20 June 2012
On 20 June 2012 Hans Hoogervorst, Chairman of the IASB, addressed the International Association for Accounting Education & Research (IAAER) conference in Amsterdam.
The imprecise world of accounting
Accounting should be the most straightforward of topics for policymakers to deal with. Accounting is mainly about describing the past—to reflect faithfully what has already happened. This should be dull business, better left to 'bean-counters'. Surely counting beans cannot cause too many problems?
Yet, over the years, many securities regulators have told me of their surprise upon finding out that accounting policy is one of the most difficult and controversial topics to deal with. It is the same around the world. Just ask the Japanese FSA, the US SEC or the European Commission.
So, why is it that accounting is the source of such heated debates?
Of course, there are many reasons why this is the case. Sir David Tweedie, my predecessor as Chairman of the IASB, used to say that it was the job of accounting to keep capitalism honest. It is no wonder that accounting standard setters come under so much pressure! Some business models can thrive off a lack of transparency. Just think of the pre-crisis Special Purpose Vehicles in the banking industry.
There is second reason why accounting can be so controversial, and that is the inescapable judgement and subjectivity of accounting methods. Put simply, there is a lot to disagree about.
When I became Chairman of the IASB in July last year, I knew enough about accounting to know that I was not entering a world that was governed by the iron rules of science. I knew that Accounting has the same problems as its sibling Economics: you need maths to exercise it, but you should not count on outcomes with mathematical precision. In short, I did not have naive expectations of accounting. Or so I thought.
One year later, however, now that I am well ahead on a steep learning curve, I must admit that I may have been a bit naïve after all. Let me give you a couple of examples that served to open my eyes.
First of all, I was struck by the multitude of measurement techniques that both IFRSs and US GAAP prescribe, from historic cost, through value-in-use, to fair value and many shades in between. In all, our standards employ about 20 variants based on historic cost or current value. Because the differences between these techniques are often small, the significance of this apparently large number should not be over dramatised.
Still, the multitude of measurement techniques indicates that accounting standard-setters often struggle to find a clear answer to the question of how an asset or liability should be valued.
It is also remarkable that our standards can cause one and the same asset to have two different measurement outcomes, depending on the business model according to which it is held. For example, a debt security has to be measured at market value when it is held for trading purposes, but it is reported at historic cost if it is held to maturity. In this case, the business model approach certainly provides a plausible answer. Still, some may find it counterintuitive that a government bond that is held to maturity would be valued at a higher price than the same bond held in a trading portfolio, where it may be subject to a discount. In the exact sciences, such a dual outcome would certainly not be acceptable.
One of the biggest measurement dilemmas relates to intangible assets. We know that they are there. While the value of Facebook's tangible assets is relatively limited, its business concept is immensely valuable (although 25% less immense than a month ago).
Likewise, the money-making potential of pharmaceutical patents is often quite substantial. However, both types of intangible assets go unrecorded (or under-recorded) on the balance sheet. Under strict conditions, IAS 38 Intangible Assets allows for limited capitalisation of Development expenditures, but we know the standard is rudimentary because it is based on historical cost, which may not reflect the true value of the intangible asset.
The fact is that it is simply very difficult to identify or measure intangible assets. High market-to-book ratios may provide indications of their existence and value. However, after the excesses of the dot.com bubble, there is understandable reluctance to record them on the balance sheet.
Although our accounting standards do not permit the recognition of internally generated goodwill, our standards do require companies to record the premium they pay in a business acquisition as goodwill. This goodwill is a mix of many things, including the internally generated goodwill of the acquired company and the synergy that is expected from the business combination. Most elements of goodwill are highly uncertain and subjective and they often turn out to be illusory.
The acquired goodwill is subsequently subject to an annual impairment test. In practice, these impairment tests do not always seem to be done with sufficient rigour. Often, share prices reflect the impairment before the company records it on the balance sheet. In other words, the impairment test comes too late. All in all, it might be a good idea if we took another look at goodwill in the context of the post-implementation review of IFRS 3 Business Combinations.
It is not only the balance sheet that is fraught with imprecision and uncertainty. We also have a problem defining what income is and how to measure it. We report three main components of income: the traditional profit or loss or net income, other comprehensive income and total comprehensive income. Total comprehensive income is the easy part; it is simply the sum of net income and other comprehensive income, or 'OCI'. Not too many people seem to be paying attention to it, even if they should.
The distinction between net income and OCI, however, lacks a well-defined foundation. While the P&L is the traditional performance indicator on which many remuneration and dividend schemes are based, the meaning of OCI is unclear. It started as a vehicle to keep certain effects of foreign currency translation outside net income and gradually developed into a parking space for 'unwanted' fluctuations in the balance sheet. There is a vague notion that OCI serves for recording unrealised gains or losses, but a clear definition of its purpose and meaning is lacking.
But that does not make OCI meaningless. Especially for financial institutions with large balance sheets, OCI can contain very important information. It can give indications of the quality of the balance sheet. It is very important for investors to know what gains or losses are 'sitting' in the balance sheet, even if they have not been realised.
In the future, OCI will most certainly be an important source of information about insurance contracts. A couple of weeks ago, both the FASB and the IASB proposed that changes in the insurance liability due to fluctuations in the discount rate would be reported in OCI. Many of our constituents requested us to do so.
Both preparers and users wanted to prevent underwriting results being snowed under by balance sheet fluctuations. As a result, OCI will become bigger and will contain meaningful information, such as indications of duration mismatches between assets and liabilities.
This decision for the use of OCI was not easy to make. Our fellow board member Stephen Cooper showed us in a razor-sharp analysis that in this presentation, both Net Income and OCI—if seen in isolation—might give confusing information. We will try to tackle some of these problems with presentational improvements. But it is also clear that a full picture of an insurer's performance can only be gained by considering all components of total comprehensive income. We will point this out explicitly in the Basis for Conclusions of the new standard.
More fundamentally, we will look at the distinction between net income and OCI during the upcoming revision of the Conceptual Framework. All of our constituents have asked us to provide a firm theoretical underpinning for the meaning of OCI and we will endeavour to do so. For now, while we may not always know how important OCI exactly is, we can be sure that net income is not a very precise performance indicator either. Both need to be used with judgement, especially in the financial industry.
What is the reason for all this ambiguity and lack of precision in accounting? Well, to a great extent it is simply the nature of the beast. Valuation is as much of an art as a science and we are fully aware of that. Our Conceptual Framework says: "General purpose financial reports are not designed to show the value of a reporting entity; but they provide information to help users to estimate the value of the reporting entity." Value is ultimately in the eye of the beholder. There is often not a clear cut answer to the question as to which measurement technique is most appropriate to capture it.
These comments about the imperfections of accounting should not be interpreted as a sign of wary relativism about the significance of our standards. Quite the opposite: I am deeply convinced that our accounting standards are an essential ingredient of trust in our market economy. In an economic system in which so many parties are working with other people's money, high quality accounting standards that provide transparency to the market are of paramount importance.
IFRS as a global standard has had a tremendously beneficial impact for global investors, who lacked all comparability in the pre-IFRS days. Several academic studies have shown that the introduction of IFRS has contributed to lowering the cost of capital.
Moreover, financial reporting does not need to be mathematically exact to be useful. It is a tool to help investors on their way. Warren Buffett is known to use financial reports as a rough-and-ready checklist: more than five or six questions marks are enough for him to decide against making an investment.
One has only to look at the insurance industry to see how essential proper accounting standards are. Currently, IFRS does not have a full-blown standard for insurance. As a result, financial reporting by the industry is riddled with non GAAP measures and there is a serious lack of comparability. Because the industry's reporting lacks the underlying rigour of uniform accounting, investors demand a higher price for capital to make up for the lack of transparency.
Public sector accounting also demonstrates the primitive anarchy that results without the discipline and transparency that good financial reporting provides. While the IPSASB has created good standards for the public sector, based on IFRS, they are used only haphazardly. Around the world, governments give very incomplete information about the huge, unfunded social security liabilities they have incurred. Many executives in the private sector would end up in jail if they reported like Ministers of Finance, and rightly so.
So there can be no question about the relevance and importance of our standards. As the convergence programme comes to a close and the IASB is ready to take on a new agenda, we should concentrate on further improving the quality of our standards. Although we know that some of the imprecision and ambiguity I mentioned before is inevitable, it is our job to push back the grey areas in accounting as far as possible.
So how should we go about it? I believe we should be guided by the following three terms: Principles, Pragmatism and Persistence.
Principles: for the very reason that accounting is not an exact science, principle-based standard setting remains the right way forward. If the use of judgement is inescapable, it should be guided by clear principles and not by detailed, pseudo-exact rules.
We will strengthen our basic principles by finishing the review of the Conceptual Framework and by tackling thorny issues such as measurement, performance indicators, OCI and recycling.
While I am not so naïve to think that a new Conceptual Framework will solve all our problems, I think it can serve to give us firmer ground under our feet. Even if precise answers are not always available, a completed Conceptual Framework should give us more guidance on the recognition of assets and liabilities, measurement techniques and performance indicators.
Pragmatism. If we know that there is not always a precise answer to every question, our work needs to be grounded in pragmatism and common sense. As Keynes said, it is better to be roughly right than to be precisely wrong. We should avoid trying to get companies to achieve precision without accuracy.
Pragmatism also means we need to look very carefully at any possible undesirable use of our standards. Whenever we are confronted with a high degree of uncertainty, we should act with great caution. I just gave the example of intangible assets. We know they are there, but measurement is a big problem. If our standards were to provide too much room for recognition of intangible assets, the potential for mistakes or abuse would be immense.
In such circumstances, it is better for our standards to require more qualitative reporting than pseudo-exact quantitative reporting.
By the way, people always tell us we should not set our standards from an anti-abuse perspective. I think that is nonsense. If we see ample scope for abuse in a standard, we had better do something about it. There are sufficient temptations and incentives for creative accounting as it is.
Pragmatism is important, but it should not be confused with opportunism. That is why we need Persistence, too. In the face of the pressures we are continually facing, persistence is an important quality for standard setters. Accounting standard setting should be sensitive to legitimate business concerns, but should also be firm and independent in the face of special interests. Many times, doomsayers have predicted their business would come to an end as a result of our standards. Just as often, the industry in question miraculously seemed able to survive our rules very well indeed. We always need to listen, but we have to take decisions, too.
For the IASB to persist on a steady course, it would be hugely beneficial if the investor's view was heard more loudly and clearly than currently is the case. While investors are our prime audience, their voice is too often drowned out by vociferous business interests. In the coming years, we are determined to further invest our relationship with investors, to ensure we get more balanced feedback on our proposals than currently is the case. We are especially
interested in strengthening our relations with what I would like to call our "end-users".
With this term I refer to true investors, in the sense that they actually own assets, such as institutional investors. The support of the investor community will make it easier for us to stay our course.
So it is with principles, pragmatism and persistence that the IASB will take on its new agenda. We should use the coming years to strengthen the underlying principles of our work. We should improve the significance of the quantitative outcome of our standards where possible. Where this is impossible, we should make this clear and put more emphasis on qualitative information.
This is all much easier said than done, but my Board looks forward to taking up this challenge with our highly motivated staff. Whatever the coming years may bring: a period of calm they will not be.