Accountants tend to think about liabilities mainly in terms of their settlement. If settlement seems unlikely to require an outflow of economic benefits, it is often concluded that no liability exists. And if a liability does exist, the settlement amount is often used to quantify it.

This often seems entirely reasonable, but perhaps it is not the whole story. For example, suppose a supplier receives £100 as full payment for goods that he can supply for £60. If the liability is seen only in terms of its settlement, it would appear that it is only £60, so presumably a profit of £40 would be recognised on receipt. This would be a radical departure both from present practice and the proposals of the ASB’s Discussion Paper on Revenue Recognition: usually revenue and profit is recognised when the goods are delivered, and not before.

One solution to this puzzle may be to think of liabilities as the mirror images of assets. Businesses purchase assets because they believe that they will contribute returns that are greater than their purchase price: accounting however, records the purchase at cost, and does not reflect the expected future returns until they are earned. The same rationale suggests that it is right to record liabilities at the amount received for taking them on.

In the IASB’s Framework for the Preparation and Presentation of Financial Statements and corresponding documents issued by other standard-setters, the definition of a liability is linked directly to the occurrence of a future outflow of economic benefits. The example of payment received in advance suggests that this may be mistaken. A flaw in the definition of one of the most basic elements of financial reporting might be expected to have important consequences.

Arguably, more satisfactory answers to many accounting issues would be derived from a clear distinction between the existence of an obligation to transfer benefits and its expected settlement. Questions that might be illuminated include those relating to the distinction between debt and equity and derecognition issues—as well as several other questions concerning revenue recognition. These ideas are explored in a paper presented by Andrew Lennard, ASB Director of Operations, at an academic conference in early July.

A further product of this line of thought is that it suggests that entry values, rather than exit values, are often the most relevant measures to use in accounting, as is implied by the value to the business model.

Those favouring exit values tend to associate entry values with attitudes based on historical cost. They say that historical cost accounting cannot do a good job in reporting on the modern business environment. They favour widespread use of exit values (selling prices) as an alternative.

However, the essential failure of historical cost is not, perhaps, that it is a cost based measure, but rather that it is historical, and hence of limited relevance at the balance sheet date. The same does not apply to current values reflecting current replacement costs. And using entry values preserves the perspective of reporting gains from selling assets or settling liabilities only when the transaction occurs, and not before.