Those who take a contrary position say that no matter what ‘might be’ its bene- fits, fair value accounting raises substantive concerns – which they try to iden- tify through contradictory statements. For instance, because changes in economic environment and risk profile are better accounted for, the fair value’s net result would be to increase volatility of earning, assets and liabilities reflected in financial reporting.
The first half of this argument on the fact that both risk profile and changes in economic environment are better accounted forthrough fair value is very posi- tive in regard to IFRS. This is precisely what can make both financial accounting and management accounting much more:
● Relevant
● Accurate, and
● Informative to its user.
The second half of the contrary, that fair value increases volatility of earnings, assets, and liabilities, confuses the messenger with the wrong-doer. IFRS is sim- ply the messenger, who says that the type of instruments in the banking book and in the trading book are such that market changes greatly impact on:
● The entity’s asset values, gains, and losses, and
● On the volatility characterizing its assets and liabilities.
According to the same critics, this volatility in the profit and loss account ‘may even be misleading, with the result that unrealized profits are taxed’, a cost which may not be offset by the tax deductibility of unrealized losses. This is, of course, a fool’s argument, essentially amounted to the statement: Better to have losses in order not to pay taxes.
By contrast, a better documented contrary argument is the fact that determining the fair value of instruments like inventoried derivatives, which have no relevant market price, could be difficult, except by marking to model. As has already been discussed:
● Models involve assumptions leading to model risk, and
● Different models could give very different results for instruments with comparable risk features.
Furthermore, the reader should appreciate that the value resulting from a mod- elling procedure is only as good as the data used as input, and as sound as the algorithms of the model itself. Most banks lack model experience and, moreover, all too often they use too short a time perspective for estimation of model para- meters. This works against the accuracy of model results. There is also the fact that different entities can use:
● Different models, and
● Significantly different assumptions underpinning their valuation procedures.
When this happens, the resulting fair values, and their effects on the profit and loss account, may not be comparable across different financial institutions or other firms. And because value estimates always involve subjective judgment, both external auditors and supervisors find it challenging to verify whether fair values obtained through modelling are reliable.
True enough, what the last couple of paragraphs have stated is the negative side of the fair value approach. But ifthe critics had a better method to propose, all
they needed to do is to come forward with it, rather than suggest either of two totally unacceptable alternatives:
● Staying with the meaningless book value, or
● Espousing opaque proformas, the greatest cheat of them all.
Still another argument advanced by critics is that a deterioration in a bank’s credit rating would result in a reduction in value of its own bonds. This will decrease the fair value of its liabilities and, if the value of the assets were to remain unchanged, it would simultaneously result in an increase in sharehold- ers’ funds calculated as the difference between the fair value of the assets and liabilities.
This particular argument divides into two parts. The one is the effect of credit downgrading on the bank’s liabilities – which might have a perverse aftermath if banks see their credit downgrading as a way to improve their balance sheet. The counter argument to such a statement is that banks simply cannot afford to do so;
they have to maintain a high rating in order to:
● Attract funds, and
● Be considered worthy partners by correspondent banks and other parties.8 The likelihood of a misleading improvement in the entity’s solvency position resulting from deterioration of its own credit risk is a different case – both counter-intuitive and controversial from a supervisory viewpoint. It might allow banks to game the system in respect of regulatory capital requirements, ifsuper- visors don’t watch out for this case, and consequently apply the brakes as well as the penalties.
To better appreciate whether or not this particular issue might become a loop- hole, it is advantageous to recall that no matter what the supervisory regime or accounting system, banks always found incentives for cherry-picking. For instance, under the accounting framework, which (for European firms) had been valid till the end of December 2004, the economic value of instruments:
● Was only recognized at the moment they were actually realized
● Therefore, a bank might have had an incentive to realize certain transac- tions purely to boost its accounting profit.
A frequent practice in this regard has been that assets which show substantial latent surplus values, such as hidden reserves, are sold to offset poor results for
core business activities. Both new regulatory regimes, like Basel II, and new accounting rules, IFRS being an example, are designed to stop institutions and other entities from manipulating their financial results.
Another sound test of a new accounting system is how entities implementing it would react to economic shocks. Take as an example an unexpected change in interest rates (leaving for a moment the derivatives portfolio out of considera- tion). In principle, an interest rate change will have a different impact on the accounting value of items in the banking book under accruals and fair value approaches.
● An interest rate increase would result in a lower economic value of fixed value instruments.
● By contrast, a reduction in official interest rate would result in an increase of the economic value of fixed interest investments.
Under fair value accounting, value changes will be recognized in the financial statements. This is not true of accruals accounting, because value changes result- ing from interest rate volatility will not affect the value of loans.
For securities in the banking book, when the lower of cost or market valuation is used, the recognition will only occur in the case of an interest rate rise. Even this valuation, however, will not recognize the latent value increases resulting from an interest rate decrease, which is not the case with IFRS.
In practically all entities today, and most particularly in the case of big banks with trillions in inventoried derivatives investments (in notional amounts), for- getting about the impact of derivatives on balance sheet and P&L is an oversim- plification tantamount to a lack of management accountability (see section 8).
The impact of derivatives gains and losses is so great that it can tilt the balance in terms of:
● Creditworthiness, and
● Financial staying power.
In more classical investments like equities and bonds, price declines could be absorbed, up to a point, by hidden reserves. But there are no hidden reserves big enough to cover huge derivatives losses as Barings Bank and tens of other docu- ments. Therefore, it is right that under IFRS all price declines must be fully reflected in the profit and loss account – leaving no way for cherry-picking.
To better appreciate this message, let us apply a similar approach to the credit institution’s deterioration in credit quality. In principle, deterioration in credit- worthiness of a financial asset, such as a loan or a bond, will be reflected in lower than expected cash flow.
● Ifthe fair value of the instrument were to be calculated by discounting its expected cash flows,
● Then, its fair value would decline in parallel with the credit downgrading.
Prior to IFRS, in most jurisdictions the value of an asset was (usually) adjusted through creation of a specific provision, when that asset has been non-perform- ing. A certain event reflecting deterioration in quality, like delay in interest pay- ments or outright default, has to occur prior to such an adjustment. As a consequence:
● If provisioning decisions are perfectly forward-looking and reflecting likely changes in expected cash flows,
● Then, the accounting effects of accruals and fair value accounting, on credit risk, will be more or less identical.
Notice, however, that forward-looking provisioning is handicapped by cost- based accounting and tax regulations, which tend to narrowly define impairment and non-performing loans. Typically, the loan loss provisions have largely been backward-looking – a default which has not been corrected through Basel II because of abandoning the expected losses (EL) formula. Another current prac- tice which has to change is that:
● As a rule credit ratings and probabilities of default are estimated in a point- in-time
● This limits vision to the short time horizon of one year, which is totally inadequate.
Only the better-managed banks use a longer time horizon for their credit risk assessment, taking into account expected average performance of a borrower over an economic cycle. In terms of management accounting, if not also for financial accounting reasons, the longer horizon is by far the better solution. Short-time estimates need to be revised most frequently, and can lead to volatility in state- ments greater than what might result from IFRS. Surely even a blind man could see this – yet IFRS critics don’t.