Presentation

This article contains the proceedings of the open debate that followed the plenary panel on ‘Accounting Regulation and the Public Good’ at the international workshop on ‘Which Accounting Regulation for Europe’s Economy and Society?’ organised under the auspices of the European Parliament in Strasbourg, on 20 May 2015.

Identified participants comprise Hélène Rainelli (University of Strasbourg), Karthik Ramanna (University of Oxford), Shyam Sunder (Yale University), Jean-Claude Thierry (MESA).

Raised questions were: (i) How robust are the data that you are using?; (ii) Can various constituencies’ interests be served by your proposed system?

How robust are the data that you are using?

Raised by Jean-Claude Thierry, MESA (Maison de l’Europe Strasbourg-Alsace)

You have been speaking about accounting: Accounting is based on information, databases and their curation. I am asking to the last orators: The main problem is database curation to derive at results in evaluation of a process. How robust are the data that you are using in your different processes to discuss and derive at the conclusion of all what you have been speaking about this morning?

Karthik Ramanna (University of Oxford)

If I understand correctly, the question is how robust are the data underlying the conclusions, correct? I think this is a very important question and one that goes to the very heart of the challenge we have when it comes to setting the rules of the game and the notion of manipulating the definition of profit in the spirit of increasing profit. That could not possibly be the spirit in which we conceived a system of market capitalism to serve society. Capitalism is intended to be a process through which societies achieve higher goals, higher purposes, certain normative purposes such as individual freedom, democratic values, etc. And we enable corporations to go about generating profits, or creating profits, or extracting profits, in the expectation that this will create value and prosperity across the board. So if you go about setting up a system to define the rules of generating profit, or the rules of measuring profit, that is in itself fundamentally flawed or compromised, then the legitimacy of the whole enterprise comes into play.

In order to be able to draw the conclusions that I draw, and to use the data, I have to start from what I see as a conceptual basis, a sort of conceptual framework. I do not end up using the conceptual frameworks of standard setters because that would result in a logical circularity. If you are evaluating the standard-setting process and you use the same basis that they offer for their own evaluation, there is a problem. Particularly if you look at the evolution of the conceptual framework for standard setting – at least in US GAAP – that conceptual framework has shifted over time. In the 1980s, when US GAAP was predominantly based on historical cost estimates, there was a different conceptual framework, which was partly defined by what was known as CON-2. By the time we get to the 2000s, a substantial fraction of US GAAP is characterised by fair value accounting, and it turns out that fair value accounting is inconsistent at times with the conceptual framework that the FASB had offered in the 1980s. So they changed the conceptual framework and in 2010 they introduced a new conceptual framework called CON-8. So if you use the conceptual framework that standard setters offer to measure the performance of the standard-setting enterprise and the performance of accounting rules, you end up with a logical circularity.

To get around this problem what I do for my purposes is to use the conceptual framework that emerges from the economic theory of accounting. There has been, at least since the post-war neoclassical consensus in economics, a conceptual framework for the role of information, the role of audited financial information in particular, in a capital market economy. Now a number of academics over the last fifty years including Mr Shyam Sunder have contributed to this body of knowledge, and I use this body of knowledge as a basis to evaluate outcomes of the standard-setting process. The economic theory of accounting is something that characterises certain principles such as verifiability of accounting estimates, or conservatism in accounting practice, or the relevance of accounting estimates, as being fundamental characteristics of accounting. You cannot assess the performance of any system without a benchmark; you have to make a normative judgment as to which benchmark you will choose. For reasons of logical consistency, I have chosen not to use the benchmarks offered by the standard setters and chosen to use instead the benchmark that emerges from economic theory. Recognise full well that is a normative choice that I have made. In some sense because the question you raise is one that is ultimately a paradox and does require one to sort of step out of the system and make a normative choice, and that is the one I have made.

Shyam Sunder (Yale University)

The fundamental difference between natural and social sciences is that the former analyse games against nature while the latter, including accounting, can be represented as games among people. Nature does not change its strategy in response to what humans do; it remains fixed. In social systems – such as in business and accounting – the environment is fluid and interactive. A takes one action, and B reacts to it; then A reacts to B’s reaction, and so on. The outcomes of such systems are more difficult to analyse and predict. For this reason, an analogy between science and social sciences is of limited value.

Can various constituencies’ interests be served by your proposed system?

Raised by Hélène Rainelli, Professor of Finance at University of Strasbourg

I am interested in financial regulation, and my question is more to the last of the speakers about your sort of innovative proposal to have a sort of competition between regimes. You are very optimistic about avoiding the race to the bottom – and I hope you are right – but the question is: If you look at Europe’s tax systems, companies will tend to choose what is more interesting for them as you said before, and maybe it would be the same for financial regulation. Because you said investors would provide, that the best system is eventually chosen. But maybe some issues are not the issue of the investor. There are many constituencies in the right accounting reform. The one of the other speakers spoke about a sort of a tension between transparency and stability: Too much transparency might entail more instability. This problem would be the same: Investors might choose the best accounting system as regards to transparency because they are interested in transparency. But what is not very certain, even if competition goes well and this does not put us to the bottom that other constituencies’ interest would be served by your systems.

Shyam Sunder (Yale University)

Thank you for making this important point. Regulatory competition, like other forms of competition, calls for regulatory oversight of its own. As you rightly suggest, managers, investors and auditors can be expected to push financial reporting in directions that serve their own respective interests. If any one of the interest groups is given the power to shape the financial reports, we should not expect the outcome to serve the interests of society as a whole. That end is better achieved in two layers. The first of these layers is the push and pull, and compromises, among the conflicting interest groups through which they settle down on the form and content of financial reports that they each find acceptable, but hardly optimal, from their respective point of view. The second layer is the regulatory oversight over the process of competition and compromise. This is not different from the market for cars, for example. Manufacturers and customers through dynamic competitive interaction among their interests end up determining what kind of cars are made, offered, and bought. Yet, we need a regulatory layer on top of this competition to set safety and environmental standards. But this regulatory layer is ‘light touch’ in the sense that it does not try to set the size, colour, horsepower and any of the thousands of other features of cars. In accounting, I believe the balance between the roles of the competitive layer and the regulatory layer is wrong, with the regulation being ‘heavy handed’ instead of ‘light touched’ by getting into far too many details – akin to setting the colour and size of cars.

No regulatory competition can exist with zero regulation. We still need regulatory oversight, as we do with respect to competing universities and stock exchanges which have to conform to light touch regulation in their domains. All regulatory competition has to be within certain bounds. Call for competition does not eliminate that oversight, it just allows a lot more room for innovation and variation.