standards. In the United States, foreign exchange gains and losses on forward contracts and hedges are recognized in net income or a component of equity in the period in which they occur. IFRS do not specify an accounting method. The United States requires the use of the current exchange rate when translating goodwill and fair value adjustments on foreign acquisitions. IFRS permit a choice between current and historical exchange rates. 4. Research and development expenditures: U.S. GAAP requires expensing of all research and development expenditures. IFRS permit the capitalization of development expenses. 5. Pensions: U.S. GAAP requires the use of the accrued-benefit method and current market-based assumptions. U.S. GAAP requires recognition of a minimum pension liability for under funded plans. IFRS permit the use of both accrued-benefit and projected benefit valuation methods and requires the use of long-term assumptions. IFRS has no requirement to recognize any liability for under funded plans. 6. Tangible asset revaluations: U.S. GAAP values property, plant and equipment on the historical cost basis subject to impairment. IFRS permit upward as well as downward revaluations of property, plant and equipment. Under IFRS this upward revision would result in additional depreciation expense. 7. Other: Any adjustments that do not fit into any of the above categories. The reconciliation items as described in the original 20-Fs were categorized using the above taxonomy, and the total amounts in each category were accumulated. In order to adjust for size, the total net income adjustments as well as the partial adjustments were deflated by IFRS net assets. This procedure results in a total net income adjustment between net income per IFRS and net income per U.S. GAAP expressed as a percentage of IFRS net assets. There are also a set of partial adjustments relating to each of the above categories again expressed as percentages of


61

IFRS net assets. The partial adjustments will add up to the total adjustment, so it is possible to determine what the major causes are of any reported difference between IFRS net income and U.S. GAAP net income. 5.2. Convergence

For those companies deemed not comparable, tests for convergence were conducted in two steps. The first step involves calculating difference measure for each company and each year in the study. Each difference measure captures the magnitude of the difference between a particular accounting number (e.g., net income) calculated under IFRS, and the corresponding accounting number calculated under U.S. GAAP. The second step involves calculating the trend over time of each of the difference measures for each company in the study. A negative trend would indicate declining differences over time, i.e., convergence. In the first step of the convergence test, four difference measures for each company and year in the study are calculated: a net income difference measure (DIFF NI ), a net assets difference measure (DIFFNA), a return on net assets difference measure (DIFF RONA), and an earnings per share difference measure (DIFFEPS ). DIFFNI was constructed by computing a yearly measure of the magnitude (absolute value) of the differences between net income per IFRS and net income per U.S. GAAP. This difference was then deflated by net assets per IFRS in order to correct for the effect of size of the firm on income, and is consistent with Harris and Muller (1999). The index of comparability was not used here to avoid the effect of the extreme values of the index when U.S. GAAP net income is close to zero. The net income difference measure is: DIFFNI =

|Net incomeUSA − Net incomeIFRS |

Net assetsIFRS

(2)

DIFFNA is constructed in a similar fashion by computing a yearly measure of the magnitude (absolute value) of the differences between net assets per IFRS and net assets per U.S. GAAP, again deflating by net assets per IFRS in order to correct for the effect of size of the firm: DIFFNA =

|Net assetsUSA − Net assetsIFRS |

Net assetsIFRS

(3)

DIFFRONA is the absolute value of the difference between return on net assets as computed under U.S. GAAP and return on net assets as constructed under IFRS: DIFFRONA = |Return on net assetsUSA − Return on net assetsIFRS |

(4)

DIFFEPS is the absolute value of the difference between earnings per share as computed under U.S. GAAP and earnings per share as computed under IFRS: DIFFEPS =

|Earnings per shareUSA − Earnings per shareIFRS |

Earnings per shareIFRS

(5)

In the second step of the convergence test, the trend over time of each of the above four difference measures for each company is examined. The beta coefficient of the trend of each difference measure over time is the convergence measure for each company in the study. Four convergence measures are obtained: a convergence measure of net income differences over time (BETANI ), a convergence measure of net assets differences over time (BETA NA), a convergence measure of return on net assets differences over time (BETA RONA), and a convergence measure

62


of earnings per share differences over time (BETA EPS). A negative coefficient would indicate declining differences over time, i.e., convergence. 6. Data analysis and results 6.1. Comparability

The data pertaining to the comparability of accounting numbers of PRC companies under IFRS and U.S. GAAP are presented in Table 2. This table reports the index of comparability by company for each year in the study, also the average comparability index for each company over all the years available. For example, the 1996 index of comparability for China Eastern Airlines was 0.90, indicating that net income for China Eastern Airlines per IFRS was approximately 90% of net income for China Eastern Airlines per U.S. GAAP. The last two columns in the table show the comparability determination at two different materiality thresholds. Net income per IFRS and net income per U.S. GAAP were deemed comparable at a 5% materiality threshold if both the average comparability index and the index for the most recent year in the study were between 0.95 and 1.05. They were deemed comparable at a 10% materiality threshold if both the average comparability index and the index for the most recent year in the study were between 0.90 and 1.10. The results show that IFRS and U.S. GAAP do produce materially different measures (10 out of 11 companies) of net income, at a materiality threshold of 5%, for PRC companies listed on the NYSE. Even at the more liberal materiality threshold of 10%, a majority (6 out of 11) companies in the study still exhibit materially different measures of net income per IFRS and net income per U.S. GAAP. For virtually all companies listed in this study (10 of 11), the average comparability index is less than 1.00, indicating net income per U.S. GAAP is generally higher than net income per IFRS. Reasons for the observed lack of comparability are explored in Table 3. Each row of this table shows an analysis of the total difference between net income per IFRS for a particular company and net income per U.S. GAAP for the same company for the year 2002. The first column shows the name of the company. The second column shows the total amount of the adjustments between IFRS net income and U.S. net income for that company, expressed as a percentage of net assets for that company. For example, for China Eastern Airlines, the total amount of the income difference is 2.19% of China Eastern Airlines IFRS net assets. This amount is called the total adjustment. A positive percentage here indicates that U.S. net income was higher than IFRS net income by an amount equal to approximately 2.19% of IFRS net assets. The next five columns show the amounts of five adjustment classifications from the Harris and Muller (1999) taxonomy. These are called partial adjustments. This taxonomy originally had seven adjustment classifications, but no PRC company in this study reported either a pension adjustment or an adjustment for research and development, so only five of the seven categories from the Harris and Muller taxonomy are used in this study. Thus, China Eastern Airlines reported a partial adjustment for deferred income taxes of −0.39%, meaning that the effect of differences in deferred taxes reduced U.S. net income from IFRS net income by an amount equal to 0.39% of IFRS net assets. This negative partial adjustment for China Eastern Airlines was offset by positive adjustments of 2.25% for tangible asset revaluations and positive 0.32% for other. Thus, for China Eastern Airlines, the category, tangible asset revaluations, is the most significant factor affecting the overall difference in net income per IFRS and net income per U.S. GAAP. In fact, the category tangible asset revaluations is the largest partial adjustment category for 8 of the 11 companies in the sample.


63

Table 3 Analysis of net income adjustments from IFRS to U.S. GAAP for the year 2002 Company

IFRS-U.S. net income adjustments Total adjustment as a percentage of IFRS net assets

China Eastern Airlines Corporationb China Petroleum and Chemical Corporationb China Southern Airlines Company China Telecom Corporation Guangshen Railway Huaneng Power Internationalb Jilin Chemical Industrial Company PetroChina Companyb Sinopec Beijing Yanhua Petrochemicalb Sinopec Shanghai Petrochemical Company Yanzhou Coal Mining Company

Partial adjustments as a percentage of IFRS net assetsa Goodwill

Deferred income taxes

2.19

−0.39

2.22

−0.98

−1.06

0.50

−0.86

0.40 −0.08

−0.07 −0.29

0.11

16.20 0.92 0.41

−0.86 −0.55

1.49

0.21

−0.23

1.04

−0.27

−0.64

Foreign exchange adjustments

0.05

Tangible asset revaluations 2.25

0.32

3.02

0.12

0.34

−1.90

0.42 0.47 0.09

−1.28

0.07

16.13

0.27

Other

2.54 1.07

−0.75 −0.10

0.93

0.31

1.95

a

None of the companies in this study reported an IFRS-U.S. GAAP adjustment due to pensions or research and development. b Some of the partial adjustments do not add to the total adjustment due to rounding.

In order to understand the effect of these revaluations, the procedures under IFRS are briefly outlined, and then the revaluations used by the companies in the study are analyzed. IFRS permits valuation of property, plant, and equipment at either historical cost, or at historical cost subject to periodic revaluation. U.S. GAAP permits only historical cost. A summary of IAS 16 (Hussey & Ong, 2005, p. 99) outlines the basic accounting for property, plant and equipment revaluations: • Revaluations should be carried out regularly, so that the carrying amount of an asset does not • • •

• •

differ materially from its fair value at the balance sheet date. The entire class of assets to which that asset belongs should be revalued. Depreciation is charged in the same way as under the cost basis. Increases in revaluation value should be credited to equity under the heading “revaluation surplus” unless they represent the reversal of a revaluation decrease of the same asset previously recognized as an expense, in which case it should be recognized as income. Decreases as a result of a revaluation should be recognized as an expense to the extent that they exceed any amount credited to the revaluation surplus relating to the same asset. Disposal of revalued assets can lead to a revaluation surplus that may be either transferred directly to retained earnings, or it may be left in equity under the heading “revaluation surplus.”

64


Table 4 Analysis of property, plant and equipment revaluations Company

Valuation basis except for land use rights

Revaluation dates

Purpose of revaluation

Direction of revaluation

China Eastern Airlines Corporation

Historical cost subject to revaluations

6/30/1996

Restructuring

Upward

12/31/2002

Periodic

Downward

9/30/1999

Restructuring

Upward

12/31/2000

Acquisition

Upward

12/31/1996

Restructuring

Upward

12/31/2001

Restructuring

Downward

3/6/1996

Restructuring

Upward

N/A

N/A

N/A

9/30/1994

Restructuring

Upward

2/28/1995

Upward

12/31/2002

HK Stock Exchange listing Periodic

Downward

6/30/1999

Reorganization

Upward

4/23/1997

Reorganization

Upward

Prior to 2001 N/A

Reorganization

Upward

N/A

N/A

China Petroleum and Chemical Corporation China Southern Airlines Company China Telecom Corporation


Huaneng Power International

Historical cost subject to revaluations Historical cost subject to revaluations Historical cost subject to revaluations Historical cost

Jilin Chemical Industrial Company


Guangshen Railway

PetroChina Company Sinopec Beijing Yanhua Petrochemical Sinopec Shanghai Petrochemical Company Yanzhou Coal Mining Company

Historical cost subject to revaluations Historical cost subject to revaluations Historical cost subject to revaluations Historical cost

An upward revaluation of property, plant, and equipment assets would result in net assets per IFRS being higher than net assets per U.S. GAAP. This would result in a downward adjustment from net assets per IFRS to net assets per U.S. GAAP on a 20-F reconciliation for a company that had revalued its assets upwards. Since the depreciation base would now be higher for a company that had revalued its assets upward, that company’s depreciation expense would also be higher resulting in a lower net income. This would result in an upward adjustment from net income per IFRS to net income per U.S. GAAP on a 20-F reconciliation for a company that had revalued its property, plant, and equipment upward. Downward revaluations of property, plant, and equipment are also possible. A downward revaluation under IFRS would require an upward adjustment of fixed assets from IFRS to U.S. GAAP on a 20-F reconciliation, and a downward adjustment of net income from IFRS to U.S. GAAP on a 20-F reconciliation. In addition, a downward revaluation could result in a direct effect on net income if the downward revaluation were in excess of any previous upward revaluation of the same asset. This would result in a lower net income per IFRS after the downward revaluation, and would require an upward adjustment of net income on a 20-F reconciliation of IFRS net income to U.S. GAAP net income.


65

Fixed asset revaluations for the companies in the study are analyzed in Table 4. Land, according to socialist theory in the PRC, is owned by the people. Therefore land is not owned by a corporation but is rather leased from the people through land use rights for a specified period of time. Land is normally found on the balance sheet of a PRC company as an operating lease, at historical cost subject to amortization. Prior to the adoption of IFRS by PRC companies, land was sometimes revalued upwards. Recently, these revaluations have been reversed since the land was usually granted from a related party. In these cases, the historical cost of the land has been considered nil. Adjustments to land use rights have been relatively minor, and are not considered in this study. Of the 11 companies in the study, only 2, Huaneng Power International and Yanzhou Coal Mining Company, use historical cost to value their property, plant, and equipment. These two companies are comparable at the 10% level of materiality. The remaining nine value their fixed assets on the balance sheet using historical cost with revaluations, and follow procedures similar to those outlined above. The vast majority of the revaluations have been upward, at the initial reorganization or restructuring of the enterprise. In the PRC, a huge amount of industrial restructuring took place as a result of the PRC’s movement from a planned economy to a more market-oriented and more open economic system. There have been some downward revaluations, however, notably China Eastern Airlines in 2002, China Telecom in 2001, and Jilin Chemical in 2002. There is some suspicion, however, that PRC state-owned entities have large amounts of unproductive property, plant, and equipment that may generate downward revaluations in the future. Thus, the most significant cause of lack of comparability between IFRS and U.S. GAAP in the PRC companies in the study is the fact that IFRS permits revaluation of fixed assets and U.S. GAAP requires historical cost as a basis for valuing property, plant and equipment. Nine of 11 PRC companies in this study have chosen to revaluate their property, plant and equipment. It is worth noting, however, that the concept of historical cost may not be appropriate in the PRC. U.S. GAAP makes the implicit assumption that historical cost has been determined by the free interplay of market forces. This assumption is clearly not appropriate in the PRC. 6.2. Convergence

The examination of comparability indicated that only 1 of the 11 companies, Huaneng Power International, was deemed to be comparable at a 5% materiality threshold. The remaining 10 companies were deemed to be not comparable and were tested for convergence. The results of the convergence testing are shown in Table 5. Each row of Table 5 shows the convergence testing results for a single company. Negative BETA coefficients on this table indicate evidence of convergence for each particular convergence measure. For example, the first row shows the results for China Eastern Airlines. For this company, BETANI , the coefficient of the trend line of the net income differences over the 7 years of the study, is 0.004. This indicates no evidence of net income convergence. The R2 measure for this coefficient is 0.0866. BETA NA, the net assets convergence measure for China Eastern Airlines, is −0.031. This negative coefficient indicates that the differences between net assets per IFRS and net assets per U.S. GAAP are becoming smaller over time, demonstrating convergence. This finding is supported by a relatively high R2 of 0.6533. BETARONA , the return on assets difference measure is −0.004 and the corresponding R2 is 0.0499 indicating weak convergence. BETA EPS, the convergence measure for earnings per share, is 0.336, demonstrating no evidence convergence. Table 6 summarizes the results of the study by company for both comparability and convergence.

6 6

Table 5 Convergence measures for companies with material differences in net income per IFRS and net income per U.S. GAAP Company

Years of data

Net income convergence measure BETANI

China Eastern Airlines Corporation China Petroleum and Chemical Corporation China Southern Airlines Company China Telecom Corporation Guangshen Railwayb Huaneng Power International Jilin Chemical Industrial Company PetroChina Companyb Sinopec Beijing Yanhua Petrochemical Sinopec Shanghai Petrochemical Companyb Yanzhou Coal Mining Company b a b

R2

Net assets convergence measure BETANA

Return on net assets convergence measure

R2

BETARONA

7 5

0.004 0.004

0.0866 0.4401

−0.031 −0.054

0.6533 0.9087

−0.004

7 3 7

−0.007 −0.072

0.3578 N/Aa 0.0000

−0.018

0.3573 N/Aa 0.2191

−0.019

5 7 4 7

Comparable—no material difference at 5% materiality 0.016 0.5525 0.013 0.1870 −0.003 −0.042 0.6085 0.9322 0.000 0.0049 0.000 0.0025

0.000

0.020 0.003

Earnings per share convergence measure

R2

BETAEPS

R2

0.0499 0.0002

0.336 0.016

0.1744 0.1017

0.4085 N/Aa 0.6899

0.018 0.021 0.008

0.001 N/Aa 0.6716

−0.013 −0.001

0.4315 0.9057 0.0449

−0.139 −0.022

0.012

0.5661 0.6253 0.0063

0.000

0.061 −0.001

0.048

7

0.000

0.0803

−0.008

0.7046

−0.000

0.0009

0.041

0.0117

7

0.000

0.0003

−0.030

0.5687

−0.005

0.2011

−0.054

0.3415

Data for China Telecom is N/A due to the small number of observations. These companies are comparable at a 10% level of materiality.

Table 6 Summary of results by company Company

Comparability

Convergence

5% Materiality

10% Materiality

Net income

Net assets

Return on net assets

Earnings per share

Number of measures

Yes Yes

Yes

2 1

Yes

Yes

3 1 1

China Eastern Airlines Corporation China Petroleum and Chemical Corporation China Southern Airlines Company China Telecom Corporation Guangshen Railway

No No

No No

No No No

No No Yes

Yes Yes a


Yes

Yes

Jilin Chemical Industrial Company PetroChina Company Sinopec Beijing Yanhua Petrochemical Sinopec Shanghai Petrochemical Company Yanzhou Coal Mining Company

No No No

No Yes Yes

Convergence was not tested since comparability at a 5% materiality threshold was indicated Yes 1 Yes Yes Yes Yes 4 Yes 1

No

No

Yes

No

Yes

Yes

Yes

Yes

Number of companies

1

5

6

6

3

a

Yes

3

China Telecom data were available for only 3 years, so a conclusion on convergence might be questionable.

1 3 18



6 7

Table 6 Summary of results by company Company

Comparability

Convergence

5% Materiality

10% Materiality

Net income

Net assets

Return on net assets

Earnings per share

Number of measures

Yes Yes

Yes

2 1

Yes

Yes

3 1 1

China Eastern Airlines Corporation China Petroleum and Chemical Corporation China Southern Airlines Company China Telecom Corporation Guangshen Railway

No No

No No

No No No

No No Yes

Yes Yes a


Yes

Yes

Jilin Chemical Industrial Company PetroChina Company Sinopec Beijing Yanhua Petrochemical Sinopec Shanghai Petrochemical Company Yanzhou Coal Mining Company

No No No

No Yes Yes

Convergence was not tested since comparability at a 5% materiality threshold was indicated Yes 1 Yes Yes Yes Yes 4 Yes 1

No

No

Yes

No

Yes

Yes

Yes

Yes

Number of companies

1

5

6

6

3

a

Yes

3

1 3 18

China Telecom data were available for only 3 years, so a conclusion on convergence might be questionable.


6 7

68


The first column of Table 6 shows the name of the company. Columns 2 and 3 show the results of comparability testing. For example, China Eastern Airlines was deemed not comparable at either 5 or 10% levels of materiality. Columns 4–7 summarize the convergence testing for each company. China Eastern Airlines, for example, showed evidence of net assets convergence and return on net assets convergence. The last column shows the total number of measures indicating convergence for each company. Only one company, Huaneng Power, was deemed comparable at a 5% materiality threshold. At a more liberal 10% materiality threshold, however, 5 of the 11 companies in the study were deemed comparable. Convergence testing was performed on the 10 companies not deemed comparable at the 5% materiality threshold. Three of these 10 companies indicated that measures of net income per IFRS and per U.S. GAAP were converging over time; 6 of the 10 companies indicated that measures of net assets per IFRS and per U.S. GAAP were converging over time, 6 of the 10 companies indicated that measures of return on net assets per IFRS and per U.S. GAAP were converging over time, and 3 of the 10 companies indicated convergence of earnings per share per IFRS and per U.S. GAAP. Of these 10 companies not deemed comparable, only 1, PetroChina Company, indicates convergence on all 4 measures. Two of the companies indicate convergence on three measures, one of the companies indicates convergence on two of the measures, and six of the companies indicate convergence on only one measure. The most frequently appearing convergence measure is net assets, since 6 of the 10 companies tested in this study demonstrate convergence on this measure. This is not surprising since property, plant, and equipment revaluations were shown to be the most significant cause of lack of comparability. As Table 4 shows, most PRC companies in this study revalued their assets upward

68


The first column of Table 6 shows the name of the company. Columns 2 and 3 show the results of comparability testing. For example, China Eastern Airlines was deemed not comparable at either 5 or 10% levels of materiality. Columns 4–7 summarize the convergence testing for each company. China Eastern Airlines, for example, showed evidence of net assets convergence and return on net assets convergence. The last column shows the total number of measures indicating convergence for each company. Only one company, Huaneng Power, was deemed comparable at a 5% materiality threshold. At a more liberal 10% materiality threshold, however, 5 of the 11 companies in the study were deemed comparable. Convergence testing was performed on the 10 companies not deemed comparable at the 5% materiality threshold. Three of these 10 companies indicated that measures of net income per IFRS and per U.S. GAAP were converging over time; 6 of the 10 companies indicated that measures of net assets per IFRS and per U.S. GAAP were converging over time, 6 of the 10 companies indicated that measures of return on net assets per IFRS and per U.S. GAAP were converging over time, and 3 of the 10 companies indicated convergence of earnings per share per IFRS and per U.S. GAAP. Of these 10 companies not deemed comparable, only 1, PetroChina Company, indicates convergence on all 4 measures. Two of the companies indicate convergence on three measures, one of the companies indicates convergence on two of the measures, and six of the companies indicate convergence on only one measure. The most frequently appearing convergence measure is net assets, since 6 of the 10 companies tested in this study demonstrate convergence on this measure. This is not surprising since property, plant, and equipment revaluations were shown to be the most significant cause of lack of comparability. As Table 4 shows, most PRC companies in this study revalued their assets upward upon reorganization, creating a revaluation surplus. This surplus was then reversed when reconciling from IFRS net assets to U.S. GAAP net assets on a firm’s 20-F reconciliation. Unless there were additional upward revaluations, this surplus remained the same. If a company’s assets were growing, the value of this surplus relative to total assets would decrease, resulting in net assets convergence in this study. The effect of property plant and equipment revaluations on a firm’s net income is less predictable, however. Property, plant and equipment revaluations affect a firm’s net income via the additional depreciation expense as the result of upwardly revalued assets. If income is volatile, however, the size of the additional depreciation expense under IFRS accounting would fluctuate relative to net income. Thus, IFRS net income and U.S. GAAP net income are less likely to converge over time than would IFRS net assets and U.S. GAAP net assets. This is shown in Table 6 in which 6 of the 10 companies indicate net assets convergence while 3 out of the 10 companies indicate net income convergence and earnings per share convergence. 7. Conclusions

This study provides evidence that despite movement toward harmonization and convergence, the PRC companies listed on the NYSE report materially different measures of net income under IFRS and U.S. GAAP. Despite much progress toward de jure harmonization between IFRS and U.S. GAAP, de facto harmonization has not been achieved at least as evidenced by PRC firms reporting under IFRS and reconciling to U.S. GAAP. Generally, net income for PRC companies reported per IFRS is less than net income per U.S. GAAP. The most significant reason for the observed lack of comparability was the revaluation of fixed assets. Since IFRS permits upward revaluation of property, plant and equipment, many PRC firms adopt this option for their IFRS reporting, most likely due to its widespread use under Hong Kong GAAP which also makes use of


69

revaluations. U.S. GAAP, however, requires historical cost accounting for its property, plant and equipment. These revaluations obviously affect net assets, but upward revaluations of property, plant, and equipment also cause reported net income differences due to the larger depreciation amounts reported under IFRS on upwardly revalued property, plant and equipment. Despite these comparability difficulties, there is evidence of de facto convergence among those PRC companies that were not deemed to be comparable. It is expected that the de jure convergence efforts of the IASB and the U.S. FASB should result in some evidence of de facto harmonization, and this is evidenced by the convergence measures in this study. 7.1. Limitations

This study is limited since it is not generalizable beyond the small but important subset of PRC companies that are NYSE-listed. NYSE-listed PRC firms are most likely different from nonNYSE-listed PRC firms simply due to the fact that their accounting is subject to U.S. Securities and Exchange Commission review. In addition to this review, the PRC government must itself approve a foreign listing. Other PRC firms would most likely have less sophisticated accounting systems since they are not subject to as much external scrutiny. In addition, the study only analyses certain accounting numbers: net income, net assets, return on net assets, and earnings per share. The possibility of cash flow differences between IFRS and U.S. GAAP was not addressed in this study. Finally, this study only shows that the accounting numbers produced under the various sets of accounting standards are different. It does not purport to show that the differences are important in the valuation of the firm. 7.2. Implications for investors

This research demonstrates that IFRS and U.S. GAAP still produce different accounting numbers. It is crucial that investors be aware of the GAAP under which an accounting number is calculated when comparing it to other companies or to benchmark statistics. Despite a long march toward accounting harmonization, investors may be misled if they compare an accounting number for a PRC Company reporting under IFRS to a corresponding benchmark statistic for U.S. GAAP reporting companies. To be specific, reported net income per U.S. GAAP is generally higher than net income per IFRS. On the other hand, reported net assets per U.S. GAAP are generally lower than net assets reported per IFRS. Most likely, these differences are related: U.S. GAAP requires assets to be valued at historic cost, while IFRS permits revaluation of net assets to market value. If a PRC company chooses to revalue its assets under IFRS, net assets per IFRS will likely be higher than net assets per U.S. GAAP. This upward revaluation would result in a higher annual depreciation charge under IFRS than under U.S. GAAP, causing a higher reported net income under U.S. GAAP. Investors in PRC firms are advised to pay particular attention to the valuation basis for assets at a PRC company. In addition, the suspected obsolescence of the plant assets of many state-owned enterprises in the PRC makes write-offs of assets highly likely. This research clearly shows that IFRS and U.S. GAAP are not comparable using Choi et al. (2001) definition of comparability: “ . . . similar in enough ways that financial statement users can compare it (at least along some dimensions) without needing to be intimately familiar with more than one system.” To be specific, users must know the process by which property, plant and equipment are valued and revalued. In the PRC, this difficulty is compounded by the reportedly massive amount of obsolete and non-productive property, plant and equipment held by the PRC

70


SOE’s. Additionally, the role of judgment in declaration of any impairments leads to the real opportunity for income manipulation. 7.3. Implications for researchers

This research underlines the need for researchers to develop viable operational metrics for comparability and convergence. Comparability of a set of financial statements prepared under IFRS and a set of financial statements prepared under U.S. GAAP, in this research, is determined by comparing reported net incomes. If net incomes are within plus or minus 5% of each other, the financial statements are deemed to be comparable. It is possible that other dimensions of a company might not be comparable even if the reported incomes under two different sets of accounting standards are within 5% of each other. The huge impact of upward revaluations of property, plant and equipment leads to an interesting dilemma. Disclosing of property, plant and equipment values at both revalued cost (under IFRS) and historic cost (under U.S. GAAP) would provide more information to shareholders than simply disclosing one of the property, plant and equipment values. Researchers need to investigate the value-relevance of these dual disclosures. In this scenario, a stated goal of convergence might actually work against the interests of the financial statement users. Particularly in the PRC, historical cost is a questionable concept, since much of the property plant and equipment is still owned by the state in a socialist market economy, and the valuation is not determined by the interplay of free market forces. The development and testing of an operational metric for convergence is an equally complex task. This research shows that convergence is a multidimensional concept, involving many different accounting numbers. Mathematical convergence of one set of accounting numbers does not imply mathematical convergence of a second set of accounting numbers, and this is demonstrated clearly by this research. Net assets convergence does not imply convergence of net income. Researchers need to introduce multiple measures of convergence. Future research directions suggested by this study include using this methodology to compare PRC standards to IFRS and U.S. GAAP. The sample size of this study might be expanded by including PRC firms that are listed on other stock exchanges than the NYSE such as NASDAQ. Accounting numbers other than net income and net assets, such as cash flow from operations might be considered in further research. Finally, the metrics introduced in this study might be tested in other contexts in which there is dual or three way reporting by a single entity under different sets of accounting standards. Acknowledgements

I would like to thank Kathleen Sinning (the editor) and two anonymous reviewers who provided helpful comments on earlier versions of this manuscript. References Adams, C. A., Weetman, P. E., Jones, A. E., & Gray, S. J. (1999). Reducing the burden of US GAAP reconciliations by foreign companies listed in the United States: The key question of materiality. The European Accounting Review, 8(1), 1–22. Archer, G. S., Delvaille, P., & McLeay, S. J. (1995). The measurement of harmonization and the comparability of financial statement items: Within-country and between-country effects. Accounting and Business Research , 25(98), 67–80.


71

Canibano, L., & Mora, A. (2000). Evaluating the statistical significance of de facto accounting harmonization: A study of European global players. The European Accounting Review, 9(3), 349–369. Choi, F. D., Frost, C. A., & Meek, G. K. (2001). International accounting (4th ed.). Upper Saddle River, NJ: Pearson Education, Inc. Deloitte Touche Tohmatsu (2002). GAAP differences in your pocket: IAS and U.S. GAAP. www.iasplus.com/ resource/IASUS.pdf , accessed October 18, 2002. Financial Accounting Standards Board (1998). International accounting standard setting: A vision for the future. Norwalk, CT: FASB. Financial Accounting Standards Board (2002). Memorandum of understanding: The Norwalk agreement . www.fasb.org/news/memorandum.pdf , accessed June 30, 2005. Gernon, H., & Meek, G. K. (2001). Accounting: An international perspective (5th ed.). New York: Irwin McGraw-Hill. Gray, S. J. (1980). The impact of international accounting differences from a security-analysts perspective: Some European evidence. Journal of Accounting Research , 18(Spring), 64–76. Harris, M. S., & Muller, K. A., III. (1999). The market valuation of IAS vs. U.S.-GAAP accounting measures using Form 20-F reconciliations. Journal of Accounting and Economics , 26 , 285–312. Hussey, R., & Ong, A. (2005). International financial reporting standards desk reference . Hoboken, NJ: John Wiley and Sons, Inc. International Accounting Standards Board (2002). IASB constitution (2000 ) as amended on 5 March 2002. www.iasc.org.uk/cmt/0001, accessed May 8, 2002. Larson, R. K., & Kenny, S. Y. (1999). The harmonization of international accounting standards: Progress in the 1990s? Multinational Business Review, 7 (1 (Spring)), 1–11. Steinfeld, E. S. (1999 (September)). Beyond the transition: China’s economy at century’s end. Current History, 98, 271–275. Street, D. L., & Gray, S. J. (1999). How wide is the gap between IASC and U.S. GAAP? Impact of the IASC comparability project and recent international developments. Journal of International Accounting, Auditing & Taxation , 8(1), 133–164. Street, D. L., Gray, S. J., & Bryant, S. M. (1999). Acceptance and observance of international accounting standards: An empirical study of companies claiming to comply with IAS’s. The International Journal of Accounting, 34(1), 11–48. Street, D. L., Nichols, N. B., & Gray, S. J. (2000). Assessing the acceptability of international accounting standards in the U.S.: An empirical study of the materiality of U.S. GAAP reconciliations by non-U.S. companies complying with IASC standards. The International Journal of Accounting , 35(1), 27–63. Taplin, R. H. (2004). A unified approach to the measurement of international accounting harmony. Accounting and Business Research, 34(1), 57–73. van der Tas, L. G. (1988). Measuring harmonization of financial reporting practice. Accounting and Business Research, 18(70), 157–169. Weetman, P. E., Jones, A. E., Adams, C. A., & Gray, S. J. (1998). Profit measurement and UK accounting standards: A case of increasing disharmony in relation to U.S. GAAP and IAS’s. Accounting and Business Research , 28(3), 189–208. Zhang, Y. (1997). China’s accounting standards: Development and implementation, Joint CICPA/HKSA Conference Paper . www2.jaring.my/capa/article/142.htm, accessed on February 12, 1999 (translated from speech in Mandarin, July 10, 1997).

Are IFRS and U.S. GAAP Converging

Recommend Documents