International Accounting Standards Board
30 Cannon Street
London EC4M 6XH
United Kingdom
12 September 2008
Discussion Paper, Preliminary Views on an Improved Conceptual framework for Financial
Reporting: The Reporting Entity
Dear Sirs
Roche, a Basel-based healthcare group, has a turnover of CHF 46 bn a year (EUR 28 bn.) derived
from pharmaceuticals and diagnostics. We employ nearly 80,000 people worldwide and have a
market capitalisation (end 2007) of CHF 171 bn. (EUR 103 bn.) As a large multinational group
which has been using IAS/IFRS since 1990 and has been closely involved in their development, we
greatly appreciate the opportunity to give you our comments on the Discussion Paper (DP.) First a
general remark, then we respond in a more focussed way to the specific questions raised in the DP.
We have developed our views in discussions and collaboration with other Swiss multi-nationals, so
our remarks parallel very closely those brought together in the letter from SwissHoldings.
A.
General Remarks
As we mentioned in our comments on the ED on chapters 1 and 2, we very strongly advocate that
no part of the revised Framework should be finalised separately, i.e. only when all of the component
parts are coherent and known. Please refer to that letter for further explanation. Below our
responses to the questions asked in the DP.
B.
Responses to Specific Questions
Q 1:Do you agree that what constitutes a reporting entity should not be limited to business
activities that are structured as legal entities? If not, why?
In principle we agree that what constitutes a reporting entity should not be limited to business
activities that are structured as legal entities. We can, however, envisage many circumstances in
F. Hoffmann-La Roche AG
CH-4070 Basel
Switzerland
Corporate Finance Accounting &
Controlling
Bldg/Room 52/1205
Tel. +41 61 68 84234
Fax +41 61 68 84282
alan.dangerfield@roche.com
1/7
which this may give rise to significant practical problems, e.g. in respect of strategic alliances, but it
is sufficient for the IASB to bear such problems in mind when considering individual standards.
Q 2: Do you agree that the conceptual framework should broadly describe (rather than precisely
define) a reporting entity as a circumscribed area of business activity of interest to present and
potential equity investors, lenders and other capital providers? If not, why? For example, do you
believe that the conceptual framework should establish a precise definition of a reporting entity?
If so, how would you define the term? Do you disagree with including reference to equity
investors, lenders and other capital providers in the description (or definition) of a reporting
entity? If so, why?
The notion of a reporting entity is central to IFRS, which prescribe how economic phenomena in
such an entity are to be reported. Consequently, we think that the Framework needs to say
something on the subject and to contain some sort of definition of a reporting entity.
However, pending the outcome of a proper debate and due process on the entity and parent
company perspectives, we leave this definition open.
On the reference to capital providers, we are very concerned that this implies that financial
statements should be presented from an entity perspective. We would prefer explicit reference to
equity investors, lenders and other capital providers.
Q 3: Do you agree that the risks and rewards model does not provide a conceptually robust basis
for determining the composition of a group reporting entity and that, except to the extent that it
overlaps with the controlling entity model (as discussed in paragraphs 102 and 103), the risks
and rewards model should not be considered further in the reporting entity phase of the
conceptual framework project? If not, why?
It seems to us that the DP is right in concluding that the risks and rewards model is not the right
single criterion for determining the composition of a group reporting entity. But neither do we think
that the control model on its own is adequate, as the whole question of SPEs demonstrates. It is
therefore not appropriate to stop further consideration of the risks and rewards model other than as a
“stand-alone” criterion. Since we would tend rather to favour putting economic interest more to the
forefront of determining the composition of a group, we would ask the Board to pursue risks and
rewards further.
Q 4: Assuming that control is used as the basis for determining the composition of a group
reporting entity, do you agree that:
(a)
control should be defined at the conceptual level?
(b)
the definition of control should refer to both power and benefits?
If not, why? For example, do you have an alternative proposed definition of control?
In Q 3, we explained our doubts about the assumption that control is used as the sole basis for
determining the composition of a group reporting entity. However, our answer to Q 4 presumes that
2/7
we do accept that assumption.
If control is to be used as the basis for determining the composition of a group reporting entity, it
should indeed be defined at the concepts level, and that definition should refer to both power and
benefits. We broadly accept the proposed working definition of control.
Q 5: Do you agree that the composition of a group reporting entity should be based on control? If
not, why? For example, if you consider that another basis should be used, which basis do you
propose and why?
See our answer to Q 3.
Q 6: Assuming that control is used as the basis for determining the composition of a group
reporting entity, do you agree that the controlling entity model should be used as the primary
basis for determining the composition of a group entity? If not, why?
We agree that the controlling entity model should be used as the primary basis for determining the
composition of a group reporting entity if control is used as the determinant. This is because we
believe that the group reporting entity should include the controlling entity as well as the entities it
controls.
Q 7: Do you agree that the common control model should be used in some circumstances only? If
not, why? For example, would you limit the composition of a group reporting entity to the
controlling entity model only? Or would you widen the use of the common control model? If you
support the use of the common control model, at least in some circumstances, do you regard it as
an exception to (or substitute for) the controlling entity model in those circumstances, or is it a
distinct approach in its own right? Please provide reasons for your responses.
In some circumstances, the common control model may lead to additional information which is
useful to capital providers, e.g. when entities are under common control of a family and cash is
transferred between the entities, so that an aggregate view might be required to give a complete
picture. However, in those circumstances in which the common control model would provide
additional information, we would not wish it to be used instead of the controlling entity model but
in addition to that provided by applying the controlling entity model.
Q 8: Do you agree that consolidated financial statements should be presented from the
perspective of the group reporting entity, not from the perspective of the parent company’s
shareholders? If not, why?
May we repeat the comments we have made in our response to the ED on Phase A, Objectives and
Qualitative Characteristics (emphasis added):
“While we can accept as a working assumption the view expressed in OB 5 that the
“information provided by general purpose financial reporting focuses on the needs of all
capital providers (those with a claim on the entity´s resources), not just the needs of a
particular group”, we do not accept the following sentence: “Financial reports reflect the
perspective of the entity rather than the perspective of the equity investors, a particular
3/7
group of equity investors or any other group of capital providers.” This statement is not
justified in the main text itself, nor does it obviously flow from anything that is said in that
text. The Basis for Conclusions argues simply that the entity perspective is more consistent
(than the proprietary perspective) “with the fact that the vast majority of today’s business
entities engaged in financial reporting have substance distinct from that of their capital
providers. As such, the proprietary perspective generally does not reflect a realistic view of
financial reporting”. As we argued in our General Remarks (q.v.), we believe that a hierarchy
is necessary in the definition of users which would first consider the needs of current and
potential shareholders – as the ultimate bearers of the enterprise risks - before seeing how
those of other capital providers are to be met. The entity approach does not do this, and we
have no indications that the parent shareholder approach currently used does not best provide
users – most of whom have a similar parent shareholder perspective – with the information
they need. We are also concerned that the ED is proposing the adoption of the entity
perspective even though, as the boards themselves have admitted, they do not know what the
implications of this proposal are for the rest of the Framework. The superiority of the entity
perspective is not apparent to us, and we believe that a full debate on the topic is vital before
the entity perspective could be considered as acceptable. Constituents need to properly
understand the differences between the perspectives and their implications. In any case the
entity approach does not align with the equity investor´s interest in the reporting entity´s
ability to generate cash flows. Owners of NCI do not always have access to the reporting
entity´s cash flows, but the entity´s perspective implies that also owners of NCI have an
interest in this cash flow. It is completely unacceptable to us for the IASB to reach a
conclusion on the perspective that the Framework should refer to without a comprehensive
debate of this issue.”
It is essential for there to be a comprehensive and in-depth debate about the perspective from which
general purpose financial statements should be presented before a conclusion is reached on the
subject, and for that debate a much more comprehensive analysis of the issue is needed than that set
out in either the ED or the DP.
Q 9: Do you agree that consolidated financial statements provide useful information to equity
investors, lenders and other capital providers? If not, why?
Yes, we believe that consolidated financial statements provide useful information to equity
investors, lenders and other capital providers. However, we must refine and expand upon that
answer. First, we would like to repeat certain further comments which we have made in responding
to the ED on Phase A, Objectives and Qualitative Characteristics (emphasis added):
“There are significant differences between the information needs of current and potential
shareholders and those of other capital providers. Without going deeply into detail, one
could for instance refer to the interest in maximizing the return on invested capital in the
former case while the “sufficing” generation of cash and earnings to cover payments falling
due must be the focus in the latter case. Many of the needs of both groups of capital
providers can be met by the same information, but there will nevertheless be areas in
which that is not the case and where the optimal reporting solutions for each group
conflict. Here it is necessary to assert a hierarchy among “capital providers”, with the
needs of current and potential shareholders being given priority in the financial
4/7
statements. The needs of other capital providers must also be met but, where there is a
conflict, by other means such as statistical disclosures.”
In our view consolidated financial statements are most useful to current and potential shareholders
of the parent company when prepared from the parent company perspective. They are useful – but
probably rather less so – to other capital providers whose business relationship is with specific legal
entities and who may also be particularly interested in (e.g.) the extent to which the parent can
actually get hold of assets lying in the subsidiaries.
Q 10: Do you agree that the conceptual framework should not preclude the presentation of
parent-only financial statements, provided that they are included in the same financial report as
consolidated financial statements? If not, why?
As follows on from our answer to Q 9, we would argue that the Framework should not preclude the
presentation of parent-only financial statements.
Q 11: With regard to the concept of control, in the context of one entity having control over
another, do you agree that:
(a)
establishing whether control exists involves assessing all the existing facts and
circumstances and, therefore, that there are no single facts or circumstances that evidence
that one entity has control over another entity in all cases, nor should any particular fact or
circumstances—such as ownership of a majority voting interest—be a necessary condition
for control to exist? If not, why?
(b)
the concept of control should include situations in which control exists but might be
temporary? If not, why?
(c)
the control concept should not be limited to circumstances in which the entity has sufficient
voting rights or other legal rights to direct the financing and operating policies of another
entity, but rather should be a broad concept that encompasses economically similar
circumstances? If not, why?
(d)
in the absence of other facts and circumstances, the fact that an entity holds enough
options over voting rights that, if and when exercised, would place it in control over
another entity is not sufficient, in itself, to establish that the entity currently controls that
other entity? If not, why?
(e)
to satisfy the power element of the definition of control, power must be held by one entity
only? In other words, do you agree that the power element is not satisfied if an entity must
obtain the agreement of others to direct the financing and operating policies of another
entity? If not, why?
(f)
having ‘significant influence’ over another entity’s financing and operating policy
decisions is not sufficient to establish the existence of control of that other entity? If not,
why?
5/7
We agree with statements (a), (c) and (f).
On (b) we agree that the concept of “control” should include temporary control but believe that, at
the standard level, some specific treatments for situations involving temporary control may be
appropriate for practical reasons.
On (d) we agree that options over voting rights are insufficient justification for control. This is
equivalent to recording values in the financial statements on the basis of what they would have been
if certain non-existing transactions had taken place. By analogy one should not record the principal
value of an FX option just on the assumption that it could have been exercised. However, we think
this may best be dealt with at standards level.
We have considerable difficulties to accept (e). We are extremely unhappy with ED 9, Joint
Arrangements, which takes the position that control cannot be shared and therefore excludes JVs
from consolidation. Firstly, our experience of joint ventures confirms that these form part of the
overall business of the entity and are not just add-on investments. Secondly, if neither of the
venturers controls the JV, who does control it? Or is it something which is not controlled (and this
does not match with our perception of the business realities)? Just as there are constraints in running
entities in which we control 90% of voting rights (e.g. US minority protection rules), so there are
also constraints with 50/50 JVs in terms of the contractual agreements and the need to decide
together with the other venturer what the operating and financial policies are. We have no difficulty
with having a separate (sub)concept of “joint control” in which control is shared, but in our view
this still has all of the practical, economic characteristics of control.
Q 12: Should any of the above control issues be addressed at the standards-level rather than at
the concepts level? If so, which issues and why?
Our initial view is that issue (d) may be better dealt with at the standards level. The question of the
treatment of put as well as call options would need to be clarified in this context.
Q 13: Are there any other conceptual issues, relating either to the control concept or to some
other aspect of the reporting entity concept, that are not addressed in this discussion paper and
should be addressed at the concepts level? If so, which issues and why?
No.
Sincerely,
F. Hoffmann-La Roche AG
Dr. Erwin Schneider
Head of Corporate Finance
Accounting & Controlling
Alan Dangerfield
Corporate Finance Accounting & Controlling
External Relations
6/7
7/7