1 Measuring the Convergence of National Accounting Standards with International Financial Reporting Standards: The Application of Fuzzy Clustering Analysis ☆ Xiaohui Qu a and Guohua Zhang b* a Center for Accounting Studies, Xiamen University, Xiamen, Fujian, 361005 P.R. China Tel.: +86-592-2181172; Fax: 86-592-2181520 E-mail address: [email protected](Xh Qu) b Post Dr. Working Station in Business Administration, School of Management, Xiamen University, Xiamen, Fujian, 361005 P.R. China Tel.: +86-451-89158253; Fax: +86-451-82533823 E-mail address: [email protected](Gh Zhang) ☆ Project Study on Quantitative Methodology for International Accounting (70572091) supported by NSFC and Key Project Study on the International Harmonization Effect of China Accounting Standards (05JJD630030) supported by the MOE, China. * Corresponding author.
28
Embed
Measuring the Convergence of National Accounting Standards with ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Measuring the Convergence of National Accounting Standards with International Financial
Reporting Standards: The Application of Fuzzy Clustering Analysis☆
Xiaohui Qua and Guohua Zhangb*
aCenter for Accounting Studies, Xiamen University, Xiamen, Fujian, 361005 P.R. China
☆ Project Study on Quantitative Methodology for International Accounting (70572091) supported by NSFC and Key Project Study on the International Harmonization Effect of China Accounting Standards (05JJD630030) supported by the MOE, China. * Corresponding author.
The globalization of the world’s economies has inevitably brought with it moves to establish a
single set of financial reporting standards. In May 2000, the IOSCO completed the assessment of
IASC core standards, including their related interpretations (the IASC, IAS2000). Members of IOSCO
were encouraged to use the IASC standards to prepare their financial reporting for cross-border
offerings and listings, supplemented where necessary to address outstanding substantive issues at a
national or regional level, or to use waivers of particular aspects of IASC standards without requiring
further reconciliation under exceptional circumstances. The IASB, after its reconstruction in 2001,
3
positioned its objectives1 to develop and promote the use and rigorous application of a single set of
global accounting standards as set out in IASCF Constitution2. The single set of accounting standards
is conceived to be the path of bring accounting convergence globally. More and more countries and
regions are obliged or volunteered to accept completely or reconcile its national accounting standards
to the international ones. Listed companies in EU countries have been required to adopt IFRS since
2005 for preparing their consolidated financial statements, and non-listed companies are encouraged.
Financial statements ended after 15 November 2007 and prepared using IFRS by foreign private
issuers in the US, will be accepted without reconciliation to US GAAP. Some countries have directly
converged its accounting model with the IFRS, such as Australia, while some have not because of
environmental differences or legal process. Being a member of IASB, China has made great efforts to
converge its accounting standards with IFRS. In February 2006, The Ministry of Finance issued a new
set of accounting standards including 1 fundamental standard and 38 specific standards, which is
officially said to have realized substantial convergence with IFRS, except few items such as related
party scope, reversion of impairment, and business combination within co-control.
Though achievements gained, it is worthy to aware that numerous and significant differences
from IFRS still exist in many national accounting systems. And even for those countries that have
adopted IFRS directly, certain differences may still exist during the implementation of the standards. It
is generally accepted that standards are not only the means of achieving the convergence of financial
reports but also the objectives of convergence. In such circumstances and given the important role of
formal convergence, reliable measurements of the progress in achieving convergence are critical.
Extant research in the evaluation of accounting convergence has mainly focused on the measurement
of material convergence, while the methods and methodology for the measurement of formal
harmonization are scarce and inconsistent. This study focuses on exploring the method and
methodology for measuring formal convergence, and proposes a new method of matching and
fuzzy clustering analysis to assess the convergence progress of national accounting standards
with IFRS from whole and single standards, and then employs it to evaluate the achievements
made in China in its efforts to converge CAS with IFRS. It is expected that this paper could advance
the study of formal convergence, meanwhile, could benefit the globalizing capital markets and other
users of financial reporting in helping them to assess the quality and comparability of the financial
information provided by Chinese companies.
4
It is necessary to differentiate some terminologies first so as to be convenient for our discussion.
Firstly, we need to make a distinction among harmonization, standardization, and convergence to
understand the target of our measurement. Van der Tas (1988) originally defined harmonization “a
coordination, a tuning of two or more objects”. Tay and Parker (1990) made a further differentiation
between harmonization and harmony, standardization and uniformity. “Harmonization (a process) is a
movement away from total diversity of practice. Harmony (a state) is therefore indicated by a
'clustering' of companies around one or a few of the available methods. Standardization (a process) is a
movement towards uniformity (a state). It includes the clustering associated with harmony, and
reduction in the number of available methods. Harmony and uniformity are therefore not dichotomous.
The former is any point on the continuum between the two states of total diversity and uniformity”
(Tay and Parker, 1990). “Converge” means to tend or move toward one point or one another.
Convergence is the act of converging and especially moving toward union or uniformity1. The
destination of both standardization and convergence is uniformity. Similar with the relationship of
harmony and uniformity, harmony and convergence are also not dichotomous. These terminologies, in
fact, reflect different stages of international accounting standards development. IASB’s new objectives
of developing and promoting a single set of accounting standards in 2001 means that accounting
standards has moved from international harmonization towards global convergence. We conceive this
convergence a standardization process. The degree of harmonization and convergence reveals
progresses made in accounting internationalization process. For convenience, we will not differentiate
their differences in our discussions below.
Secondly, we need to make distinction between formal harmonization and material harmonization
so as to make clear the contents and methods employed in measuring international accounting
convergence. Consensus of opinion has formed that accounting harmonization includes accounting
standards harmonization and accounting practice (financial reports) harmonization. Accounting
standards harmonization refers to harmonization between regulations, and is called formal or de jure
harmonization. Accounting practice harmonization refers to harmonization between practices applied
by companies, regardless of whether such practices are affected by regulations, and is called material
or de facto harmonization. Formal accounting harmonization is the path and basis of achieving
material accounting harmonization. Giving the importance of formal accounting harmonization, we 1 http://www.merriam-webster.com/dictionary/convergence
5
will focus on measuring the convergence of accounting standards, regardless of whether such
standards are followed in practice or not. The measuring methods should be the ones that are suitable
to the features of accounting standards.
The rest of the paper is constructed as follows. First, we begin by reviewing the methods existed
in literature for measuring the success achieved in effecting convergence between any two sets of
accounting standards. We then propose and demonstrate matching and fuzzy clustering analysis
methods for assessing the convergence progress of national accounting standards with IFRS. Finally,
China’s new accounting standards are taken as an illustrative example, its convergence level is
measured by using this new method, and conclusion is drawn.
2. Literature review
Most of prior studies on assessment of accounting convergence have focused on the measurement
of material harmonization. The indices, such as I and C indices, as well as H index introduced by Van
der Tas (1988), are popularly used in the literature. Several authors made further improvements or
developments to those indices, such as Archer et al. (1995), Hermann and Thomas(1995), Morris and
Parker(1998), Aisbitt(2001), Pierce and Weetman(2002), Taplin(2004). The data resource used
for investigating the level of harmonization among the practices and treatments in above studies was
accounting information prepared by companies. However, because material vs. formal harmonization
studies are substantially different, indices used in the measurement of material harmonization are not
valid for evaluating formal harmonization. Some researchers such as Adhikari and Tondkar (1992),
Lainez et al. (1996), and Rahman et al. (1996), have developed, though tentatively and scarcely, new
methods to measure advances in formal harmonization. In more recent years, further contributions
have been made to this area of research by Garrido et al. (2002) and Fontes et al. (2005). Their
research advances accounting harmonization studies from an experimental stage, where methodology
and analytical techniques were proposed and tested on particular samples of accounting issues of
different countries, to method and methodology stage, where any accounting issues and countries
could be evaluated within the framework of the research design.
The early exploration on formal harmonization concentrated on the requirement of regulations in
different countries. Adhikari and Tondkar (1992) examined 35 stock exchanges’ listing requirements
as the source for accounting regulation, and identified some determining environmental factors for
6
such listing requirements. Lainez et al. (1996) analyzed and quantified the existing discrepancies
among the information requirements imposed by the stock exchanges of 13 countries. They found that
there are more differences among periodical reporting requirements than among additional
information to be disclosed in the case of private offerings.
At the same time, Rahman et al. (1996) conducted a tentative research on measuring formal
harmonization between countries. They measured the formal harmonization level that has been
achieved between the two ‘neighboring’ countries of Australia and New Zealand. Taking the
disclosure and measurement requirements stipulated in accounting standards, legislative requirements,
and stock-listing requirements as their data source, they identified the different categories of
requirements that had achieved higher or lower degrees of harmonization between the two countries
by multiple-discriminating analysis. The requirements of disclosure and measurement in each item
were considered discrete data. Mahalanobis-like distances was used to measure the distances among
categories. The results indicate a higher level of harmony on measurement requirements, and a lower
level of harmony on disclosure requirements. The study of Rahman et al.(1996) gives us a new
perspective on studying formal convergence. The major problem in their measurement is that
Mahalanobis distance is primarily defined for continuous variables, whereas requirements of
disclosure and measurement are discrete data, which should be treated purely in a descriptive fashion.
Besides, it is difficult to explain the degree of harmonization by absolute distances.
We conceive research on measuring formal harmonization between countries a lateral study.
Thereafter, Garrido et al. (2002) conducted a longitude study (within single standard over different
period) on formal harmonization. They focused their study on evaluating the progress of IASC. The
IASC achievements all through three stages of its standard-setting activity were evaluated by
Euclidian distances. The results proved that the IASC had made great progress in regard to the level of
harmony achieved through the accounting standards it had issued or revised. Though progress has
been made in measuring formal harmonization, the main query about this study is the use of Euclidian
distances. As it is an absolute distance, the result of Euclidian distances can only shows the difference
between the items compared, but could not reflect similarities or dissimilarities of the items compared.
This flaw makes it unsuitable for use in analyzing the convergence among different standards (lateral)
or the progress achieved within one standard (longitude).
Aware of the flaws existed in the literature, Fontes et al. (2005) proposed Jaccard’s coefficient
7
and Spearman’s coefficient to assess the progress of formal harmonization between any two sets of
accounting standards. They measured formal harmonization between Portuguese Accounting
Standards and IFRS in three phases of accounting convergence by using Euclidian distances, Jaccard’s
coefficients and Spearman’s coefficients. The result proved that coefficients methods are better than
Euclidian distances. As Euclidian distances yields absolute values and is not easily interpretable, it can
only be used to assess the progress in dynamic terms. In contrast, the results yielded by Jaccard’s
coefficients can be interpreted both in dynamic terms (increasing results over time denote formal
harmonization advances), and static terms. The calculation of Spearman’s correlation coefficients
reinforced these results and provided further evidence of the progress achieved in converging
Portuguese accounting standards with IFRS. The use of coefficients offers similarity of standards
between countries and makes up the flaws of distances. This methodology is believed to be applicable
to analyze the level of convergence between different regulations at different points in time or among
different countries.
Synthesizing the methods and methodology employed in extant literature on measuring formal
harmonization (including between standards and within standard), we can find following features:
Firstly, data source used in formal harmonization studies are mostly regulations, standards, and
stock exchanges’ listing requirements. These data are featured with qualitative (nominal) variables
instead of quantitative variables. This determines that measuring methods used in evaluating formal
accounting harmony should be the one that is suitable to the features of nominal variables. We learn
from statistics that the most suitable means to calculated nominal variables are coefficients. So far, in
extant research only Jaccard’s coefficients and Spearman’s coefficients proposed by Fontes et al. (2005)
can meet this requirement. Euclidian distances and Mahalanobis distances are mostly suitable to
calculating ordinal or interval variables.
Secondly, samples in formal harmonization studies are mostly accounting issues with variety
choices on accounting treatments. Measurement and disclosure requirements in regulations are not
exhausted. This leaves doubt that even though the accounting choices are exactly the same between
standards compared, there are still possibilities that the two set of standards are actually not
convergent when their scope, the terminologies and the measurement criteria are not exactly the same.
And these differences would eventually affect material harmonization.
Thirdly, variables chosen in formal harmonization studies are similar with the variables chosen in
8
material harmonization studies, in which accounting treatments are classified as “required”,
“recommended”, “allowed”, or “not permitted”. This type of classification may work when measuring
the convergence of material harmonization as the strength of accounting methods contemplated will
affect accounting practice. But as for regulations, there will be no substantial difference for which
treatment is required or allowed, as there are no favorites in regulations, while there are favorites in
practice. Better comparison between regulations, we conceive, should be similar or un-similar
requirements and stipulations (comparison items) in regulations compared, which include details of
standard’s scope, terminologies concerned in the standard, measurement criteria on accounting
elements, accounting methods, re-measurement and disclosure etc., as these details will also affect
reliability and relevance of financial reporting eventually.
Fourthly, each formal harmonization study in extant literature concluded from the whole without
considering single standard’s effect on the convergence of two sets of accounting regulations. This
will miss some important information such as which single standard has achieved higher convergence
while others not. The latter is very important for researchers and standard-setters in their research and
further harmonization efforts, and for financial information users in their understanding and making
use of accounting information as well.
Considering features existed in the context of formal harmonization literature, this study aims to
provide a new method for measuring the convergence between any two sets of accounting standards
both from totality and from single standard, and to cluster single standard according to its level of
convergence by fuzzy clustering analysis.
3. Fundamental principles of fuzzy clustering analysis
Clustering aims at dividing a data set into groups or clusters that consist of similar data. Fuzzy
clustering analysis is a method used in multivariate statistical analysis. It employs mathematical
methods to identify close or estrange relationships of cases so as to classify them objectively. The
close or estrange relationships are expressed by using some measure of similarity or distance. The
former is usually measured by simple relevant coefficients; the latter is measured by distances. Our
study aims at to measure the similarity or dissimilarity between any two sets of accounting standards,
so coefficients could be the most suitable method.
Matching coefficients is mostly used in measuring close or estrange relationships of nominal
9
variables. The similarity degree ( 12d ) between any two cases of 1 and 2 is formulated in formula (1).
21
112 mm
md+
= (1)
Where 1m denotes the numbers of indicators that matches, 2m denotes the numbers of indicator
that does not match. 12d is between 0 and 1. If 21d =1,the two cases are exactly the same; if 21d =0,
the two cases are completely different.
As object per se possess ambiguous features under many circumstances. It will correspond to
reality even more when fuzzy algorithm is introduced into clustering analysis. The essence of fuzzy
clustering analysis is to construct fuzzy matrix according to the attributes of the object studied.
Classifications are then made according to assigned membership degrees based on the fuzzy matrix.
We suppose x overall domain. If A is a function of x with values of [0,1], A is called fuzzy set,
and is denoted as:
AA
A∉∈
=xx
01
)x( (2)
Its membership grades is between close section of [0,1].
The procedures of fuzzy clustering analysis are as follows:
Step 1: Choosing indicators for fuzzy clustering analysis. { }n21 x,,x,x:X is supposed to
represent overall objects classified. Features of each object ix are labeled by a group of datum
( )imi2i1 x,,x,x .
Step 2: Transforming original data so as to eliminate the effect of different dimensions. The idea
and methods of the transformation are as same as the ones used in systematic clustering analysis,
which include standardization, gradation, and logarithm etc.
(4)recognition prerequisites for inventories (5)recognition prerequisites for net realizable value
(4)4/in Framework (5)16、17/30-32
Measurement criteria
(6)elements of inventories costs (7)purchase costs (8)manufacturing costs (9)other costs (10)expenses excluded from inventories costs (11)expenses included in inventories costs (12)inventory costs from service providers (13)agricultural products costs from living assets (14)inventory costs from investors (15)inventory costs from debt restructuring and merge (16)inventory costs from non-monetary transactions
(7)6/11 Net price method is used in IAS. Total price method is used in CAS.
(13)12/20 CAS recognizes it on
book value, while IAS on fair value less selling expenses.
(14)11/0 (15)12/0 (16)12/0
Measuring methods
(17)standard cost (18)retail method (19)specific identification (20)FIFO (21)weighted average cost
(17)0/21 not regulated in CAS (18)0/22 not regulated in CAS (19)14/23-24 (20)14/25-27 (21)14/25-27
Re- measurement at the end of the
period
(22)amortization on packages and supplies (23)damages and losses in inventories taking should be recognized as reporting period expense (24)adjustment on book value when sold (25)measured at lower of cost and net realizable value (26)allowance method for inventories shrinkage (27)revision of allowance for inventories shrinkage
Graph 1 Fuzzy Clustering Graph of CAS Convergence with IFRS
26
References:
Adhikari, Ajay & Rasoul H. Tondkar.1992. “Environmental Factors Influencing Accounting
Disclosure Requirements of Global Stock Exchanges,”Journal of International Financial
Management and Accounting, 4(2): 75-105
Aisbitt, Sally.2001. “Measurement of Harmony of Financial Reporting Within and Between Countries:
The Case of the Nordic Countries,”The European Accounting Review, 10(1): 51-72
Archer, Simon, Pascale Delvaille & Stuart McLeay. 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
Fontes, Alexandra, Lúcia Lima Rodrigues & Russell Craig. 2005. “Measuring Convergence of
National Accounting Standards with International Financial Reporting Standards,” Accounting
Forum, 29: 415-436
Garrido, Pascual,Ángel León & Ana Zorio. 2002. “Measurement of Formal Harmonization Progress:
The IASC Experience,”The International Journal of Accounting, 37: 1-26
Herrmann,Don & Wayne Thomas. 1995.“Harmonisation of Accounting Measurement Practices in the
European Community,”Accounting and Business Research, 25(100): 253-265
Lainez, Jose A., Susana Callao & Jose I.Jarne. 1996. “International Harmonization of Reporting
Required by Stock Markets,”The International Journal of Accounting, 31(4): 405-418
Morris, Richard D. & R.H.Parker.1998. “International Harmony Measures of Accounting Policy:
Comparative Statistical Properties,”Accounting and Business Research, 29 (1): 73-86
Pierce, Aileen & Pauline Weetman.2002.“Measurement of de facto Harmonization: Implications of
Non-Disclosure for Research Planning and Interpretation,”Accounting and Business Research,
32(4):259-273
Rahman, Asheq, R., Hector Perera & Siva Ganeshanandam. 1996.“Measurement of Formal
Harmonization in Accounting: An Exploratory Study,”Accounting and Business Research, 26(4):
325-339
Taplin, Ross H.. 2004.“A Unified Approach to the Measurement of International Accounting
Harmony,”Accounting and Business Research, 34(1): 57-73
Tay, J.S.W. & R.H. Parker.1990.“Measuring International Harmonization and
27
Standardization,”Abacus, 26(1):71-88
Van der Tas Leo G.1988. “Measuring Harmonisation of Financial Reporting Practice,”Accounting and
Business Research, 18(70): 157-169
28
1 IASB adopted its objectives in 2001 when reconstructed as follows: (a) to develop, in the public interest, a single set of high-quality, understandable, and enforceable global accounting standards, which require high-quality, transparent, and comparable information in financial statements and other financial reporting, to help participants in the world’s capital markets and other users make economic decisions; (b) to promote the use and rigorous application of those standards; and (c) to bring about the convergence of national accounting standards and IAS to effect high-quality solutions. On 21 June 2005, the IASC Foundation added the following objective in its constitution: (c) in fulfilling the objectives associated with (a) and (b), to take account of, as appropriate, the special needs of small and medium-sized entities and emerging economies. 2 The objectives of the IASC, as set out in its Constitution, revised in 1982, are (a) to promote worldwide acceptance and observance of the IAS it formulates, and (b) to work generally for the improvement and harmonization of regulations, accounting standards, and procedures relating to the presentation of financial statements.