Panel on Tax, Audit and Accounting

Written evidence submitted by Deloitte (SH 007)

Thank you for your letter of 5 December 2012 and request for information concerning the role of tax, audit and accounting in banking standards and culture. We have restricted our comments to those areas, rather than any wider causes of the financial crisis or concerns about banking standards and conduct. Our responses to the detailed questions set out in your letter are included in an appendix to this letter.

In summary, our views are that:

· International Financial Reporting Standards (IFRSs) should continue to be used for the general purpose financial statements of all listed companies, including banks. IFRSs are designed to provide a neutral view of the business. To the extent that regulators wish to see the business differently, this should be dealt with by supplementary information in regulatory returns. Banks could be requested to provide additional disclosure reconciling key accounting to regulatory information, either in Pillar 3 disclosures or in the annual report and financial statements. Development of a special purpose accounting regime for banks could result in user confusion and a lack of comparability.

· We are supportive of the efforts of the International Accounting Standards Board (IASB) and their US equivalent, the Financial Accounting Standards Board (FASB) to develop an expected loss model for loan loss provisioning to replace the existing incurred loss model. We are keen to see a standard that is both capable of practical application and which gives transparent, useful information to all users of the financial statements.

· We are supportive of efforts to improve communication about risk and uncertainty to users of the financial statements and enhance the related information contained within the audit report. The increased transparency that will be introduced for listed companies by the FRC’s 2012 revisions to the UK Corporate Governance Code is supported by investors as an appropriate way to do this. Global and UK moves towards a more informative audit report summarising the key risks of material misstatement and the auditor’s response will also improve communication and better demonstrate the value of the audit to users of financial statements. We continue to believe that the existing pass/fail model for the auditors’ report remains appropriate. Graded audit opinions could be confusing, and are unlikely to solve the issue whereby a qualified opinion (or one containing an emphasis of matter) triggers a run on a bank – any downgrading could trigger a similar run.

· In our experience, the FSA’s Code of Practice for the Relationship between the External Auditor and the Supervisor is already working well. We welcome the fact that, under the Financial Services Act 2012, the Code will be required to cover communications in both directions between auditor and supervisor, contributing to a strong two-way dialogue to supplement tri-partite meetings between auditor, supervisor and audited entity. We are supportive of moves for a similar code on a global basis.

· We support continued alignment of the tax base with accounting (and hence oppose moves towards an Allowance for Corporate Equity (ACE) or Comprehensive Business Income Tax (CBIT)). Further alignment by focussing on the income statement (rather than, as currently, taxing volatile fair value movements recorded in other comprehensive income) would also provide improved transparency for users of the financial statements and policy makers as to the true amount of tax paid compared to profit. This approach minimises the temptation to tax arbitrage. For the same reason, we do not support the development of a separate tax system for banks, nor a UK-only financial transaction tax which could divert business offshore or to non-bank financial institutions and/or alternative markets.

· Whilst the recent environment has been one in which many banks do not have significant taxable profits, HMRC’s Code of Practice on Taxation of Banks has already shown that it can be effective in addressing tax planning with no commercial substance.

Appendix – Responses to detailed questions set out in your letter of 5 December 2012

We respond to both your key and expanded questions below. Our responses are immediately below the question or group of questions to which they relate.

Q1 How, if at all, does the tax system encourage leverage in banks? What is the effect of having tax relief for debt interest but not for dividends on equity? What effect does this have on the stability of the banking system?

Q1.1 How, if at all, does the tax system encourage leverage in banks?

Q1.1.1 How large is this effect?

Q1.1.2 What are the determinants of the size of the effect?

Excessive leverage is generally accepted as being one of the main causes of the banking crisis and hence the extent to which the tax system influences leverage is an important question. Some economists argue that the tax system encourages businesses to finance themselves more by debt and less by equity than would be the case if interest was not tax deductible. In principle it is not clear that the application of this argument to banks is any different to its application to other businesses, except that banks raise a great deal of debt finance.

This economic argument is not without challenge. Focusing on public companies, banks operate under an increasingly internationally harmonised accounting and regulatory framework under which the costs of debt finance are deducted from reported profit and the economic costs of raising equity are not so deducted. Those reported profits are widely accepted by the markets as a key performance indicator and a basis for management incentives.

In the closely held company sector the arguments around the choice between debt and equity finance have a different dynamic. In our response to Question 1.1.6 about the de Mooij study of potential debt bias in the tax system, we note that this study is based at least as much on privately held companies as publically held ones. In the UK, owners of incorporated businesses can typically secure tax relief personally for the cost of their own borrowing to provide either debt or equity finance to their companies. Economically, it is the debt raised by the individuals that presents the true picture of the capitalisation of the business: whether that funding is passed between them and their companies in debt or equity form is less significant. However because the company level information is typically publically available and individuals’ information is not, it is the former information which is generally used for research into such issues. Omitting the personal level of debt would seem to call into question the value of the research. Comparable factors, specific to other jurisdictions, may be at work in relation to data included from other countries.

Banks’ decisions to finance via debt or equity are not driven solely by tax and indeed there are fundamental commercial differences between them. Debt is issued for a fixed term, and hence can offer a return appropriate for the place in the current economic cycle, whereas issuing equity requires a consistent return with existing equity in perpetuity. The use of debt issued to match current levels of lending (for example) is therefore designed to match costs with income.

Banks are necessarily leveraged. Market forces and regulatory rules impose restrictions on the maximum amount of leverage a bank can have. Banks are already incentivised to operate as close to this maximum as they think they can by the ability that affords to write more business from a given equity base, irrespective of any tax effect.

Banks based in jurisdictions with a low (or nil) effective corporate income tax rate are also leveraged. It appears to be generally accepted that UK banks increased leverage in the period to 2007, even though UK corporation rates were then at an all-time low of 28% (compared with rates of 30-35% in the 1990s and as high as 52% in the early 1980s). This is contrary to what one might expected based on the correlation identified in the de Mooij study and similar research, which predict that leverage will increase when tax rates rise.

At different times and in different countries, some forms of debt have been accepted as regulatory capital and some forms of regulatory capital have been tax deductible. For example, tax deductions for interest on some regulatory capital instruments, in particular the innovative tier one capital instruments that became popular in the late 1990s, may have encouraged banks to focus on this type of capital as an alternative to equity funding. We note that due to international regulators’ concerns with the loss-absorbing ability of some of these instruments, Basel III (and hence via proposals in Capital Requirements Directive IV the requirements of the Financial Services Authority and the new Prudential Regulation Authority) now puts a greater focus on core tier one capital. Given that these regulatory limits restrict the amount of tax deductible tier one capital instruments that can be issued, their impact is unlikely to have been material to the capitalisation levels of the banking sector.

If there is a policy desire to reduce leverage then our view is that regulatory rules and the attractiveness to the markets of bank equity (or at least some alternative plan for obtaining such equity) should be the main avenues explored. Without additional equity, tighter regulatory rules would risk less banking business being written, rather than more capital supporting the same amount of business, with a consequential impact on the supply of credit to the rest of the economy.

Our experience suggests that most banks have reduced leverage since the onset of the financial crisis for commercial and regulatory reasons. It seems likely that non-tax factors would continue to exert pressure on leverage ratios whether the cost of deducting debt were deductible or not. Tax system bias towards debt has only a relatively minor impact on a bank’s leverage or the stability of the banking system.

Q1.1.3 Do taxes other than the corporate system introduce any distortion?

The UK bank levy (introduced in 2011) discourages leverage, since debt liabilities are generally chargeable to the levy whereas Tier 1 equity funding is specifically exempt.

Stamp duty reserve tax is a cost borne by typical investors in UK equities but not by investors in debt instruments. To the extent that this becomes reflected in a greater investor appetite for debt rather than equity instruments, it may have a marginal effect in encouraging leverage.

Q1.1.4 To what extent is the distortion offset by the personal tax system?

The distortion between the treatment of debt and equity finance typical of most corporate tax systems is to some degree offset in many countries by a more favourable treatment of capital gains and of dividends than of other forms of income such as interest. Clearly such offset will not be exact.

For listed banks, the personal tax offset is likely minimal as many categories of market investor such as pension funds or insurance companies are either tax exempt or pay tax on a different basis.

Q1.1.5 Would regulatory constraints on capital no longer bind if the distortion were removed?

The focus should be on ensuring appropriate capital requirements for banks as part of the regulatory regime. Substantial work has already been undertaken on this front since the financial crisis. We expect the regulatory and market pressures on banks (including the threat of regulatory sanction) will continue to bind banks’ behavior and financing strategies, regardless of the tax treatment of debt and equity.

Q1.1.6 Has the responsiveness of the effect to its determinants changed over time? ("We find that, typically, a one percentage point higher tax rate increases the debt-asset ratio by between 0.17 and 0.28. Responses are increasing over time, which suggests that debt bias distortions have become more important." de Mooij (2011), The Tax Elasticity of Corporate Debt: A Synthesis of Size and Variations, IMF WP11/95)

The study to which you refer is a high level 'meta-study' combining results of various historical studies with widely varying methodologies, definitions of key terms and other features. None of the surveys included appear to focus on banking.

The evolution of these studies reflects a search for methodologies that will yield the results predicted by theory. For example, on page 4, commenting on earlier studies that looked at time series variations, "These estimates, however, often yield insignificant effects or coefficients of the wrong sign". The reason given for dismissing these inconvenient results is that "variation in CIT rates over time is generally small". This glosses over the fact that, for example, in the UK corporation tax rates nearly halved from 52% in the late 1970s to 28% on the eve of the credit crunch (and have fallen further since), that the tax base moved more in the direction of accounting profits with its clear differential treatment of debt and equity, and that the corporate tax take in the same period multiplied from less than GBP 5bn per annum to about GBP 50bn per annum.

The studies generally focus on solo company results in specific countries. It would be unsurprising if, for example, multinationals tended to issue more debt in high tax countries than in low tax ones or if family businesses tended to issue more debt out of the corporate entities under their control and less debt personally where corporate tax rates were relatively high. Such results are of interest to tax policy. However, what is of most relevance to judgments about whether there is excessive leverage in the business is the debt/equity ratio at global level in the case of multinationals and at the level of the overall business structure of the family in the case of family businesses. There appear to be little or no evidence in this study of any impact of varying tax rates on these matters.

The studies are reported as suggesting that the sensitivity of leverage to tax rates is reduced where countries have 'thin capitalisation' rules (essentially, rules restricting interest deductions in cases of excessive leverage). Such rules can indeed be viewed as the consensus international response to the problem suggested by the meta-study. They exist in various forms in the UK. In banking, the regulatory capital rules operate as thin capitalisation rules by preventing excessive leverage (according to prevailing international definitions) within one legal entity. Therefore, even in its own terms, the meta-study is of reduced relevance to banking.

Q1.2 What is the best way of removing any such bias? 

Q1.2.1 ACE vs. CBIT

Q1.2.2 Are there any other workable options?

We think the UK should not act unilaterally to change fundamentally the tax treatment of debt and equity. As set out in our response to Question 1.1, we are not convinced that there is a problem with the tax system, as regards bank leverage, that requires fixing. At the very least, we believe no action should be taken without a much more detailed study of the scale of the supposed problem, the role of global accounting, regulatory and management incentivisation rules and market practices, and the consequences for bank lending to business if removal of the distortion led to less business being written.

Introducing an Allowance for Corporate Equity (ACE) would give a tax deduction for the cost of equity and thereby reduce the tax base. This would therefore either reduce the tax take for the Exchequer or lead to a significant increase in marginal tax rates, possibly to uncompetitive levels internationally. In our experience, ACE is not commonly used internationally and has been abandoned after a few years by most of the few countries that have introduced it. Global banks are therefore unlikely to regard the introduction of ACE as a reliably enduring feature to the extent of investing more in the UK (certainly as compared to a reduction in the rate of corporation tax which conforms to a long established international trend toward broader base, lower rate, but higher take).

A Comprehensive Business Income Tax (CBIT) model denies a tax deduction to the cost of borrowing. This would be a major extension of the tax base, disallowing a typically large business cost and bringing corporation tax closer to being a tax on turnover than a tax on profits. This would be particularly the case with banks for which debt costs are akin to the 'cost of sales' and could be prohibitively expensive unless the rate were reduced substantially.

It could be argued that a distinction should be drawn between interest and other financing costs, which are integral to a bank’s trading activities and more structural debt funding, such as debt incurred in purchasing shares in a subsidiary company. Analysing financing costs between those that are integral to a bank’s trading activities and those that are for other purposes would introduce additional complexity, given the fungibility of funds.

A CBIT model could reduce the competitiveness of the UK economy as a whole, as well as of UK banks, as it could penalise expanding banking business in the UK relative to other countries. The trend of tax policy in recent decades has been to produce a less distortive corporation tax raising more revenue. CBIT would be closer to this trend (expanding the tax base, which might enable further rate reduction) than ACE. It is still at odds with that trend in that base-broadening has typically been done by bringing tax rules into line with accounting rules and international norms. CBIT would be a major and uniquely large step away from them.

Q1.3 Can and should the reform be restricted to banks, as opposed to other financial and non-financial companies?

As indicated above, we do not believe that fundamental reform of the taxation of debt and equity is warranted. A detailed review of the consequences would be needed prior to any major change.

Q1.4 What are the administrative and international complexities involved in such a reform?

Q1.4.1 What would happen if the UK unilaterally introduced an ACE reform?

Q1.4.2 Are there any international legal obligations preventing a reform in the UK?

Q1.4.3 How should the balance sheet of UK banks’ foreign subsidiaries, and intra-group debt and equity positions, be treated for the purposes of the reform?

We are concerned that any fundamental change in the UK tax system would place the UK out of line with the norms of internationally accepted practice with which global institutions are familiar. This would inevitably lead to uncertainty for some years and could damage the appeal of the UK as an attractive place for new business.

Q1.5 What are the implications of a reform for the Bank Levy?

Her Majesty’s Treasury (HMT) and Her Majesty’s Revenue and Customs (HMRC) have already proposed a review of the Bank Levy in 2013 and we will be contributing our views to that review. As it is not clear what changes might be proposed, any comment on the implications would be speculative. We note that the rate of Bank Levy has increased five times since it was introduced, apparently in order to achieve the anticipated yield for the Exchequer. The cost falls mainly on UK-headquartered banks (as it is charged on their global consolidated balance sheet). Whilst some other jurisdictions such as France and Germany have also introduced bank levies, the tax base and rate differ from the UK. The USA and many other countries currently have no equivalent levy. These factors therefore tend to increase the cost of undertaking banking business in the UK.

Q1.6 How much would the reform cost? 

Q1.6.1 How can the distributional effect be neutralised?

Q1.6.2 Is it practical to apply the allowance only to new equity?

Q1.6.3 Can the rate be increased to offset the narrower base?

We would need more detail about any proposed reform in order to estimate costs and potential effects.

Q1.6.4 Could the banks issue shares to the Government to compensate it for the loss of tax revenues?

The only circumstances where we would expect a bank to want to issue shares to the Government, in lieu of a tax payment, would be if the bank were in dire financial circumstances. In this case valuing such shares would be difficult and potentially impractical. The Government could be left holding illiquid shares with no clear means of realising the asset. This would therefore appear to introduce an unnecessary mechanism and additional legislative complexity for little benefit.

Q1.7 What are the economic consequences of financial services not being subject to VAT?

The financial services exemption from Value Added Tax (VAT) is a long-standing practice, enshrined in EU law. VAT is not charged on a wide range of financial transactions due to the practical difficulties of identifying the value added to charge to tax. This exemption benefits the personal customers, as if VAT were chargeable, this cost would be passed on to the consumer. Banks suffer considerable amounts of VAT on their inputs, which they are unable to recover from HMRC, which adds to their cost base and their 'cost to income ratio' (another key market-accepted performance measure).

Q1.8 Should financial transactions or ‘financial activity’ be taxed?

There has been significant debate on the subject of taxing financial transactions (a "Tobin" tax) over the last few years. We believe that this could be detrimental to the UK’s position as a leading international financial services centre unless applied and properly enforced globally. The likelihood of achieving such international consensus appears remote and so we do not believe that such a tax should be introduced.

There are already examples of taxes on certain financial transactions in the UK, notably Stamp Duty and Stamp Duty Reserve Tax (SDRT) on most purchases of shares (with special related regimes for investments in Unit Trusts and Open Ended Investment Companies, and for securities held in depositary receipt schemes such as ADRs and clearance services). The focus of the tax on equity investments and the exemptions available (notably for market makers) tends to make it a cost borne by banks’ customers rather the banking sector itself. Accordingly, it does not have a significant impact on the competitiveness of UK banks compared to overseas banks.

The EU is currently considering a financial transactions tax, under the enhanced cooperation procedure, to apply in particular Member States. France has also recently implemented a financial transactions tax on the value on all transfers of ownership of securities issued by publicly traded companies with a market value of over Euro 1bn. The rate of tax of 0.2% compares with the main rate of SDRT in the UK of 0.5%. Although levied on brokers and custodians, and applying to a somewhat wider range of transactions than SDRT (for example to certain sovereign credit default swaps) we understand that, as with SDRT, the main economic burden is borne by investors in equities.

Q1.9 What are your views on suggestions that there should be an additional bank levy to bail out future failures, or that certain levels of remuneration be treated in a similar way to distributions?

Existing regulatory reforms, such as those on capital, recovery and resolution planning, are intended to ensure no future bank bailouts by taxpayers. Retail depositors will be protected (up to a set amount) through the Financial Services Compensation Scheme and the cost of this already falls on the bank (and building society) sector. With wholesale depositors potentially being "bailed in" we do not see an obvious role for an additional bank levy.

Remuneration of banks’ employees and directors has been the subject of significant review and change in the last couple of years by regulators. Steps have being taken on governance, deferral and transparency of remuneration, culminating in the FSA’s Remuneration Code. The Code includes a requirement that at least 50% of variable remuneration should consist of shares (or similar instruments) subject to a suitable retention period. This aligns the employee’s interests more with the medium-term and away from short-term profit at the expense of risk. The FSA is collecting data on remuneration practices and we believe that it is important to consider the output of that exercise before considering any further changes.

Q2 What are your views on alternative systems to level the playing field?

We assume this question refers to the tax treatment of debt and equity from Question 1. As suggested earlier, we believe that the regulatory system is being reformed in a way that will address this issue by stipulating minimum levels of equity. We believe that a regulatory response is the appropriate approach to reduce bank leverage.

Q3 Do banks’ attitudes to tax planning affect banking standards and culture, and does this have any effect on the wider economy?

In our experience the opposite is the case: it is the culture in banks that affects the banks’ attitude to tax planning. Different banks have different risk appetites, so there is no single standard or culture.

A distinction needs to be drawn between tax planning undertaken by the bank as a corporate and transactions entered into with clients. For the former, banks are no different to other large corporates in seeking to structure their affairs in a tax-efficient manner. In respect of transactions entered into with clients, we have observed that teams that had been primarily focussed on structured tax transactions with clients have either been disbanded or re-assigned to other activities.

We believe a number of factors have contributed to the fall in activity in the customer business. These include the pro-active work of UK and other tax authorities to close loopholes in domestic law and gaps between different tax systems; HMRC’s Code of Practice on Taxation for Banks; a lack of taxable profits (which limits the benefit of any tax planning); together with constraints imposed by banks’ desire to shrink their balance sheets and focus on liquidity.

Q4 Do you have any views on the role and purpose of structured capital markets teams in banks? Does the volume and type of structured tax transactions have any effect on bank stability, and did this play a part in the banking crisis?

"Structured capital markets teams" will mean different things to different people. Many of the operations we see described as such provide genuine commercial benefits to their customers. Examples would include:

(i) structuring project finance or leasing transactions in a way that is commercially attractive to the customer (which may include UK or other tax considerations) whilst meeting security concerns for lenders;

(ii) asset-backed finance for various asset classes to enable a cheaper or longer term or alternative source of finance; and

(iii) innovative financing structures such as Islamic financial products.

Structured capital markets activity could have included the development of complex instruments such as collateralised debt obligations (CDOs) and other structured products. We believe that this activity, which was one aspect of structured capital markets business, did play a part in the banking crisis. It created complexity and made it harder clearly to identify and to monitor the risks inherent in the original lending.

In the desire to earn profit, banks and others invested some of their surplus liquid funds in what were then thought to be relatively low–risk debt securities (such as securities rated AA by a reputable ratings agency). In retrospect some of those securities, such as CDOs or investments in Structured Investment Vehicles, were not low-risk and resulted in significant losses. These losses together with reduced liquidity were factors contributing to the financial crisis.

Tax structured transactions in particular in the years immediately preceding the credit crunch tended to exploit potential tax arbitrage opportunities with minimal creation of additional credit or liquidity risk. The public policy issue they pose is therefore the impact on national tax bases rather than to the stability of the banks. Such arbitrages tended increasingly to be between different national tax systems given that successive anti-avoidance legislation had largely shut off such opportunities within the UK.

Q4.1 Is there any evidence that post-crisis this activity has moved into the sphere of shadow banking and if so, does this have any effect on bank stability and regulation?

To the extent that structured capital markets activity is customer-driven, customers are still likely to seek these services; hence if banks cease to offer these products or services, non-bank businesses (sometimes referred to as ‘shadow banks’) may seek to fill the gap. This is more likely to be the case if regulation or other factors allow other providers to be more nimble or more cost-effective. This poses a potential issue for regulators as if banks or pension funds lend to ‘shadow banks’ who enter into structured transactions (rather than the banks doing so directly), systemic risk may become harder both to spot and quantify. We do not, however, have any evidence that such structured activity has moved to a shadow banking sector.

Q5 What are your views on the effectiveness of the Code of Practice on Taxation for banks? Would the Code benefit from having sanctions and if so what should these be?

Due to the approach adopted by HMRC since at least 2006 to build an open working relationship with large taxpayers, many large banks had already moved to meet many of the requirements of the Code. When the Code was first proposed in 2009, the main area of concern for banks related to the tax planning section; in particular defining what was acceptable, both in terms of the bank’s own tax position and any assistance provided to the banks’ customers in their tax planning. HMRC took on board some of those concerns in developing the final wording of the Code. Since 2010, it is clear from figures published by HMRC, that most banks operating in the UK have voluntarily adopted the Code.

Whilst we consider the Code has been effective in influencing the behaviour both of banks and HMRC, as noted in our response to question 3 above, the focus on balance sheet size and liquidity, increased regulatory insight and lack of taxable profits had already played a part in reducing banks’ tax planning activity. So, for some banks, adopting the Code meant little or no changes were required at that time.

Having adopted the Code, banks are now committed to operating within its scope, even if economic conditions again become more favourable to tax-based arbitrage. We are aware that HMRC’s view has been sought on a number of occasions as to whether certain transactions fall within the Code. This suggests the Code is having an impact. Clearly, banks cannot now undertake transactions that are not compliant with the Code without breaching the Code. In that respect the Code could have long-term benefits for the Exchequer.

As the Code is voluntary, sanctions appear to be inappropriate. HMRC’s published Governance Protocol on compliance with the Code makes clear their approach, including public comment by HMRC in certain cases if it thought a bank were not in compliance.

Q6 How effective has the Senior Accounting Officer legislation been with particular regard to banking standards and culture?

The Senior Accounting Officer (SAO) legislation is relatively new and a ‘light touch’ approach was adopted by HMRC in year one, so it is too early for us to be able to make considered comments on the long-term impact of the SAO legislation. We note that the SAO legislation applies only to specified tax obligations and many of the specific financial services tax requirements are not within its scope. SAO legislation applies to UK companies; so inbound organisations that operate largely through branches in the UK are not in scope.

Banks have operated within regulatory requirements for many years, so had a need to have appropriate systems and controls in place before the introduction of the SAO legislation. We are aware that banks have undertaken projects to ensure compliance with the legislation, sometimes with external assistance. Some projects have helped banks to improve the quality of information, thereby leading to tax savings (for example, by being better able to identify allowable deductions, rather than prudently assuming no tax deductions are available).

Q6.1.1 Would the introduction of a power akin to that of the FSA in S166 FSMA 2000 be a useful addition to HMRC’s toolkit?

No. The FSA’s powers under section 166 mean it already has powers to review a bank’s systems and controls in respect to tax matters; tax risk is one form of operational risk which banks are required to monitor and manage. There may be a benefit to HMRC and the FSA sharing information or otherwise working together (see also our response to Question 9 below).

Q7 Do we need a special tax regime for banks? If so, what would this look like and what would be priorities for change? Should tax continue to follow accounting with respect to banks? Should the tax system actively seek to influence banking standards and culture?

It is not clear why a special tax regime would be desirable for banks.

Profits computed for UK corporation tax purposes are largely based on accounting profits, as calculated in accordance with UK Generally Accepted Accounting Practice (GAAP) or ‘international accounting standards’ [1] . The reliance of tax calculations on accounting treatment has become increasingly important over the last two decades. As accounts are primarily prepared for non-tax reasons, basing tax calculations on the accounting measure of profits ought to reduce compliance costs, uncertainty for taxpayers, the risk of manipulation and the opportunity for domestic and cross-border arbitrage. Were bank-specific tax rules to be introduced, this could be an additional encouragement to arbitrage between banks and non-banks.

Although we generally favour basing tax on the accounting results for these reasons, it is generally accepted that there should be departures from the accounting results for good and clear public policy reasons. These include avoiding potentially large volatility of annual cash tax liabilities as a result of particularly volatile items reflected in accounts, such as fair value movements. Although the UK’s tax rules address this issue in several important respects, there is one issue which it exacerbates (by including in the tax result fair value movements which in accounting terms are taken directly to equity rather than in the income statement). This affects all corporates but is particularly acute for banks – they are effectively required for regulatory reasons to hold significant volumes of assets such as gilts and listed corporate bonds which give rise to such fair value movements. Although this was identified before the introduction of international accounting standards to UK tax law in 2004, no change in the law has been forthcoming. This seems overdue for review.

Q8 Are banks exploiting regulatory and information arbitrage between FSA, HMRC and auditors? If so, what is needed to address this?

We are not aware that banks are exploiting regulatory and information arbitrage between FSA, HMRC and auditors.

Q9 Should there be a ‘safe environment’ in which the tax authority, regulator and auditors can share confidential information and concerns, possibly on varying levels of seniority?

It should be possible for regulators, tax officials and auditors acting within the constraints of their current statutory and professional obligations to share in good faith key information at the appropriate level of generality. As auditors, we have regular discussions with regulators. Our own ethical values as a firm and specific regulatory requirements on auditors mean that if we came across misconduct in the course of our audit we would relay this to the appropriate body. For example, we are already required to report suspected money laundering (which could include suspicions of tax fraud) to the Serious Organised Crime Agency. The FSA’s Code of Practice for the Relationship Between the External Auditor and the Supervisor, introduced in September 2011, has seen an increased commitment to active information sharing.

We think it important that as information sharing increases, the respective roles of regulators, auditors and tax authorities remain distinct and clearly demarcated. In this context, we would be reluctant to see the auditor become an extension of the regulatory authorities or tax-collecting agencies, as we do not think this would result in better financial statement audits. The FSA’s Code of Practice for the Relationship Between the External Auditor and the Supervisor shows how cooperation can be improved without undermining these different roles.

It is a question for HMRC and the FSA as to the extent to which they can already and/or wish to share information with each other. The HMRC Code of Practice on Taxation for Banks includes consideration of tax risk, which may be of interest to the FSA (and its successors) as part of their wider remit into risk within financial institutions.

Q10 What was the role of accounting standards and reliance on fair value principles in the banking crisis? What does a ‘true and fair view’ really represent to the market?

We address this question in the three parts - what was the role of accounting standards, what role did reliance on fair value principles have and what does a ‘true and fair view’ really represent to the market? We also refer you to written evidence and supplementary evidence we have submitted to the Lords Economic Committee enquiry, ‘Auditors: market concentration and their role’, which addresses a number of the accounting, audit and regulatory issues raised by your questions.

What was the role of accounting standards?

The role of accounting standards is to provide consistent, transparent information to the market, and they continued to perform this role during the financial crisis. During the crisis, International Financial Reporting Standards achieved the objective of providing neutral, transparent, reliable and comparable information on financial performance, including fair values and loan loss provisioning, within the accounting framework that operated at the time of the crisis. We elaborate further, in our response to Question 11, our views on loan loss provisioning.

We support the goal of a single set of high-quality global accounting standards applied consistently across industry sectors. IFRSs as developed by the International Accounting Standards Board are the only globally recognised set of accounting standards and are capable of filling this role. We also support the G20 Leaders’ call for increased international regulatory cooperation and intensive work towards the completion of a single set of high-quality global accounting standards.

The financial crisis demonstrated that financial markets are global and the consequent importance of consistency and transparency in financial reporting across national boundaries. Further the performance of global markets before, during and subsequent to the crisis illustrates the volatility of asset markets through cycles. The question of whether accounting reflected this volatility accurately or somehow was a driver of that volatility is at the heart of the debate about fair value accounting.

What role did reliance on fair value principles have?

IFRS in certain cases requires and in some cases permits the use of fair value for subsequent measurement of certain financial assets and financial liabilities. It is a common misconception that fair value is the only measurement basis for financial instruments, specifically for banks. This is not correct. Financial assets are measured either at amortised cost, which is the case for most bank lending, or fair value, which typically applies to positions held for trading and some investing positions.

One cannot answer whether the question of whether fair value is appropriate without giving consideration to what is actually measured at fair value and whether this is the appropriate measurement basis for those assets. We believe IFRS’s mixed measurement model has struck the right balance of fair value and amortised cost for financial instruments. There is very little dispute that derivatives are most appropriately measured at fair value. As evidenced in JP Morgan Cazenove’s research of 44 European major banks for 2011 on average only 22% of non-derivative financial assets are measured at fair value – in other words 78% are measured at amortised cost. For the top 5 leading UK banks the split is 29% at fair value, 71% are amortised cost.

Where fair value measurement is used, it provides a timely measure of value; failure to report those values would keep investors and policy decision makers in the dark about market, credit and liquidity challenges. Fair value reporting is the best method for providing the levels of transparency that markets need to function effectively.

Critics of fair value assert that its use either caused the financial crisis or made it significantly worse. The assertion is that market values were inflated in the years preceding the crisis and artificially depressed as the crisis deepened. Thus, before the crisis the method produced unduly optimistic results and during the crisis unduly pessimistic ones; exaggerating the boom and the bust (‘pro-cyclicality’).

Good accounting is necessarily pro-cyclical, in that during the good times it reports good news and during the bad times it reports bad news. However, this is true of historical cost accounting as well as fair value. Company financial reporting is only one example of this alternating good and bad news. All indicators of economic activity tend to give similar messages, optimistic or pessimistic, depending on the prevailing mood in the economy. Nevertheless, the concern about procyclicality has some validity in that there is a feedback loop. As accounting profits increase, regulatory capital also increases, enabling increased lending and/or larger positions to be taken. With losses, the reverse occurs. To limit the accounting information would be counterproductive - such opacity merely reduces investor confidence in financial reporting (see paragraph 83); better would be a regulatory approach which is appropriately flexible in response to unrealised profits and losses. We will continue to work with the Bank of England, the FSA and its successors to develop workable solutions to these challenges.

The financial crisis did highlight that when market volatility is extreme and liquidity is limited, the determination of fair value is complex and companies claimed that the ‘current’ fair values did not reflect their expectations of the ultimate economic value of these assets. The IASB and the US Financial Accounting Standards Board (FASB) have now issued converged guidance on determining fair value and additional disclosure requirements to address fair value matters, including the approach to valuation in illiquid markets.

What does a ‘true and fair view’ really represent to the market?

It is a requirement of both UK and EU law that financial statements must give a true and fair view. Section 393 of the Companies Act 2006 (CA 2006) requires that the directors of a company must not approve financial statements unless they are satisfied they give a true and fair view. True and fair is applied in the context of existing accounting standards and the legal framework, both of which may change over time. Section 393 was introduced in response to concern from certain investors who were nervous that the introduction of IFRS may have removed the requirement for truth and fairness.

In parallel with this change, the Companies Act 2006 reintroduced the requirement for an "unfettered" opinion as to whether the financial statements gave a "true and fair view" (with a separate opinion on compliance with IFRS or UK GAAP, as the case may be). This difference was, in fact, one of semantics. As explained in our response to Q13, the auditor always had to consider a wider range of factors in forming an opinion on truth and fairness; it was simply the way in which this was reported. Between 2005 and 2008 (when the 2006 Act came into force), an auditor who did not believe financial statements gave a true and fair view despite compliance with IFRS would still have been required to qualify their opinion – this change simply made it more obvious to users of financial statements that this was the case.

The introduction of IFRS in the UK did not change the fundamental requirement for accounts to give a true and fair view. CA 2006 authorises two alternative systems of accounting: UK GAAP and IFRS. IFRS consolidated accounts are required for EU groups listed on recognised exchanges. IFRS uses the term ‘present fairly’ rather than ‘give a true and fair view’.

In a 2008 legal opinion addressed to the UK Financial Reporting Council, Martin Moore QC stated that ‘although the routes by which the requirement to give a fair presentation or a true and fair view have become embedded in relation to the financial statements of UK companies differ slightly in each case that requirement remains paramount.’ [2] Further, he notes that the requirement set out in IFRSs to present fairly ‘is not a different requirement to showing a true and fair view but is a different articulation of the same concept.’

The IASB’s Conceptual Framework uses a similar phrase, ‘faithful representation’ to describe the qualitative characteristics of financial information. To portray an economic event faithfully, information must be complete, neutral and free from error. The IASB’s objective in any situation is to maximise these characteristics to the extent possible.

Neutrality is central to many measurement issues: a neutral depiction is ‘not slanted, weighted, emphasised, de-emphasised or otherwise manipulated to increase the probability that financial information will be received favourably or unfavourably by users’.

In August 2012, the Department of Business Innovation and Skills (BIS) wrote to the chair of the House of Lords Economic Affairs Committee to state that they "remain satisfied about the legality of the conformity of IFRS with UK law." This must include the requirement in the Companies Act for directors not to approve accounts unless they are satisfied they give a true and fair view. We support the conclusion of BIS that compliance with IFRSs results in financial statements that are true and fair.

The concept of true and fair (including the requirement to go beyond simple compliance with accounting standards) and the foundation of this concept in law are therefore clear. The extent to which the market’s perception is that true and fair is a simple box-ticking exercise,is best answered by investors. To the extent there is a difference between their view and that of preparers and auditors, this is an expectation gap that should be bridged. Proposals by the FRC and the IAASB for reforming the audit report may help in this regard (see Question 17 below).

We also draw your attention to our response to Q13 below which explains how auditors are required, as a matter of auditing standards, to go beyond considering pure compliance with IFRS when deciding if a set of financial statements do indeed give a "true and fair view". A set of financial statements that does not comply with IFRS is unlikely to give a "true and fair view", a set that does is likely to, but in both cases, the auditor must form a view on wider grounds of truth and fairness.

Q11 What are your views on the current incurred-loss impairment model and its role in the banking crisis? Do you consider that proposals to move to an expected-loss model will address criticisms of the current accounting rules?

The ‘incurred loss’ method is an impairment approach for financial assets measured at amortised cost – i.e. those where the intention is to hold an asset to maturity rather than actively trading it. Examples would include mortgage lending by a retail bank. This approach discourages entities from manipulating reported net income by making provisions for future losses for which there is no objective evidence at the balance sheet date. The incurred loss model avoids companies making provision for loss events that have yet to happen. This is consistent with the general approach in IFRS and US GAAP.

We acknowledge that the British Banking Association’s Statement of Recommended Practice (SORP), which guided UK banks’ financial reporting prior to 2005, allowed for the use of a ‘general provision’ for loan losses, but it also made clear that this was for incurred but not identified losses – not for losses expected to occur in the future. The incurred loss provisioning methodology under IFRS also captures incurred but not identified losses. This is commonly referred to as ‘incurred but not reported’ (IBNR).

Thus, a bank putting aside a set percentage of all loans as a ‘general provision’ – an additional reserve to cater for some possible future event, such as an unforeseen recession, which may or may not occur at some point in the economic cycle-was not permitted before IFRSs were introduced and was not permitted following the introduction of IFRS in the UK in the 2005. Any fixed percentage provisioning was simply an attempt to estimate, for a population, a suitable specific provision. General provisions are by definition not specific and their opacity may ultimately reduce investor confidence in the financial statements. A general provision may appear to bolster the balance sheet but this is at the cost of clarity as to the results in a specific year.

The IASB and the FASB have responded to requests from the G20 Leaders and regulators that they develop an impairment model that is more forward-looking than the current incurred loss model, and so are developing a new model for loan loss provisioning incorporating ‘expected losses’. An expected loss model is conceptually different from an incurred loss model as it aims to include in the current impairment provision the impact of future loss events that have not happened but which, across the portfolio, are reasonably expected to occur; something that banks attempt to do when they price the original lending or make their investing decision.

In the crisis (as in other times) fair value captured economic/market expectations of the future, incurred loss provisioning does not. Banks that accounted for certain loans at fair value may have accelerated loss recognition at the cost of increased volatility; banks that accounted for such loans at historic cost using incurred loss will have recognised losses later and reported less volatile results. There are valid arguments in favour of both approaches. Nevertheless, the introduction of expected loss provisioning may shift the historic/amortised cost model towards the fair value approach, reducing the apparent anomaly of a "choice".

We are supportive of the efforts of the IASB and FASB to respond to the call from prudential supervisors and the G20 countries for a more forward-looking provisioning approach. We would support standards that include an expected loss model that enhances the transparency of entities’ exposure to credit risk, be capable of applying in practice and can be incorporated into a cost-based measurement approach which is the measurement basis for most bank lending.

We note that banks’ provisions are currently a matter of particular concern for regulators, including the Bank of England and FSA in the UK. Rather than seeking to impose regulatory requriements in the general purpose financial statements of banks (that are used by a wide range of stakeholders), we would recommend that any additional regulatory provisions should be dealt with in separate regulatory returns. This approach is already required by Basel II in existing returns to the FSA and publicly available Pillar III reporting which requires provisioning for one year’s worth of expected future credit losses. We note in paragraph 102 below that these could be reconciled at a summary level in the notes to the financial statements to give investors information about regulatory capital if that would be helpful to them. In this way, whatever degree of regulatory prudence thought appropriate by the regulator can be introduced whilst banks’ financial statements within their annual report continue to give the neutral picture of their activities which investors want.

Q12 What is the best method of accounting for profits and losses in trading instruments? Are there any alternatives to mark-to-market or mark-to-model that might better represent a ‘true and fair view’?

We support the accounting treatment in IAS 39 Financial Instruments: Recognition and Measurement and IFRS 9 Financial Instruments for profits and losses on ‘trading instruments’. That is, such instruments should be measured at fair value with gains and losses recognised in profit or loss in the period in which they arise.

Fair value accounting provides a timely measure of value; failure to report those values could potentially keep investors and policy decision makers in the dark.

Despite its criticisms, fair value measurement for certain assets is the best method for providing the levels of transparency that the markets need to function effectively. As noted in paragraph 63 above, most financial assets are not measured at fair value.

Q13 Did IFRS accounting standards contribute to a box-ticking culture to the exclusion of promoting transparency and a ‘true and fair view’ of the business?

No. We do not believe that IFRS accounting standards contributed to a ‘tick box’ culture to the exclusion of promoting transparency and a ‘true and fair view’ of the business.

IFRSs are principles-based. Contrary to the box-ticking mentality, they require significant exercise of professional judgement by both preparers and auditors. This is particularly true of fair value measurement and loan loss provisioning.

In deciding whether the directors have prepared financial statements that give a true and fair view, ISA (UK and Ireland) 700 (revised) The Auditor's Report on Financial Statements requires a wider consideration than a check for compliance with standards. For example, not only must the auditor consider if the accounting policies adopted comply with IFRSs, but it must also consider whether they are appropriate in the circumstances.

IFRSs have also promoted increased transparency of information. For example, IFRS 7 Financial Instruments: Disclosures (and prior to that, IAS 32 Financial Instruments: Disclosure and Presentation) requires far more information about financial assets and financial liabilities held by all companies (not only banks) than UK GAAP did prior to the adoption of IFRSs in 2005. This includes information about financial instrument risk.

As auditors, we also exercise professional judgement in the area of transparency. ISA (UK and Ireland) 700 also requires us to consider whether "the disclosures are adequate so as to enable the intended users to understand the effect of material transactions and events on the information conveyed in the financial statements."

We have given more detail on the primacy of the "true and fair view" requirement in our response to Question 10 above.

Q14. Do we need a special accounting regime for banks? If so, what should it look like?

We do not believe a special accounting regime for banks in the UK is required. Many UK banks operate on a global basis and consequently need to use a common reporting language if they are to be able to be compared to their peers. That reporting language is, with the exception of US banks, increasingly IFRS.

Accounting standards for specific industries generally lead to complexity, as they tend to dive into greater detail to reflect each nuance of particular transactions in that industry. The increased complexity can make the annual reports of banks less accessible to investors and other stakeholders. Industry-specific standards can also cause problems when dealing with diversified financial groups that include both banking and non-banking operations.

We would also be concerned if application of bank specific accounting standards led to regulatory arbitrage. Banking business should not be driven to or from non-banks (including mortgage lenders, credit card companies and other consumer credit providers) and the shadow-banking sector solely as a result of accounting differences.

Developing industry-specific guidance runs the risk of a proliferation of potentially conflicting industry-specific requirements. In the US in 2008, the US Securities and Exchange Commission (SEC) published the findings of the Pozen Committee [3] , which recommended that industry -guidance should be eliminated from US GAAP to reduce avoidable complexity. Significantly, the Pozen Report went on to recommend that the SEC should encourage the IASB to limit future industry-specific guidance.

The Basel Committee on Banking Supervision, through its Accounting Task Force, is working with banking regulators around the world to ensure international accounting standards and practices promote sound risk management at banks, support market discipline through transparency, and reinforce the safety and soundness of the banking system. To the extent IFRS financial statements need to include additional information for banks the regulatory authorities have the tools to ensure such information is provided.

It can be argued that separate accounting standards for banks could provide a useful tool to regulate a bank’s ability to remunerate staff or make distributions to shareholders. We believe that these aims are best accomplished through existing tools such as the FSA’s Remuneration Code and prudential capital requirements.

We note that listed UK banks would still be required to produce audited IFRS financial statements, because they would continue to be subject to the requirements of the IAS Regulation (1606/2002/EC). Consequently, a ‘special accounting regime’ in the UK would be a supplementary, regulatory-based regime, which could lead to confusion with investors.

Q15. Are there any interim measures (such as mandatory disclosure) which could be introduced in the meantime?

As we do not support a separate reporting regime for banks, any ‘interim measures’ would likely be either specific reporting to the regulator or additional mandatory disclosure in the financial statements. However, these need to be targeted, preferably by the appropriate regulator, on specific areas of banking activity.

The recent report to the Financial Stability Board from the Enhanced Disclosure Task Force, ‘Enhancing the risk disclosures of banks’ highlights ways in which banks’ financial reports can be improved with respect to transparency around impairment, financial instruments generally, and the relationship between financial statement capital and regulatory capital requirements. We participated in this initiative and are supporting banks in implementing the recommendations. Investors had a key role in developing these recommendations and we support the coordinated implementation of the recommendations. Our recent briefing paper ‘Promoting Stability’ provides more detail on the EDTF’s work.

The European Banking Authority has proposed a regulatory return that reconciles shareholders’ funds as reported under IFRS to the regulatory capital resources. IAS 1 Presentation of Financial Statements already requires the financial statements to include information on capital management which includes how externally imposed regulatory capital limits are managed. This could be expanded to include a quantitative summary of the reconciliation between regulatory capital and the balance sheet to help investors understand the impact of regulation on the banks they have invested in. This would, however, indirectly bring the calculation of regulatory capital within the scope of the audit (see our response to Question 16.3 below).

Q16. How likely are current proposals for improving disclosure and dialogue to be successful (with particular reference to discussion papers issued by FSA/FRC)?

We believe that the FSA’s Code of Practice for the Relationship Between the External Auditor and the Supervisor is already working well and resulting in a better exchange of information between auditors and supervisors, to the mutual benefit of both. We remain committed to working with the Bank of England, the FSA and its successors to effective dialogue and, if appropriate, further improvements to the Code of Practice.

Q16.1 Should there be enhanced powers to better align auditors’ incentives with those of the regulator?

We do not believe that there is currently a mismatch of incentives. Auditors are required by law to raise matters of regulatory concern with regulators. The Code requires regulators to discuss matters with the auditor relevant to their audit. It is in the interests of both that this exchange be as full and frank as necessary.

Q16.2 Should auditors of banks be obliged to have a primary responsibility to the regulator, rather than the client?

No. The primary responsibility of the auditor is to the shareholders, with the audit committee acting as their proxy where appropriate. The existing statutory mechanism under FSMA (and the further changes that are proposed in the Financial Services Bill) already puts the onus to report matters of regulatory interest to the regulator, including, where necessary, the obligation to do so without informing the audited entity’s management. This would include, for example, situations where an auditor feels that there may be a material uncertainty as to an entity’s ability to continue as a going concern. This gives the regulator time for further discussion with the entity’s management, the auditor, and other players including HM Treasury and potential action, before the audit report is finalised.

Q16.3 Should regulatory return be audited?

We believe that the FSA (and its successors) should be able to take a risk-based approach to the audit of regulatory returns. With increasing complexity, returns will become harder to complete, with increased potential for error. External assurance may give the regulator increased confidence that they are taking regulatory decisions based on accurate information. However, for more straightforward returns and/or simpler banks, the potential for error may be lower and the benefit of assurance may not outweigh the cost. The FSA already has the power to request assurance on a regulatory return under the s166 skilled persons’ regime. We would recommend that the FSA (and its successors) formalise their approach by identifying risk factors that may suggest assurance would be beneficial. This could include complexity of a bank’s operations, headroom in capital adequacy, complexity of the return, and potentially a cyclical approach – riskier returns being assured more often, less risky ones less often, low risk returns not being assured at all.

Q17 Is there a problem arising from the difficulty of qualifying the accounts of a bank? Should auditors be able to ‘grade’ accounts – from AAA down? What would be the effect of this?

The consequences of a qualification or emphasis of matter with the auditors’ report are difficult, as this may precipitate a run on the bank. It is partly for this reason the Financial Services and Markets Act 2000 (FSMA) requires auditors to discuss such matters with the FSA if they are considering a qualification or believe that a bank may not be a going concern. This allows regulatory action to be taken, if necessary, practicable and appropriate, which often results in the resolution of the matter and hence the qualification or emphasis of matter is not needed.

The Sharman report Going Concern and Liquidity Risks: Lessons for Companies and Auditors examined this issue and concluded "liquidity support from central banks may be a normal funding source for a bank and therefore reliance on such support if reasonably assured, does not mean that the bank is not a going concern or that material uncertainty disclosures or an emphasis of matter paragraph are required."

In evidence to the House of Lords Economic Affairs Committee [4] we confirmed that, together with other audit firms, we held discussions with the FSA and with the then City Minister around the availability of government and Bank of England support in concluding on our going concern judgments for banks in 2008. The financial statements of the banks in question contained disclosure of the potential need for support, support was available, and hence there was no material uncertainty relating to going concern as defined in accounting and auditing standards. In another situation, we discussed matters with the FSA and confirmed that we would be unable to issue an audit report without an emphasis of matter (or qualification if the material uncertainty was not disclosed). That institution subsequently collapsed.

We observe that the ‘AAA’-style grading referred to in Question 17 has similarities to that used by credit ratings agencies. They have a particular role to play in financial markets, but the role of auditors is different, and we would not wish them to become confused with similar sounding ratings issued by auditors.

FRC developments

We believe that the recent changes introduced by the Financial Reporting Council (FRC) including its revised UK Corporate Governance Code, supported by the revised Guidance for Audit Committees and revisions to International Standards on Auditing (UK and Ireland), provide a better approach to providing shareholders with better information. The revised Code requires audit committees to describe in the annual report the significant judgements that they have formed in approving the financial statements. If the auditor disagrees with such a description, then they have to both report their disagreement and provide the missing information. These changes apply for financial years commencing on or after 1 October 2012.

We support the FRC’s changes as providing a proportionate response. They allows users of the financial statements to understand the significant judgements taken, which may include not only going concern but (for example) asset valuations, decisions on consolidation (or otherwise) of special purpose entities and the level of loan loss provisioning.

We believe that this will be more informative and useful for stakeholders than any specific grading scheme. For example, how uncertain does a valuation need to be to move from AA to A? Users’ attention will be drawn by the audit committee who will report in one place on the most significant judgements in the financial statements. And the sanction that, if the auditor disagrees with that summary, they will not only mention this in their report but also provide the corrected or missing disclosure provides users with comfort that this has been done properly.

IAASB developments

The International Audit and Assurance Standards Board (IAASB) have recently consulted on changes to audit reports. In the papers for their December 2012 meeting, they analysed the feedback received from a wide range of stakeholders. There was strong support from all quarters for the existing pass/fail requirement of the audit report (i.e. a report is qualified or not). There were no significant calls for a graded audit report from any group of stakeholders, including investors and regulators.

There was, however, support for auditor commentary within the auditor’s report to provide either signposting of important matters in the accounts and information about what the auditor has done in response. Almost all respondents expressing support did, however, express a concern that there should not be rival sources of information about the company being audited – management and directors provide information about the company, with auditors expressing an opinion on that information.

These proposals are worthy of further development. In particular, continued discussion with a wide range of stakeholders as to the matters that they would wish auditors to comment on is essential. With specific reference to banks, the European Banking Authority reported that they would like to see the company continue to report significant matters about the company, the audit committee to report on their deliberations on those matters, and the auditor to report on how those matters have been addressed in their audit.

Auditor reporting of going concern was the other significant area which the IAASB considered at their December meeting. This is an area where further harmonisation work may be needed given the lack of consistency between governance and accounting (as opposed to auditing) regimes in this area. Again, the principle should be that management and those charged with governance give information first, with the auditor forming an opinion on it. The UK is unusual in requiring management to make a statement around going concern in circumstances where there is no significant uncertainty; further work globally on improving reporting would be useful.

Finally, the IAASB has proposed some expansion of the explanation of the scope of an audit. This could be an appropriate place to clarify the auditors’ and directors’ respective responsibilities as to the truth and fairness of the financial statements, including their consideration of matters beyond compliance with the applicable accounting standards.

Q18 Should the scope of audit be widened so that auditors can better express a broader view of the business? For example should auditors comment specifically on issues such as remuneration policy, valuation models or risk?

Is assurance appropriate in all cases?

We agree that, in suitable cases, additional independent assurance can provide valuable confidence to users of information that the information that they are using is reliable. As ever, there will be a need to consider costs and benefits.

The costs of additional assurance will vary depending on, amongst other things, the degree to which an auditor already understands the underlying systems and controls used to prepare the information being assured. The benefits of additional assurance will depend on the degree of risk of material misstatement of that information, and the extent to which that information is important to decisions being taken by a user.

These factors need to be balanced. As set out in our response to question Q16.3 above, we support the FSA’s skilled person’s approach whereby they can ask for assurance when they feel it will be beneficial, without imposing it as a blanket requirement for all regulated entities. Any detailed proposals to amend this approach would benefit from a proper cost benefit analysis.

What conditions must be met for assurance to be possible?

In order for auditors to express an opinion on a subject matter, that subject matter would need to be objectively verifiable against suitable criteria. The characteristics for suitable criteria in the International Framework for Assurance Engagements are relevance, understandability, neutrality, reliability and completeness.

Auditors should not give opinions on subjective matters. Such opinions are likely to be of little value to stakeholders – and could potentially be perceived as harming auditors’ independence as they could be seen as more likely to give a favourable opinion if it could not be demonstrated to be objectively wrong.

We believe, however, that the developments we alluded to in our response to Q17 above would provide additional value from the audit. This is best illustrated by considering the three examples in the question.

On risk, it would be wrong for the auditor to form an opinion as to "is this risk appropriate for a bank to be taking?" The answer to this is inherently a subjective question depending on the risk appetite of the shareholders, the amount of headroom in regulatory capital, and the likelihood and potential impact of that risk. Different stakeholders will have different views. However, the auditor already forms a view on detailed, objective risk information included in the notes to the financial statements – for example, disclosure in the accounts as to the potential change in net assets if foreign exchange rates moved within a reasonably probable range.

On valuation techniques, the auditor already has to form a view as to whether the value determined by that model is within an acceptable range, and would qualify their opinion if it were not. Equally, if the valuation models were not as described, they would qualify their opinion. We believe that the current FRC changes will be helpful as, if the choice of valuation model is significant, the audit committee will need to comment on it (and the auditor will check that this comment is appropriate).

Remuneration policy is a similar issue to risk – it is for shareholders to decide whether the policy for the pay of directors is appropriate, and the current BIS proposals to improve governance and disclosure in this area will be helpful. The auditor can confirm whether the amounts actually paid are as disclosed. The auditor could, theoretically, also comment as to whether the policy had been corrected applied to determine the amount paid, provided that the inputs into that policy are objectively verifiable. So, for example, a policy that pays a fixed proportion of profit in year 1, with a further proportion if bad debts have not exceeded a certain percentage, is objectively verifiable. A policy that pays out for making "good" investments would not be verifiable.

Q19 What would be the effect of using return on assets as a performance measure in banks, as opposed to return on equity?

This is a question best directed to investors.

Q20 Are the amendments to the Financial Services and Markets Act 2000 regarding dialogue between regulator and auditor sufficient, or does further work need to be done in this area?

We think the recent amendments to the Financial Services and Markets Act (FSMA) 2000 made by the Financial Services Act 2012 are sufficient. We note that the FSA’s Code of Practice for the Relationship Between the External Auditor and the Supervisor will also support better working relationships. We welcome the fact that the amended FSMA not only puts the existing Code on a firm statutory footing, but also requires the Prudential Regulation Authority (PRA) to make a Code regarding its obligations to disclose relevant information to auditors (where permitted) and for the exchange of opinions between auditors and the PRA. This supports the two-way nature of the exchange. Further work will be needed to develop reciprocal relationships of trust, but this is not work of a legislative nature and is best dealt with in Codes which can be flexed as circumstances change. Concerns such Codes are less robust than legislation are ill-founded, as the revised FSMA allows both the PRA and the Financial Conduct Authority (FCA) to impose penalties and disqualify auditors for non-compliance with such Codes. With other members of the accountancy profession, we are also working with the Financial Stability Board to develop closer relations between auditors and regulators.

4 January 2013