Accounting standards. A question of measurement

The continuing rumble over international accounting standards

CHANGE is never easy; and the bigger the change, the more it hurts. So it is no surprise that the European Union's ambitious project to require its 7,000 listed companies to adopt international accounting rules by January 1st 2005 has been a difficult exercise.

Earlier this month, all seemed done and dusted, when the European Commission endorsed the last of the 41 accounting rules issued by the International Accounting Standards Board (IASB). This highly contentious rule, IAS39, deals with accounting for financial instruments, such as options and derivatives. It had been the subject of a row between the IASB and the EU for months, and was accepted only after the commission won over the most vocal critics, including the European Central Bank and much of Europe's financial industry, by deleting passages of IAS39 to create a looser version of the rule.

However, the commission's compromise is already fraying. Britain's Accounting Standards Board has publicly urged British companies to ignore the commission's version of IAS39 and use the IASB's original rule. Several international banks, including HSBC and UBS, have said they will use the IASB's version. Some companies that are listed in America, where accounting rules for derivatives are stricter than IAS39, could follow suit if they believe that capital markets will reward those who adopt the toughest standards.

There is a risk that the old patchwork of national accounting rules will be replaced not by a pan-European regime but by another mishmash. In an attempt to avoid this, and under pressure from the commission to amend IAS39 quickly, the IASB is reviewing its derivatives rule - but the chances of a new, generally acceptable version by January are slim.

Different banks will probably end up using different versions of the rule. One drawback of this is the perception that accounting standards will be as much a product of politics as of financial measurement. The commission is not supposed to be a standard-setting body. Yet its “hollowing out” of IAS39, after intense lobbying by European insurers and banks, especially French banks, is “tantamount to writing a new standard”, says the chief executive of one large European bank.

The repercussions could extend far beyond Europe. Almost 100 countries are due to use IASB's accounting rules by next year. Still more, such as China, are expected to follow. Some worry that the precedent set by Europe might cause others to adopt the rules piecemeal. It might also complicate the IASB's long-running attempt to close the gap between its rules and those of America's standard-setter, the Financial Accounting Standards Board (FASB). That said, lobbying and political interference in accounting standards is scarcely unknown in America either: witness the technology industry's ability to blunt FASB's efforts to have employees' share options treated as an expense.

There may be unintended technical consequences too. Jeannot Blanchet of Morgan Stanley notes that the IASB designed its rules to work as a unit. So changing one standard in isolation, as the European Commission has done, could have unforeseen effects elsewhere.

These fears are not academic. Danish regulators gave warning recently that the commission's compromise could destabilise their country's financial system, owing to the “artificial volatility” that the rule would create in Denmark's mortgage market. Some insurance companies, too, worry that the commission's fudge could make their accounts more volatile.

At the root of the arguments over IAS39 is an important question of principle: how should assets and liabilities be measured? In the past, most national accounting standards used historic cost. For financial instruments, such as stock options and many other derivatives, this was often (conveniently) zero, or close to it.

This mattered less when the use of financial instruments was limited and markets for them were small. But both have exploded in recent years. Now companies from banks to retailers use derivatives and other financial instruments to hedge currency, commodity-price and other risks. According to the Bank for International Settlements, the value of over-the-counter derivatives worldwide is almost $8 trillion. But under old accounting rules, the value of derivatives did not show up on many European balance sheets at all - a remarkable omission, given the extent of some companies' exposure.

Enter IAS39, which requires companies to use “fair value ” - -what a willing buyer would pay for an asset or liability in the market - rather than historic cost when valuing financial instruments. The growth of large, liquid markets in many of these makes their prices easier to define than in the past. Unlike historic cost, fair value fluctuates - so balance-sheet items and profits fluctuate too. Banks and insurers, which use piles of derivatives, are less than pleased.

The IASB argues that IAS39, and other fair-value accounting standards, such as IAS32, which mainly affects insurers, do not create volatility: they merely disclose it. Investors can choose to read the volatility as they see fit. Critics counter that this volatility is artificial, having little to do with companies' long-term health. Far from enlightening investors, it will confuse them. Moreover, they say, measuring “fair value” is not as easy as it seems. Not every financial instrument has a deep and liquid market, and models are inexact.

The crux of the problem is that the systems of accounting that have developed hitherto are mixed bags of both historic costs, which do not change with time, and fair values, which can vary a great deal. It is inevitable that accounts will reflect changes in the value of some things but not others, making companies' numbers jump around more than they perhaps should. Eliminating one cause of volatility in accounts, as the commission did this month, might merely create more volatility elsewhere, as the unhappy Danes have discovered. In accounting, as in many things, there is no magic bullet.

• Primary Source