I-2.4: The principle of the neutrality of standards (the example of accounting standards)

The subject of neutrality brings a certain number of preliminary ideas and reminiscences to mind.

First of all, the image of Buddha comes to mind, because it alone expresses the extent to which neutrality is an endless subject, since by thinking about the fact that he isn’t thinking about anything, Buddha is still thinking about something.
It also makes me think of contemporary music’s constant attempts to attain a form of neutrality: yet this goal remains unattainable, because it is evident that the search for the neutral, in a sound or a timbre, can probably never be found.
I also think of the concept of mathematical zero, which is so impossible to evoke in a simple way that it must be something else, something more than simply the neutral.
The neutral also evokes Edmond Jabès’ “point”: a poetic ideal of which nothing can be said, not even whether it is neutral or not.

Another example is Roland Barthes’ course on the Neutral[1], because when one reads Barthes, one understands how very multifarious the neutral is, and how its immensity cannot be reduced to one single conception of the Neutral: the neutral cannot be understood by any single person.

Lastly, I think of Vladimir Jankélévich, who described the neutral as a different and elusive reality.

Of course, I am embarrassed to mention philosophy since my article comes directly after a philosopher’s[2], but the neutral and the question of neutrality truly live on in my memories. It was the same for Maurice Blanchot’s reminiscences, when he concluded that the neutral is like something between day and night, even though we know that there is nothing between day and night.

In short, I have only found a single postulate in all of philosophy’s work on the question of neutrality, which is that the neutral does not exist.

And yet, the neutral remains no less an obsessional topic, especially in economics.
Indeed, the theory of markets leads us to believe that there is a certain homology between doing nothing and being perfectly neutral with regards to the economic system. 
Thereby, economics meet philosophy in the person of Edouard Le Roy for whom being neutral meant doing nothing. And yet, can we not still debate this point: can we not affirm that doing nothing is still doing something?

But, returning to economics, might we imagine that the ‘invisible hand’ is the embodiment of neutrality? Is it not the neutral par excellence in that it acts, and yet does nothing? It is possible, but this is something of which we will never be certain, because this hand is, indeed, invisible.
The recent economic crisis has shown that this hand is sometimes asleep, and the concept seems to me to be too basic to be a foundation of any sort.

But, taking economic thought into consideration, it is possible to get lost looking for the neutral: indeed, alongside market theories, doesn’t the theory of economic action (of which Keynesian analysis is the topical illustration) prescribe exactly the opposite by saying that we must not be neutral?

“To be neutral, or not to be neutral?” We must realize that this is an obsessional question that pits those who believe neutrality is the optimal posture against those who affirm that the absence of neutrality is ideal.

In accounting, you might think that the question of neutrality is not an issue, because the extremely powerful rule that bookkeeping must be kept in double-entries, neutrality ought to be the last property lacking from accounting.
Nothing is less true: you would be mistaken in believing this simply because neutrality is also an obsessional topic for accounts.

It would even be possible to affirm that accounting standards embody most of the debates surrounding neutrality. Some affirm that only neutral accounting standards are quality standards, and that accounting’s ideal is to be neutral.
This ideal is perfectly understandable, since accounting has set for itself the goal of representing economic activity. But, it is possible to consider that one of representation’s functions is to be neutral; that a faithful representation is doubtlessly a neutral one, and therefore neutrality should be accounting’s principal goal.

I am nonetheless going to try to show that neutrality does not exist in any sort of standard-setting, and especially not in accounting standards. I am also going to try to draw a number of practical conclusions from this observation.

But before this, what is the general organization of accounting standards-setting? This is a dualistic system because unlisted companies are obliged to follow French accounting standards set by the Autorité des Normes Comptables (ANC – French accounting standards authority), whereas listed companies use international accounting standards set by the International Accounting Standards Board (IASB).

How is it then possible to define what a standard-setter is? There are many different types of standard-setting bodies. However, according to my definition, a regulator is obliged to deal with a system’s pathologies, and therefore has a tendency to overestimate them, whereas a standard setter simply wants to deal with generalities. This is what interests us here, within our narrow point of view: the neutrality of standards.

How is it possible to demonstrate that the neutrality of standards does not exist?

• First of all, by explaining a static vision of the neutrality of standards, taken by itself, and by demonstrating that a standard is always a choice, and always has a cause.

• Then, by providing a look at the neutrality of standards ‘in motion’, or with a dynamic perspective, considered through its effects, by pointing out that a standard always has consequences.

• Lastly, by showing that the general interest is the best name that can be given to the neutral, I will outline how the art and practice of neutrality can be transformed into the art and practice of the general interest.

1. A standard is a non-neutral and motivated choice

The very act of creating a standard, like any legislation, is above all a choice. A standard is never made in order to do nothing, in order not to have an effect. For example, nobody can imagine that useless standards would be set, such as rules stipulating that we must breathe or that we must get dressed every morning.

When a standard is created, it is because action is desired, a result is desired. A standard is always created in order to change what existed before and is therefore not neutral. Standards are set in order to provide rules or to create order, not to attain the neutral.

 The paradox of a non-neutral image in accounting

The paradox concerning accounting standards is that, as I said before, such standards aim at creating a neutral representation of economic activity. But, this is far from reality and I think two examples will suffice to demonstrate this.

First of all, let us take the expression of a faithful image. Indeed, if this image were faithful by nature, why would we need to add the word faithful?

The same goes for the expression ‘true and fair view’. It is because we know that accounting does not necessarily provide a fair view that we need the word ‘fair’, in order to pretend that it is an absolute vision, whereas in reality, no image can provide an absolute vision in such cases.

Similarly, even though accounting means holding up a mirror, you will agree that there are thousands of ways to orient that mirror, to decorate the stage, and will also agree that accounting will never produce one single image. I would like to add that since we admit that using any single accounting standard it is possible to represent economic phenomena in different ways, the existence of many different accounting standards will simply multiply the number of images that we can ultimately obtain.

It is for these reasons that we can consider accounting standards to be the result of a choice, and can also identify these choices as the product of a particular context.

 A paradox that can be explained by historical contingencies

Using two pillars — the idea of an instant representation, and the idea of a long-term representation — we can easily identify the cyclical evolutions of accounting standards.

Indeed, in the beginning of the 1800s, accounting standards gave an instant representation of economic activity because the biggest fear in the pre-industrial revolution period was corporate bankruptcy. For this reason, creditors had to be sure that they had all information that could potentially protect them from bankruptcies.

But, as industrialization progressed, and despite bankruptcies and economic crises, a medium-term approach was developed, which was thought better able to protect everyone’s interests and to represent durable and prudent economic models.

Accounting standards are therefore subject to cycles. The tumultuous destiny of fair value during the twentieth century—revered during the speculative period up to 1929, banished in 1934, held in contempt until 1970, and finally adored today with the return of speculation—is enough to show that the long-term evolution of accounting standards is in function of cycles, which themselves are consequences of choices made in function of particular contexts.

 Contingent and non-neutral accounting standards taken as a language emanating from agents’ individual wills

Accounting standards are therefore choices linked to their respective historical periods. Each period invented its own accounting standards according to its needs in function of economic activities, stakes, and risks. These standards are established in function of various­—more or less legalistic, more or less codified—cultures, each one with its own contingencies.

For example, IFRS are an answer to the emergence of financial markets, and are intended to help them grow.

Setting accounting standards is far from a neutral process: it is entirely a contingency, which refutes the idea that any sort of neutrality is possible in standard setting.

Accounting’s progressive globalization might lead us to think that we are currently engaged in a trend towards the possibility of true neutrality, embodied by a single, global standard. This is not the case, because as the previous examples show that accounting standards, be they national or international, are not, and never will be, neutral.

The criterion of neutrality is therefore not a criterion for setting accounting standards. It is not mentioned in the 2002 European regulation enumerating the quality standards for IFRS to be adopted into European law. Indeed, the regulation mentions the necessity of being able to count on intelligible and comprehensible standards that allow for comparisons to be made. However, it never once says that accounting standards must be neutral.

2. Standards taken as a dynamic source of non-neutral effects

The crisis was a decisive moment, for it taught us something relatively new. After each crisis, we were generally accustomed to observing that certain standards were lacking, and that it was necessary to make new ones.

But this time, it became clear that existing standards might be defective and might have disastrous consequences. I believe that three examples will be enough to prove this point.

 Standards can have disastrous consequences

First of all, let us take the example of consolidation. There is a clear opposition between those who think that controlling a company means owing 50% plus one of its shares, and those who think that in order to identify who is in control, a much more subtle analysis of the rights and obligations of the different shareholders is required.

In the United States, this first conception, which is part of their accounting standards and which seems to make sense, is an attempt at expressing a neutral vision.
In truth, by imposing a simple, academic presentation, this standard does not go far enough in determining who is really in control.
Therefore, many companies and ‘vehicles’ bearing instruments of credit that should have been consolidated were not consolidated; they were therefore not subject to the prudential standards associated with credit, and were not included on banks’ books. The risks were swallowed up in this black hole, and the consequences are now known to all.
It is clear that in a case such as this, accounting standards have real consequences and are not neutral, even when they seem to be.

Setting aside previsions for bank loans is another interesting example. Indeed, it seemed perfectly neutral to mandate that provisions be constituted only for proven risks, and not probable future risks that were subject to hypotheses and subjectivity.
But here, too, this choice is far from neutral and bears significant consequences when considered in real-life circumstances.
By only taking proven risks into account, reserves were insufficient and left institutions unprotected when risks that were only probable became manifest, and credit to SMEs became rarified.

Lastly, mark to market once again shows us how accounting standards that were intended to be neutral are not neutral at all. First of all, market valuations might seem to be entirely neutral because the market could be seen as the best arbiter between irreconcilable opinions.
But, once again, this rule has very serious consequences when it is used in unsuitable circumstances.
Indeed, not all products have a market value because they are not exchanged, and cannot therefore be attributed a market price unless they are misleadingly valued. This was the case of ‘toxic products’: when their price crashed in a way that was unprecedented in the history of finance, it created the greatest crisis in decades.

In all of these examples, standards seem neutral, but their unintended effects on the economy are, on the contrary, considerable.

 Standards always have a long-term effect capable of surpassing their internal logic

No standard is neutral because they all generate behaviors and they intervene in situations where they create consequences that go beyond the standard setter’s initial anticipations and intentions.
Standards always have an impact that goes far beyond their internal logic. By thinking about them ‘in motion’, and by understanding that standards can have a considerable impact on the economy, this point seems even more obvious.

Therefore, the only way to come close to a certain form of neutrality would be to think about how to minimize impact. Trying to create neutral standards is not enough. We have to go beyond by thinking about how to neutralize their impact on the economy, and do away with their dynamic effects to prevent standards from creating behaviors that might destabilize economic activity.

Therefore, a neutral standard is a standard that does not destabilize actors by itself, but instead stabilizes their interactions and allows them to benefit from a certain amount of collective security.
Imagine a fence around a tennis court, or traffic lights: they are limits that do not change the rules of the game, they simply provide security and boundaries; they are a form of discipline, a definition and a measure of flows that work without changing behavior, whether it be a tennis match, a journey by automobile. When drawing up consolidated accounts, the same should be achieved.

But what is this ultimate and precious virtue of security, exactly?

In all evidence, it is the stability that was called “financial security” in the 2003 French law on financial security. It is what we today call “financial stability” with the creation in the United States, in Europe, and in France, of Financial Stability Councils.

It clearly seems that this sought-after object is a superior, common, and public good to which accounting standards must contribute, since it is the most elementary representation of economic activity. Therefore, it must provide the most neutral possible representation. It must never destabilize, instead, it must be a vector of stability.

This is the way I think we must define the goal of neutral standards.
Indeed, in the foregoing, I have highlighted the definition of the general interest.

Since neutrality is fundamentally unattainable, the only goal we can set for ourselves is the general interest. Although we must admit that it varies according to the times, customs, cultures, and epochs, we must also recognize it as the highest value.

In this respect, in the aforementioned 2002 European regulation, we must take note of the criteria of a ‘European public good’ amongst the various other technical criteria.

 The only way to take the neutral into account in standard setting: the general interest

Therefore, it seems clear that in order to create standards able to satisfy the general interest, we must avoid technical ideology, be capable of encompassing all points of view, and making sure that the results are in accordance with a broader and higher economic policy that goes beyond the simple ideal of accounting standards’ internal quality.

This is part of an “accounting policy”. Accounting is not simply a technical subject, it must be perceived as a political issue that justifies public interest in the governance of accounting standards setting through the recognition that accounting standards setting bodies will have a significant impact on the capacity of attaining the ideal goal I have just described.

It will also be necessary to prevent the many risks of excess. Two examples will suffice.
First of all, standards setters must never try to change the system by themselves, using an illegitimate will based on technical arguments that might impact the economy without previous consultation between stakeholders that goes beyond accounting’s strict limits.
Similarly, it is primordial to prevent standards setters from setting standards using anti-abuse clauses, which use standards and twist them in order to punish practices of the day, rather than setting permanent and robust rules. This is a confusion of the roles between standard setter and the verification of individual practices that are the responsibility of other public institutions.

The general interest is the best label I can give to the neutral. So how can we make the art and practice of neturality into the art and practice of the general interest?

3. Making the art and practice of neutrality into the art and practice of the general interest

Four elements are necessary to achieve this goal: a mindset, institutions, procedures, and a fundamental discussion.

 What mindset?

First of all, I have been greatly impacted by the conversations I have had with some of my foreign counterparts, who are very influenced by technical expertise or by sociological postures that are highly specific to their individual experiences.

Indeed, some of them categorically refuse to take the general interest into consideration, without really saying why. Some of them feel that the general interest need only be taken into account when dealing with major issues; otherwise, technical expertise must take the upper hand. But for French people, the general interest should be constantly taken into account, and is truly anchored in our mental and political universe.

It quickly becomes apparent that there is a fertile dimension upon which the general interest should be based: the personal ethics of standard setters.
In truth, combining very different personal ethical standards is one of standard setting’s major issues.

 What institutions?

The IASB and many other bodies have developed the concept of independence.

This concept seems to me to be extraordinarily defensive, isolating from the rest of the world, and negating the absence of standards’ neutrality towards the economy. On the contrary, we must work towards collective action and liability. How can a body whose standards have an impact on the economy refuse to participate in discussions, not speak with anyone and not allow anyone to speak to it? Or worse, to pick and choose its interlocutors?
On the contrary, we must insert standard setting into a vaster social ensemble. Of course, we must respect the independence of standard setting, but this independence must be part of a broader, continuous dialogue with all stakeholders on a global level.

The best way to achieve this open system of standard setting is to make sure that within the standard setting body itself exists a diversity of origin, profession, and even opinion.

Of course, this means that many organizational choices will have to be made in terms of members’ statuses, whether they are full-time or part-time participants, relationships between permanent services and other aspects of the standard setting body, as well as the method of funding. We have made efficient choices in France. It is by advancing towards these concrete goals that standard setting will progress towards the general interest.

This is the debate that we are spearheading today in the plans to reform the IASB.

 What procedures?

Again, we are faced with a bottomless subject. However, this subject can be found in four areas.

First of all, the method of setting standards. Should it be top down or bottom up? Should it lie in the hands of standard setting professionals or those who draw up accounts? Should it be based on the strict formalism of procedural rules, or should it be based on direct, spontaneous contributions from stakeholders? At the ANC, we have decided upon procedures that make room for practitioners’ demands and participations; ultimately, decisions are made in the general interest according to a progressive and collective approach.

Currently, all of these questions are far from being resolved. One of the true stakes for the future is finding a global approach to a balanced compromise between various paradoxes that satisfies all stakeholders.

Furthermore, it is necessary to think of standard setting as a procedure in and of itself, and making sure that it is perfect.

Accounting standard setting lags behind prudential standards setting, which has been capable of performing life-sized tests in order to play out the destabilizing effects of various standards.
Currently, we have nothing capable of performing ex ante tests to measure the expected effects of a new accounting standard, but are devoting our efforts to this subject in order to be able to compete with prudential standards setting on this point.

Furthermore, it is necessary to be able to foresee evolutions in the implementation of standards, principally in verifying the quality of standards implementation using ex post studies.
In this way, we will be better equipped to measure the coherency of a standard’s implementation with its internal logic, and to be able to discern its impact on the economy to be sure that it has not caused instability.

Finally, it is absolutely indispensible to come up with procedures able to verify that standards are homogeneously implemented. We are currently working towards developing procedures that would be better adapted to fulfilling this goal.

 What fundamental debate?

 As we have seen, thinking about the art and practice of the general interest raises a lot of questions.
Amongst these questions, we must reexamine the following fundamental issues: “who does accounting serve, and what purpose does it serve?” At the ANC, we have already fully integrated this primordial question into all of our research programs, and have begun reviewing the answers to these questions.

Of course, these discussions take a lot of time, but they are no less necessary. Indeed, we have so greatly suffered from the excesses of the 1970s’ accounting standards, which we are currently revising, and we must define our goals within the framework of an international debate.

These goals will necessarily be hybrid. They must be defined in a democratic way, even though we currently suffer from a deficit of democracy on the international level. Standard setting seen as the art and practice of the general interest, within the framework of democratic standard setting, means that we have to imagine and establish new powers and checks-and-balances.

This is the ambitious project that we are currently trying to invent.

Conclusion

Creating a global standard is part of the search for the general interest, even in areas that are at first glance technical, such as accounting.

We are attempting to reach agreement on the concepts, and within this movement we are trying to collectively define the notion of the general interest, which is at the heart of standard setting.

More than creating global standards, we are building a power structure within the framework of a competition whose debates are sometimes violent, with its traitors and heroes, in which we are currently taking part.

In this perspective, we must find a consensual and credible system that inspires confidence and that is as stable as possible. At the Autorité des normes comptables, we have called this goal “seeking a center of gravity” which is different from neutrality, and “within which the French stakeholders feel comfortable.”

We are dedicating a great deal of energy to this goal because financial stability requires it. More than ever, we need a legal framework that will balance the risks and opportunities present in the economy and the financial world. It is probably not possible for this framework to be neutral if it is to neutralize risks and excesses.

Faced with the multiple and inevitable risks of instability, we need to create factors of security. The highest goal of accounting standards setting, like all the other parts of financial regulation, is to contribute to financial security. This is the general interest.