Auditors: Market concentration and their role - Economic Affairs Committee Contents

Memorandum by Mr Timothy Bush (ADT 10)

  I have pleasure in submitting written evidence to the Committee.

  I was the chair of the Institute of Chartered Accountants in England and Wales' (ICAEW) Competition and Choice working group set up under the then Department of Trade and Industry, which then evolved into the Financial Reporting Council's (FRC) project on the same issue. I am a past member of the Governing Council of the ICAEW and an investor, an FSA registered fund manager. I am the Investment Management Association nominated representative on the Urgent Issues Task Force of the Accounting Standards Board (ASB). I have analysed banks in particular since 2006.

  I would also be prepared to give oral evidence. I am writing solely in a personal capacity with no paid interest or other conflict of interest with the subject matter.

  As the Committee is seeking evidence about the role of the auditor and risk, my evidence sets out and explains the context and purpose of the accounting and auditing provisions of the Companies Act. The Companies Act 2006 (previously the 1985 Act) has Section 837 which very simply defines the output expected of audited accounts as:

    "the [audited] accounts must have been properly prepared, or only be defective in ways immaterial in determining whether the distribution [ie a dividend] is in breach of this Part of the Act. (Source: Palmer's guide to Company Law 2006).

  The Act is seeking that accounts are sufficiently reliable so that companies will not pay dividends out of capital, such as when they are actually making losses. This objective is a stress test a sense check. However, audits have appeared to fail (clean opinions were given on banks about to collapse) because:

    — the auditing standards give auditors the objective of auditing by the "accounting standards framework". That falls short of setting out the full implications of S837

    — when the "accounting framework" is IFRS (which for banks and listed companies has been the case since 2005), IFRS accounts cannot deliver that objective, because IFRS in parts positively overstates financial assets and profits. It has the wrong risk model

    — overstating assets and false profits is almost bound to lead to companies, but particularly banks, getting out of control and misleading themselves as well misleading outsiders

  This has happened because IFRS has wholly replaced the Accounting Rules that were in the Companies Act itself and had had proper legislative scrutiny. Section 837 is still intact and it is still the law but there has been no functional mechanism to deliver it.

  The inconsistency between what accounting standards (IFRS) and auditing standards set out and what the Companies Act really requires was flagged by institutional investors in 2005 upon the launch of IFRS. The accountancy regulators dismissed those concerns. But since then IFRS has merely taken the self-referential compliance model already set by the auditing standards that bit further. So similar flaws have passed up the chain via IFRS to the directors and their business control systems as well.

  Banks traded, priced credit and paid dividends when they weren't really making the "profits" that they were showing and thought they were making. The banking crisis is largely due to faulty numbers. Post Enron accounting reforms have not worked as the medicine was the disease.

  It is therefore encouraging that Parliament is looking at this matter in some detail.


  S1.1  This evidence covers two distinct areas; the role of auditors and the relevance of accounting/auditing standards in that role (to cover questions 5-9) and audit firm concentration (to cover questions 1-4). It sets out a case which explains why the audits of some banks have come under question from politicians due to their apparent failure to identify the true state of those banks that were on the brink of collapse. Or put another way, why auditors were the dogs that did not bark.

  S1.2  Some banks were following capital destructive business models, essentially lending at below true cost, and in some cases having no capital at all. They were showing an accounting illusion of capital and profit, but they received unqualified audit opinions on their accounts, and paid dividends.

  S1.3  Given that the accounting provisions of the Companies Act were established in 1879 explicitly to serve a banking solvency function and given that Parliament was assured in 2006 that the function of auditors of accounts had not changed,[1] something has come adrift for questions on their role to be asked at all. In reading evidence recently given to the House of Commons Treasury Committee I conclude that neither the accountants, nor their regulators conveyed to that committee the true purpose of the statutory auditor. This evidence may explain why.

  S1.4  Because of the way the law relating to auditors is constructed, the audits of banks should identify true solvency. The position that regulators should instead is false. Regulators did not have the resources or global proximity to the businesses, auditors did. The contractual purpose of the Companies Act auditor is to avoid management operating under or presenting an illusion, business is more dynamic than regulation.

  S1.5  However the accounting standards introduced in 2005 can mask true solvency, and do not even purport to seek to show that. It is therefore my conclusion that there is an absolute incompatibility between Company Law and key parts of IFRS (the International Financial Reporting Standards of the IASB) that for banks means an absolute danger.

  S1.6  IAS 39 (the accounting standard that deals with most of a bank's balance sheet and profits) conflicts with the Accounting Rules of the Companies Act. It is therefore not surprising that the delivery of IFRS accounts and the audit thereof will be defective in comparison to what the Companies Act intends of both accounts and audits. Rather than functioning as a reliable basis for controlling and monitoring banks, inside and out, IFRS has had the dual effect of helping some of them to get out of control and obscuring that. The incompatibility is not merely against the intent and spirit of the law but is contrary to specific statutory provisions of the Companies Act itself.

  S1.7  Because the UK's accounting standards board (ASB) had set generally good standards whilst sitting below the Companies Act's Statutory Accounting Rules there may have been a general presumption that IFRS was merely an internationalisation of what the ASB had done. But, what the ASB had done was properly restricted by the Statutory Accounting Rules set down by Parliament.

  S1.8  However, IFRS did not merely replace UK Accounting Standards, it also replaced the Accounting Rules contained within the Companies Act that were consistent with the purpose of the law and drafted with legislative levels of scrutiny. IFRS has quite simply frustrated the delivery of the accounting model and audit that Parliament intended because IFRS is set without equivalent legislative scrutiny by a body with very different objectives. IFRS seeks to value companies at a point in time ("decision usefulness") rather than ensure their safety going forwards (stewardship). IFRS also frustrates Section 386 which requires that companies by their books know where they are with reasonable accuracy at any time. That is difficult if the accounting standards do not achieve it.

  S1.9  The way that the law is still structured, Sections 830-837 (Capital Maintenance), are still paramount outcomes, but complying with IFRS may well break that part of the law, despite being the expected method of delivery. Instead of a health check of a business, the model that part of false positives. Very dangerous indeed.


  E1.1  I set out when and how I first concluded that there were severe technical problems in accounting and then auditing that have contributed to the banking crisis to the extent of being a primal cause.

  E1.2  In early 2008, I produced a brief analysis of problems with banks and their accounts, and having sent that to HM Treasury I was immediately asked to meet officials in July 2008, where I ran a model (similar to that the FSA now has) past members of the Financial Crisis Team. I gave them my view that I thought that accounting standards were not only covering up problems in banks but were causative of them. Why cause? Because people plan and budget—and then monitor that—by numerical outcomes. If accounting standards produce a false profit as a numerical outcome, things will go awry right from the start.

  E1.3  That presentation was summarised by an official as the first time that anyone had "managed to explain [to them] what no one else had managed to explain". At that stage only Northern Rock had collapsed, although Bradford & Bingley had had a failed rights issue. Investors were firstly blamed for being unsupportive of banks and there was a prevailing view that the pre-emption rights issue period was too long.[2] In reality the market was suspecting that other banks too might be bust—and also need to be nationalised—but the numbers did not show it. Attention had been overly focussed on "liquidity" (asset/liability maturity profile) rather than capital solvency/adequacy (the amount by which assets exceed liabilities). All banks have a liquidity problem, that is their public benefit, they borrow short-term to lend long term. Liquidity matters most when people will not lend to what they suspect might be a bust bank (liabilities exceed assets). Banks were operating under and presenting an illusion.

  E1.4  The official commented that speaking to long only investors (ie not short sellers) tended to show a convergence of interest with HM Government's interest. I replied that it was because long investors also had diverse economic interests rather than narrow vested interests. I also commented that I thought that before long the government would also be an equity investor in more banks than merely Northern Rock. The team was impressive and quick on the uptake.

  E1.5  My analysis was then, and still is, that problems have arisen quite simply due to changing accounting standards in January 2005. Checks and balances inherent in the accounting model replaced in 2005 were taken out and not compensated for. These changes not only affected regulated banks (ie deposit taking institutions) making loan advances, but also unregulated non-banks undertaking similar activity, such as Cattles plc (not regulated as a deposit taking bank—but making loans to customers by a doorstep collection model). It is difficult to blame faulty regulation where there wasn't any (other than consumer protection regulation) in the case of Cattles which had existed since 1927.

  E1.6  It is positive that of all regulators working in this area Lord Turner of the FSA has best articulated the issues that others have not.[3] The accountancy regulators still don't give a coherent explanation for things. They are embarrassed. Furthermore, standard setters resemble lobbyists, selling their standards rather than writing them properly. They don't even do impact assessments.


  E1.7  The statutory requirement of auditing and accounts flows from the 1879 Companies Act, and followed the collapse of the City of Glasgow Bank in 1878, which had been a fraud. Before that when their commission was not compulsory, matters followed common law.

  E1.8  The law is constructed such that the law is framing a private contract that is albeit compulsory. The parties are the company and the auditors. This differs from regulation, which is not contractual, the parties then are the regulated party and the regulator. An analogy is the legal requirement for a driver to have third party insurance. Whilst the government demands the contract, it is not a party to the contract. The audit though, is not insurance, but assurance, to detect and deal with risk and negligence, including fraud.

  E1.9  The framework of the 1879 law is intact, and is now the 2006 Companies Act (previously the Companies Act 1985 and preceding acts, and is set out too in case law).[4] Clauses particularly relevant to accounts and the audit and its purpose are:

    1. The accounts that are published comply with the accounting framework (IAS accounts or Companies Act accounts)[5] and give a "true and fair view" (ie a sufficient output). Section 393 (and the auditor opinion thereon, Section 495(3)(a) and 495(b).

    2. The books of account that are kept privately at all times are suitable to inform the directors of where the business is at all times. Section 386 "adequate accounting records" (and the auditor opinion thereon, Section 498).

    3. The firm link between the statutory audited public accounts and the legality of dividends declared off of those accounts. Sections 830, 836, 837. Sections 837(4) is wholly about the auditor duty in that regard. He must always conclude whether the accounts are suitable to be used for declaring a dividend even if his opinion is qualified. The statutory duty is so positive so as to require the auditor to report to shareholders on that matter over the heads of the directors if necessary S837(4).

    4 Section 532 prohibits the auditor from contracting with the company to limit liability. This appeared in 1929 as auditors had been contracting with the directors, or having provisions put into company articles to limit their, and/or directors' liability. This remains subject to a modification in the 2006 Act under Section 534-538 allowing limitation to an amount that is "fair and reasonable in the circumstances".

  E2.0  The requirement in 1 above is about transparency. The requirements in 2 and 3 link both the public audited accounts and the private books of account with solvency. The Companies Act is very simply aiming to protect creditors from companies funding dividends and losses from creditors. Hidden losses in a bank will mean that it is taking in deposits to pay shareholders and managers, and hidden ongoing losses means it is making mispriced loans.

  E2.1  The requirement in 4 was to prevent "take the money and run audits", ie the statutory condition of having one was met, but the contractual efficacy was limited. Sections 534-538 were placed into the Companies Act in 2006 only after a conditional agreement with sceptical institutional investors (concerned then about poor quality audits of some public companies) that auditors would strive to adhere better to the intent of the Companies Act with respect to their statutory duties.[6] The terms of that agreement encountered problems[7] at the outset due to the problem set out below (E4.1).


  E2.2  The UK Companies Act Accounting Rules ("Accounting Rules"), and hence the audits thereof, were for banks in substance loan quality stress tests. The Companies Act positively requires this for dividend safety/capital maintenance to ensure that dividends are paid from firm, not transient or unstable asset values.

  E2.3  However the requirements of the Act have not been backed up properly by functional standards since 2005. The accounting standards have failed to support the Act and the auditing standards then failed to pick it up also. But if the accounting standards regime is faulty, then to the extent that these can apply to how companies account perpetually throughout a year (as IAS 39 does), it is difficult to see how an auditing standard can fully correct it at the end of a year. Information drives behaviour at all times.

  E2.4  The root of the accounting/auditing problem (and indeed the problem manifests before the auditor even starts work) is the change to International Financial Reporting Standards (previously called International Accounting Standards) in 2005. I will not generally refer in this evidence to the problems with "fair values" in the context of banks with trading books, but instead the "sister" of fair value, incurred loss provisioning (see in more detail later E3.5), also a part of the controversial standard IAS 39. That change took out the critical stress-test-for-dividend and capital maintenance purposes, to the extent of being wholly at odds with the intent of the Companies Act. It is possible to make a simple link between the lack of ongoing accounting failing to be a stress test from 2005 with regulators in 2010 rather late in the day having to do it instead. That is not what Parliament intended.


  E2.5  There is also another crucial test that fell away after IFRS was implemented in 2005. That is the test of not allowing values from one group of assets to cover up losses in others.[8] The Accounting Rules forbid "netting off" different things, with a rise in one asset perhaps compensating for the fall in another, but IFRS permitted/required this so as to accommodate fair values of compound assets such as CDOs (collateralised debt obligations). Under IFRS what is valued and disclosed is the sum of the parts at a point in time. IFRS was not only not revealing of the underlying components nor their individual true condition or their risk of losing value, but it also did not even account for things that were already deteriorating.

  E2.6  Under the Accounting Rules, it is forbidden in principle for the illusion of value to be created by essentially an insurance policy sitting on top of a pool of decaying assets (some CDO's carried default protection, as either insurance or derivatives). However, the IFRS methodology served to replace primary and reliably audited information from banks themselves with secondary information from markets, hence inherently second-hand, that was reliant on other agencies, such as the credit rating firms. It has had the impact of delaying the accounting for losses inside CDOs that were often there from inception. Some CDO investments (which could be held on treasury books, banking books, or trading books) were approaching £1 billion in value. So covering up losses within that package is a serious matter.

  E2.7  Under IFRS CDO's were fair valued as a whole from 2005-2007 with virtually no disclosure of what they comprised. In the first quarter of 2008, it became widely understood that there were such poor loans in some CDO's that the insurance in place was from a few and interconnected counterparties and hence that protection was also risk. Only then did the Financial Stability Forum (not the IASB) set a standard to disclose what assets CDO's comprised and what their loss rates were, and the source of the insurance/derivative cover.

  E2.8  From speaking with bank directors in 2007, 2008 and 2009, is was clear that the lack of public information was met by a lack of information on their part also. That is not surprising given the size of banks. Directors receive distilled information just like the public does. The problem was that complying with IFRS can distil information in such a way that does not match with business risk. Losses were occurring in 2006 that did not get accounted for until the first quarter of 2008. There was a mismatch between the real estate foreclosure notices in the USA and the transmission of these losses to the accounts of banks, though the losses were there. In the age of the internet, for modern accounting to delay loss recognition by at least 15 months is a remarkable phenomenon.

  E2.9  Unlike IFRS being accepting of (and incentivising the production of) complex instruments valued as one thing, the Companies Act Accounting Rules are seeking to pull things apart to expose the fundamental parts. The accounting question is what is the cost of each item, and what is the diminution/risk to values of any of the parts.

  E3.0  The difference between IFRS and the Accounting Rules for 2005-07 is stark. With the Accounting Rules any part of a CDO going down would show as a loss (and reduce profit, capital and dividend capacity), IFRS masks that, indeed the market value of the whole thing could be shown as going up (these were called by the IASB "day one gains").

  E3.1  Historic cost accounting (which IFRS has replaced in key places) can be disparaged by invoking the linguistic inference of a lack of modernity. One significant public benefit of historic cost accounting, is the simple fact that the impact of some things going down, cannot be masked by other things going up! Things held that are going down is business risk, things going up whilst not sold is merely opportunity. Under the Accounting Rules, insurance is at best a contingent asset, and a prudent view would require an auditor looking at the counterparty risk before even considering it having any accounting value.

  E3.2  Other problems with IAS 39 included a total lack of business substance. It required banks to legally designate on an asset's inception the accounting classification of either trading or to maturity or to treasury, irrevocably irrespective of the real business use of the asset. Given that IAS 39 gives trading assets profit for going up, in a rising market it incentivises designating assets as trading assets, leading to hoarding rather than selling. Yet commercially the term "trading" implies inventory, goods to be sold in the short run. Not only is the classification misleading, but, as with recent cocoa prices, prices might be high merely because banks were holding on to assets in a rising market due to other buyers being there, or a lack of liquidity due to hoarding. The Accounting Rules forbade taking a profit from merely holding assets, so such problems do not arise. With IAS 39, once a "designation" had occurred, directors and auditors were unable to change the accounting, even if they disagreed with it. It is easy to envisage how situations could get out of control in quite profound ways.


  E3.3  Churchill referred to the architecture of buildings, in particular the Palace of Westminster, as shaping the behaviour of those in it. Accounting standards are directly analogous as forming the core architecture of the financial system. When an accounting standard enables a profit from building a house of cards people will tend to build them.

  E3.4  Other than till money and office buildings, a bank exists as paper contracts and the business can only be described and controlled, by looking at numbers which pull of all of that together. A bank that gets it numbers wrong may "overtrade", essentially self-finance, by taking in deposits and issuing its own paper whilst paying out too much as either; more bad loans, pay, taxes or dividends.


  E3.5  This critical problem can be summarised by comparing these extracts from the two standard systems (ie Companies Act accounts versus IFRS accounts).

    IAS 39:  incurred loss. BC 109 "For a loss to be incurred, an event that provides objective evidence of impairment must have occurred" and "Possible or expected future trends that may lead to a loss in the future (eg that unemployment will rise or a recession will occur) do not provide objective evidence of impairment." (Chuck Prince the former CEO of Citigroup called this "Dancing until the music stops).

    The Accounting Rules. Clause 19. "The amount of any item must be determined on a prudent basis|.and in particular, only profits realised at the balance sheet date are to be included in the profit and loss account." This was then interpreted by the British Bankers Association standard ("BBA-SORP") for banks (franked by the ASB). "The balance sheet amount of advances should reflect any diminution of their ultimate realisable value below their cost." And "To cover the impaired advances which will only be identified as such in the future, a general provision should be made."

  E3.6  The BBA-SORP/Accounting Rules test required a bank in valuing loans, and taking profit, and hence in pricing credit, to take a prudent view of the future, by taking past experience of lending and recessionary cycles. The loss from the test "any diminution" is proportional to the quality of the borrower and his collateral, and general provisions should reflect that. However, IFRS, positively forbids looking at the past or the future. The auditor of a bank operating that system as its provisioning model therefore really has nothing better to go on. And for those inside a bank the question "how are we really doing?" becomes a difficult one.

  E3.7  This is an extract from Lord Turner's letter to the Chancellor, on the Dunfermline Building Society which failed and needed to be bailed out. It indicates the problem with incurred loss provisioning, suggesting that the auditor (using IFRS) felt its loans were understated.

    "In December 2007 as part of the ARROW work, the FSA discussed the Society with its auditors, who informed the FSA that overall the Society was well controlled and who suggested that the commercial loan loss provisions, given the benign market, may not have been entirely justified, ie may have been slightly higher than justified."[9]

  E3.8  If one wished to invent a bank that would collapse, but appear profitable, even when audited, right up to the point of collapse, one could (with IFRS as the standard system):

    (i) lend as much as possible.

    (ii) take as little collateral as possible and charge a premium for that lending risk.

    (iii) compete for marginal business to grow as fast as possible.

    (iv) experience no losses in reported performance, indeed show growing profits and capital until the bank falls over.

  E3.9  None of these things need to be malign, merely a product of group think, caused by rosy internal and externally audited numbers from assuming that these numbers are sustainable. This is accentuated if ones competitors are equally dysfunctional from doing the same things. Any business can sell goods at below cost, but obscuring that is precisely what IAS 39 accounting achieved for high risk lending by some banks.

  E4.0  If accounting and auditing standards do not address the full scope of the law, then clearly audits will not deliver what is expected under the full scope of the law. The audit is intended as a check (a lookout) to then be a check (a brake). With IAS 39 both functions are ineffective.


  E4.1  Other than auditor liability limitation which was new, all clauses mentioned in E1.9 by their 2006 Act references were carried forwards from the 1985 Act to the 2006 Act. However, Sections regarding the "true and fair view" were absent from 2005-09.

  E4.2  The IAS Statutory Instrument[10] (which implemented IFRS in the UK) excised the Companies Act Accounting Rules and deleted the applicability of UK Accounting Standards Board standards, for IFRS companies. However, the Statutory Instrument also excised the "true and fair view" objectives from the 1985 Companies Act for IFRS using companies, and that excision was also replicated for Companies Act (non-IFRS) accounts as well.

  E4.3  Although "true and fair view" clauses were then put back by the Companies Act 2006, different parts of the new Act were invoked at different times, therefore these clauses were not operative from 2005-09. Given that the true and fair view is a backstop precisely when an accounting standard or its application might be defective, its absence corresponded with precisely the period that problems were in the banking sector due to just that.

  E4.4  The problem of the result of the Statutory Instrument was identified by institutional investors and one large accounting firm, and then supported by politicians and then the other accounting firms, but not regulators.[11] The essence of investor concern was the following:

    (i) true and fair view as an overarching objective of Companies Act accounts is a backstop for precisely when accounting standards are defective or inadequate in the circumstances.

    (ii) Company law has also never defined the true and fair view, merely set the context for it. However, IFRS defines true and fair view as not only the product of applying IFRS, but applying IFRS for the purposes set out in IFRS.

    (iii) IFRS as to purpose, is not only not silent on dividend assurance and capital maintenance, but the IASB was opposed to the construct that audited accounts were for dividend assurance and capital maintenance for the benefit of creditors and members.[12] The stated purpose of IFRS is "to be useful for users" instead.

    (iv) IAS 39 was known by many to have been "the Enron standard"[13] and it was not apparent that the IASB had fully appreciated its problems with it. Also IAS 39 was inherently inconsistent with dividend assurance and capital maintenance.

  E4.5  A summary of the key changes from the above is set out in this table. The grey type is to demonstrate the various levels of confusion created after 2005.

PositionPre-2005 accounts of all companies Companies Act accounts 2005-09 IFRS Accounts 2005-09
True and fair view override Yes No, or at best unclear.No, and no evidence of it being used in a UK bank, (it was used in France).[14]
Prudence as an overriding accounting rule? YesAccounting Rules are intact, but the ASB put IAS 39 into FRS 26 (see below) No. Imprudent provisioning was compulsory for banks (including banking subsidiaries), and this standard applied listed entities (their consolidated accounts) and for the individual company accounts of bank holding companies.
IAS 39 (with its incompatibilities.)N/A The BBA-SORP required loans to be carried at "not more than their ultimate realisable value below cost." IAS 39 was incorporated into UK FRS 26, despite being inconsistent with the Accounting Rules of the Companies Act, (incurred loss provisioning, below).
Incurred loss provisioning. Forbids adjusting loans for inherent credit risk.

  All this meant that from January 2005 a bigger fog than usual descended around the banks' accounts.

  E4.6  However a fog in objectives was already in place around the auditing standards.

PositionUK Auditing Standards to 2005 ISA from 2005
An auditing standard dealing with the matters in the 1985 Act equivalent to the 2006 Act of 836 and 837? No. In some supplemental guidance, on a form of words when an auditor qualifies his opinion.
But, this does not indicate the dividend test as a criteria for qualification, ie it does not direct auditing in that direction. +
Materiality objective in auditing standards? Accounting materiality [ie relevance for auditing purposes is: "Material in the context of the financial statements taken as a whole"*
Fraud and error? Both are relevant to capital/dividend safety. Auditing standards consistently emphasize the role of directors/managers to find fraud. This is somewhat odd as auditors use auditing standards not directors.

  #  ISA International Standards of Auditing.

  *  But, Section 837 does not restrict accounting materiality to this very general sounding criteria. Paying dividends out of capital (depleting reserves) by only £1 is a breach of the law (when and if reserves get that low). Company Law expect dividends to be what the company can bear and expects the accounts to reflect that. FSA banking regulatory capital is merely an additional constraint to that, essentially adding to share capital as the balance sheet grows.

  +  this omission therefore might lead auditors into believing that if their audit and the accounts comply with accounting standards and auditing standards, then they have achieved their objective and do not have to qualify their opinion. The auditing standards are therefore appearing to set a passive and reactive rather than an investigative objective.

  E4.7  In contrast to that, Section 837 is very clearly expounded in HMRC material,[15] it is an acid test, but it is surprisingly absent from much regulatory material, to the extent that even some regulators seem to have missed its significance in what they are regulating.[16] What is most relevant is that Section 837 is entirely incompatible with "incurred loss" provisioning.

  E4.8  Reasons for this acid test not being more clearly set out in auditing standards might include:

    (1) auditing practice is global. The USA does not have an equivalent acid test (in statute) for the auditors, hence the US profession is unlikely to wish to see it in international auditing or accounting standards.

    (2) the UK accounting profession might prefer it was not in the law. Any focus on it betrays the concept of "an expectation gap" as somewhat of a fig leaf.

    (3) external audit quality inspection processes established under self-regulation have been founded on auditing standards as the quality yardstick, rather than these more explicit parts of the law. "The inspection gap" then matches the expectation gap by inspecting "in accordance with standards". It holds up the fig leaf.

  But—as has happened—when accounting standards are faulty, there is no clear audit objective to follow either. The law requires the audited accounts to reflect the business condition, in numbers, and an assessment of proper control (Section 386), so that dividends can be paid, or withheld. Auditing standards are a fudge.

  E4.9  The recent criticism of auditors arises despite much more audit regulation. Yet from 1879 to 2005, in the current and former sterling area, banks did not collapse systemically with no regulator and in places no lender of last resort. All of this area applied a similar accounting, reporting and governance system, the Companies Act.


  E5.0  I believe that there has been an insidious problem with a "too few to fail" attitude within the regulatory environment itself.[17] That has manifested in regulatory support for sheltering auditors from civil liability claims.[18] It is a classic manifestation of regulatory capture.[19]

  E5.1  However, as we have seen there is a widespread recognition of the moral hazard of a "too big to fail" situation with banks. The same problem arises with audit firms. If poor auditing can lead to banks failing (as was held in cases in the UK in the 1990's and 1980's), then the combination of "too big to fail" (banks) and "too few to fail" (auditors) is a lethal combination.


  E5.2  Standards (accounting and auditing) are heavily influenced by the profession itself, a part of which has a vested interest in obscuring its statutory test of audit quality, which is an acid test, and replacing it with a far woollier one for public consumption and professional inspection. I have observed regulators not believing what Parliament intended, they then overlook it and hence themselves unknowingly subvert it. A significant part of the regulatory oversight system has very little grasp of the scope of the law as opposed to the standards. I have yet to see any evidence that the UK's FRC uses information on successful auditor litigation (ie contractual failure) as a criteria for audit quality. It instead inspects according to (regulatory) standards with the lesser tests. It is checking false compliance, and driving a paradigm of "good" corporate governance on that model.

  E5.3  Essentially Enron and the then collapse of Anderson was the tip of an iceberg that has unfortunately been hit twice. The lessons of that crisis were not learned and many of the remedies since have been quack medicine.


  E5.4  Currently there is talk of auditors doing more on "risk", and auditing risk statements. It is quite easy to see its attraction as both a distraction and a way of extracting more fees. The competitive advantage to accountants should be around numeracy. The risk to a bank is not getting money back having lent it. The key issue is getting the numbers right, and delivering as the law already requires but the standards have not delivered. Banks overstated numbers to one degree or other most probably from early 2006, or earlier in some cases, due to the problems with IFRS. That was the problem, more words and more fees around the more words are not going to address that. Other suggested remedies such "talking more to the FSA" will not help either party much if the auditor has not bottomed the numbers.

4 August 2010

1   Answer to Justine Greening MP from Margaret Hodge MP, Minister, July 2006. Back

2   Evening Standard, June 2008 Back

3   "Are banks different?" Lord Turner, January 2010. Back

4   House of Lords, Caparo vs Dickman, 1990. Back

5   Also called "UK GAAP". Though UK GAAP, is Companies Act formats and rules, and then Accounting Standards Board Standards. Back

6   NAPF/Morley Fund Management, "Bringing Audit Back from the Brink" and "Undermining the Statutory Audit" 200-05. Back

7   Financial Times, Barney Jopson, 24 June 2005. Back

8   Companies Act, Schedule 2, Part 2, Clause 21.£sch2-pt2 Back

9   Letter from Lord Turner to the Chancellor, Alastair Darling MP, Back

10   SI 2004/2947. Back

11   Financial Times-John Plender. 10 July 2005, "Battle for truth in European Accounts". Back

12   IASB Conceptual Framework Project. IASB 2005. Back

13   "Following the Money", The Enron Failure and the State of Corporate Disclosure, Brookings Institution, USA, Back

14   Soc Gen in 2008 overrode IFRS, to match the losses of unauthorised dealing, with the period in which the positions had been started. The IFRS treatment would have shown a profit for 2007, as the positions were in the money, with the losses being deferred until 2008. Back

15   HMRC manual-notes on Company Law aspects of dividends, Back

16   FRC/FSA consultation July 2010 on matters arising from the results of inspections of bank audits. This does not mention the impact of potential overvaluations of assets found in the inspections with the impact on dividends. Back

17   New York Times, Back

18   FRC delegations to DTI in 2004/2005 to support a liability cap. Opposed by HM Treasury. Back

19   "Regulatory Capture" Stigler, Chicago, Nobel Prize 1982. Back

previous page contents next page

© Parliamentary copyright 2011