Description
The paper discusses some issues and problems associated with attempts to classify national systems of
accounting. Attention is given to classification experiences in other disciplines. The analysis aims to show
the incoherence of taxonomies which rely upon appeals to objectivity and tries to present a case for the
development of classifications which explicitly focus on the concept of an “accounting system
Pergamon Ac c ount f ng, Or gani zat i ons and Soc i el y, Vol. 20, No. 7/S, pp. 639-664, 1995
Copyright 0 1995 Elsevier Science Ltd
Printed in Great Britain. All rights reserved.
0361-3682/95 $9.50+0.00
03613682(95)00008-9
THE VERY IDEA OF CLASSIFICATION IN INTERNATIONAL ACCOUNTING
ALAN ROBERTS
Department of Economi cs, Uni versi ty of Readi ng
Abstract
The paper discusses some issues and problems associated with attempts to classify national systems of
accounting. Attention is given to classification experiences in other disciplines. The analysis aims to show
the incoherence of taxonomies which rely upon appeals to objectivity and tries to present a case for the
development of classifications which explicitly focus on the concept of an “accounting system”.
The history of attempts at taxonomy in the field
of comparative international accounting
reveals some contrasting methodological
approaches. The earliest attempt was by
Hatfield (1966, reprinted from a paper of
1911) who, on the basis of some observations
about differences in accounting practices in
four countries, suggested a three-group classifi-
cation: U.S., U.K. and Continental Europe. In
more recent times, one set of approaches has
relied on extrinsic classifications of accounting,
i.e. accounting in different countries was clas-
sified in terms of factors which influence the
nature and practice of accounting in those
countries. Thus, Mueller (1967) produced a
four-group classification in terms of “patterns
of accounting development” followed (Mueller,
1968) by another classificatory attempt based on
business environments. Other examples of clas-
sification relying on extrinsic factors would
include Seidler (1967) and the AAA (1977)
whose classifications in international account-
ing were based upon “spheres of influence”
and “zones of influence”, respectively. More
recent examples are provided by Gray (1988)
who, using a research paradigm of Hofstede
(1980, 1983), proposes “culture area classitlca-
tions in the context of combinations of account-
ing values” (p. 11) and Shoenthal (1989) who
considers classifications in terms of professional
competencies.
A second approach to classifying inter-
national differences in accounting has used
intrinsic tests of divergence and resemblance;
accounting in different countries is classified in
terms of the nature of accounting practices in
those countries. A number of researchers (da
Costa et al, 1978; Frank, 1979; Nair & Frank,
1980; Goodrich, 1982) have applied factor ana-
lysis to Price Waterhouse data (1973, 1975,
1979) on accounting practices to produce
clusters or groupings of countries. In the Nair
& Frank study, for example, consideration of
measurement practices was separated from
that of disclosure practices. In the former
case, a four-country grouping was obtained
which was broadly similar to the groups
obtained by Frank in 1979; in the latter case,
a seven-country classification was proposed.
Following these classifications, Nair & Frank
tried to associate economic and cultural vari-
ables with the groupings obtained, with mixed
results (Gray, 1988).
A further approach to classification in the
field of comparative international accounting
is that of Nobes (19884). Nobes focuses on
the measurement practices used in the finan-
cial reporting of public companies based in
639
640 A. ROBERTS
countries in the developed Western world in
1980, and, inspired by the example of biologi-
cal classification and its methods of numerical
taxonomy, takes 9 factors which might provide
discriminating features of those practices in 14
countries which were then scored on a scale of
O-3 in order to construct a morphology. A vari-
ety of techniques, relying on concepts of taxo-
nomic distance, were used to produce clusters
of countries labelled as “species”; further
levels of clustering led to the creation of a
family tree hierarchy (Fig. 1). Significantly, in
view of earlier criticisms of the use of Price
Waterhouse data for classification purposes
(Nobes, 1981) the scoring of the nine factors
was made on the basis of information from a
number of sources including works of refer-
ence, financial reports, detailed questionnaires
and personal knowledge obtained from work-
ing in the countries concerned. In distancing
himself from previous intrinsic classifiers,
Nobes emphasized the need for judgement
(Nobes, 1984, pp. 31-32) in attempting to
derive classifications in international account-
ing.
This paper sets out, on the basis of these
various taxonomic attempts, to explore some
abstract issues about the derivation of classif%
cations in comparative international account-
ing and to suggest some approaches which
might be developed in the future. Attention is
placed upon the experience of other disci-
plines in terms of dealing with the taxonomic
process; in particular, debates about classifica-
tion in biology are used to highlight a number
of perspectives on the past development of
international accounting classifications. The
paper has a specific focus upon the Nobes
model for a number of reasons. Firstly, unlike
previous classifications, this model has not
been the subject of major review and com-
ment. Secondly, its development followed an
explicit consideration of general issues in clas-
sification, including a review of attempts to
classify in other disciplines. A third reason is
that it represents a research paradigm in the
field and has spawned, to use Kuhn’s terminol-
ogy (Kuhn, 1970), at least one working out of
“normal science” (Al Najjar, 1987).
The structure of the paper is one which
starts with a general review of the nature of
the taxonomic process itself and, within this
Measurement Practices
I
The paper discusses some issues and problems associated with attempts to classify national systems of
accounting. Attention is given to classification experiences in other disciplines. The analysis aims to show
the incoherence of taxonomies which rely upon appeals to objectivity and tries to present a case for the
development of classifications which explicitly focus on the concept of an “accounting system
Pergamon Ac c ount f ng, Or gani zat i ons and Soc i el y, Vol. 20, No. 7/S, pp. 639-664, 1995
Copyright 0 1995 Elsevier Science Ltd
Printed in Great Britain. All rights reserved.
0361-3682/95 $9.50+0.00
03613682(95)00008-9
THE VERY IDEA OF CLASSIFICATION IN INTERNATIONAL ACCOUNTING
ALAN ROBERTS
Department of Economi cs, Uni versi ty of Readi ng
Abstract
The paper discusses some issues and problems associated with attempts to classify national systems of
accounting. Attention is given to classification experiences in other disciplines. The analysis aims to show
the incoherence of taxonomies which rely upon appeals to objectivity and tries to present a case for the
development of classifications which explicitly focus on the concept of an “accounting system”.
The history of attempts at taxonomy in the field
of comparative international accounting
reveals some contrasting methodological
approaches. The earliest attempt was by
Hatfield (1966, reprinted from a paper of
1911) who, on the basis of some observations
about differences in accounting practices in
four countries, suggested a three-group classifi-
cation: U.S., U.K. and Continental Europe. In
more recent times, one set of approaches has
relied on extrinsic classifications of accounting,
i.e. accounting in different countries was clas-
sified in terms of factors which influence the
nature and practice of accounting in those
countries. Thus, Mueller (1967) produced a
four-group classification in terms of “patterns
of accounting development” followed (Mueller,
1968) by another classificatory attempt based on
business environments. Other examples of clas-
sification relying on extrinsic factors would
include Seidler (1967) and the AAA (1977)
whose classifications in international account-
ing were based upon “spheres of influence”
and “zones of influence”, respectively. More
recent examples are provided by Gray (1988)
who, using a research paradigm of Hofstede
(1980, 1983), proposes “culture area classitlca-
tions in the context of combinations of account-
ing values” (p. 11) and Shoenthal (1989) who
considers classifications in terms of professional
competencies.
A second approach to classifying inter-
national differences in accounting has used
intrinsic tests of divergence and resemblance;
accounting in different countries is classified in
terms of the nature of accounting practices in
those countries. A number of researchers (da
Costa et al, 1978; Frank, 1979; Nair & Frank,
1980; Goodrich, 1982) have applied factor ana-
lysis to Price Waterhouse data (1973, 1975,
1979) on accounting practices to produce
clusters or groupings of countries. In the Nair
& Frank study, for example, consideration of
measurement practices was separated from
that of disclosure practices. In the former
case, a four-country grouping was obtained
which was broadly similar to the groups
obtained by Frank in 1979; in the latter case,
a seven-country classification was proposed.
Following these classifications, Nair & Frank
tried to associate economic and cultural vari-
ables with the groupings obtained, with mixed
results (Gray, 1988).
A further approach to classification in the
field of comparative international accounting
is that of Nobes (19884). Nobes focuses on
the measurement practices used in the finan-
cial reporting of public companies based in
639
640 A. ROBERTS
countries in the developed Western world in
1980, and, inspired by the example of biologi-
cal classification and its methods of numerical
taxonomy, takes 9 factors which might provide
discriminating features of those practices in 14
countries which were then scored on a scale of
O-3 in order to construct a morphology. A vari-
ety of techniques, relying on concepts of taxo-
nomic distance, were used to produce clusters
of countries labelled as “species”; further
levels of clustering led to the creation of a
family tree hierarchy (Fig. 1). Significantly, in
view of earlier criticisms of the use of Price
Waterhouse data for classification purposes
(Nobes, 1981) the scoring of the nine factors
was made on the basis of information from a
number of sources including works of refer-
ence, financial reports, detailed questionnaires
and personal knowledge obtained from work-
ing in the countries concerned. In distancing
himself from previous intrinsic classifiers,
Nobes emphasized the need for judgement
(Nobes, 1984, pp. 31-32) in attempting to
derive classifications in international account-
ing.
This paper sets out, on the basis of these
various taxonomic attempts, to explore some
abstract issues about the derivation of classif%
cations in comparative international account-
ing and to suggest some approaches which
might be developed in the future. Attention is
placed upon the experience of other disci-
plines in terms of dealing with the taxonomic
process; in particular, debates about classifica-
tion in biology are used to highlight a number
of perspectives on the past development of
international accounting classifications. The
paper has a specific focus upon the Nobes
model for a number of reasons. Firstly, unlike
previous classifications, this model has not
been the subject of major review and com-
ment. Secondly, its development followed an
explicit consideration of general issues in clas-
sification, including a review of attempts to
classify in other disciplines. A third reason is
that it represents a research paradigm in the
field and has spawned, to use Kuhn’s terminol-
ogy (Kuhn, 1970), at least one working out of
“normal science” (Al Najjar, 1987).
The structure of the paper is one which
starts with a general review of the nature of
the taxonomic process itself and, within this
Measurement Practices
I