As the FT reports this morning:

KPMG has been fined £3m by the UK accounting regulator for audit failings at collapsed alcohol retailer Conviviality, the second set of sanctions against the firm in as many days. The Big Four firm failed to gather enough evidence to support its work or to apply sufficient professional scepticism during the audit of Conviviality's accounts for the year to April 2017, the Financial Reporting Council found.

It hardly need be said that this is not good for KPMG, or for the audit profession as a whole. Serial failings are becoming a significant issue. But why is that is the real question?

Is it that KPMG do not know how to do the job properly?

Or is it that they do not want to do the job properly?

Is it that they and the Financial Reporting Council have a different understanding of what the job is?

Might it be that in the face of criticism the Financial Reporting Council is toughening up its own act and is retrospectively applying standards to firms of which they were not aware at the time?

Or could it be that the job simply cannot be done under existing auditing and accounting rules, that are so unfit for purpose that no one can do an audit because a true and fair set of accounts could not be produced using them, at least in some cases?

Of these options I think all are a possibility, to be honest. I certainly do not exonerate the Financial Reporting Council from blame in this, and KPMG has very obviously had a particular issue to address, so common are its reported failings now.

But of all the options the last seems particularly significant. As my series on audit reform suggests, the current structure of accounting and auditing is quite literally all wrong. That these two tasks are intimately related is not recognised by current regulation. There is a pretence that auditors can be independent of the accounts that they audit, which is not true because the observer and the observed will always have an impact on each other.

There is an additional pretence that auditors can form an opinion upon a set of accounts without having to question whether the framework that is used to produce them is itself reliable, and that cannot be true.

Worse still, accounts are only produced for the use of shareholders, when there are very clearly other interested parties whose concerns need to be taken into account.

And, the purposes of accounting and audit are not aligned and should be. This is why Prof Adam Leaver at Sheffield University Management School and I have suggested that:

The purpose of accounting is to provide the stakeholders of a reporting entity with financial statements that include relevant, reliable and sufficient information which allow them to make informed decisions. 

As a consequence we suggest that:

The purpose of the audit of a public interest entity (PIE) is to firstly report on whether the financial statements on which the auditor offers an opinion deliver relevant, reliable and sufficient information to users of those statements and to secondly, where there is a shortcoming, remedy that shortcoming or, if it is not possible to do so, to report why that is and what its consequences are.

As is apparent these are related objectives but with the auditor being required to take an active rather than a passive role in ensuring that stakeholder needs are met.

Right now the value-added in audit is very far from clear. In the proposal that we make the role of the auditor is decidedly proactive and the added value is clear. Audit would, as a consequence, be both an attractive career option and one that would provide a very much more obvious role than a career in box-ticking.

KPMG is failing, but the task it is failing at is one so badly defined that systemic reform is required.