The Effect of Information Asymmetry on Financial
Statement Disclosure Quality by Banks in the European
Union
Marco Bogaard 0569445 May 21, 2014
MSc in Accountancy & Control
Accounting Section, Amsterdam Business School, University of Amsterdam First supervisor: dr. S.W. Bissessur
Abstract
Will banks increase their financial statement disclosure quality to reduce information asymmetry? I will research this question by comparing disclosure quality in a time of relative high information asymmetry between banks and their stakeholders which is found during the financial crisis in 2008, and a time of relative low perceived information asymmetry in 2007. Due to this difference in information asymmetry between banks and stakeholders I hypothesize that banks provided higher disclosure quality in 2008 than in 2007. Disclosure quality is measured in terms of compliance with mandatory disclosure requirements. As assets and liabilities of banks consist for 90% of financial instruments, IFRS 7 Financial Instruments: Disclosure captures most disclosure requirements for banks and therefore compliance with this standard is used as a proxy for disclosure quality. The standard was mandatory in the European Union (EU) from 01-01-2007 but could be applied already in 2005 and 2006. This gives the opportunity to also examine a learning effect for complying with the standard. It is important to examine this potential learning effect as it can have an effect on the explanation of the observed results. Another feature of this paper is the examination of the effect that a government bail-out had on the disclosure quality. As, after bankruptcy, a government bail-out is the most concrete signal of financial distress, the paper hypothesizes that bailed-out banks had a higher increase in disclosure quality than banks which were not bailed-out. The results show significant higher disclosure scores in 2008 compared to 2007 for the complete sample of 30 EU listed banks. This implies that information asymmetry has a positive effect on disclosure quality. Furthermore, the results show no evidence that both government bail-outs and early adopting IFRS 7 have a significant effect on disclosure quality between 2007 and 2008.
Table of contents
1 Introduction ... 4
2 Theoretical background, disclosure requirements of IFRS 7 ... 9
3 Relationship between information asymmetry and disclosure ... 14
3.1 Information asymmetry ... 14
3.1.1 Adverse selection ... 14
3.1.2 Moral hazard ... 16
3.1.2.1 Moral hazard in private markets ... 17
3.1.2.2 Moral hazard in public markets ... 18
3.1.3 Agency problems ... 20
3.2 Disclosure ... 22
3.2.1 Voluntary disclosure ... 22
3.2.1.1 The effect of visibility on voluntary disclosure ... 22
3.2.1.2 Incentives for voluntary disclosure ... 23
3.2.2 Mandatory disclosure ... 26
4 Financial crisis effects on disclosure ... 29
5 Research method ... 33
6 Research design ... 37
7 Research results ... 39
7.1 2007 versus 2008 compliance (complete sample) ... 39
7.2 2007 versus 2008 compliance bailed-out banks ... 40
7.3 2007 versus 2008 compliance early adopters ... 42
7.4 Results overview ... 45
8 Conclusion ... 47
9 APPENDIX A: IFRS 7 compliance checklist for both 2007 and 2008 (for additional 2008 requirements see Appendix B) ... 52
10 APPENDIX B: Additional IFRS 7 requirements for the 2008 compliance checklist ... 64
1 Introduction
The purpose of this research is to examine the effect of information asymmetry on disclosure quality provided in the financial statements of banks in the European Union (EU). High disclosure quality is crucial for users of the financial statements to understand and interpret the presented balance sheet and profit and loss figures. As Healy and Palepu (2001) point out, information asymmetry impedes the efficient allocation of resources in a capital market economy. Researching the link between disclosure quality and information asymmetry reduction is particularly interesting because it, as Verrechia (2001) concludes, links disclosure to efficiency, and in this sense provides an economic rationale for the utility of financial reporting. Following Ball (2001), decreasing information asymmetry between banks and stakeholders will even increase the development of countries, increase the development of banks and increase the efficiency of public securities markets.
One way to reduce information asymmetry is by providing higher disclosure quality (e.g. Mishkin, 2006; Daske et al., 2008). In line with this theory my research question is the following: ‘Will banks increase their financial statement disclosure quality to reduce information asymmetry?’
The setting for my research is the EU prior and during the financial crisis of 2008. This financial crisis revealed a major information asymmetry between banks and their stakeholders. In line with the theory provided by Akerlof (1970), this information asymmetry about product quality resulted in quality uncertainty, which in turn resulted in a market malfunction, in this case particularly poor functioning credit markets (Dwyer and Tkac, 2009) and a global equity market collapse (Bartram and Bodnar, 2009). More concrete, due to this unusually severe financial crisis most banks lost trust in each other, suffered huge losses, some banks defaulted, others were acquired by competitors and many were bailed-out by their national government.
While the first signs of the financial crisis appeared in stress measures as the TED spread1 at the start of 2007, and as more problems started to become evident in
1 The TED spread is the difference between the interest rate of interbank loans and the interest rate of
T-bills. T-bills are seen as the lowest risk investments. Therefore, a relative high TED spread indicates that lenders of interbank loans expect that banks are exposed to more risk.
the second half of 2007, there was no big impact of the crisis on financial instruments disclosures in the financial statements of 2007. This has two reasons. One reason is that banks themselves maybe were not aware of the severity of the risks they faced; the financial instruments were highly complex and had received high ratings from rating agencies. Another reason is that stakeholders (among others, shareholders) certainly were not aware of the risks banks faced in 2007. This is evident from fact that the equity market deterioration started around mid-2008. This resulted in a situation where banks could have withheld information in 2007 without getting punished by decreasing stock value. This would have led to lower bonuses and pensions for the banks’ management and therefore it was not desirable for management to disclose negative information.
In contrast to 2007, the financial crisis had a major impact on banks in 2008. This is evident from the TED spread which peaked in mid-2008, the equity market deterioration around 2008 and the increased public attention for banks as of mid-20082. This awareness of banks and outsiders about the fact that banks suffered from the risks they took can have an impact on the financial instruments disclosures in the financial statements of 2008. Furthermore, banks may want to reduce the information asymmetry to obtain loans, reduce interest rates on these loans and to increase their stock value. For the years after 2008 no further incentives related to the financial crisis existed to disclose more and better information as stock value already recovered from their lowest point during the first months of 2009. Therefore, the research will only relate to the years in which there was a change in incentives which could lead to a change in disclosure quality, thus between the years 2007 and 2008.
2 By reviewing articles written during 2007 and 2008 in the biggest Dutch financial newspaper the
Financieel Dagblad (FD) I note the following. 10% Of the FD papers included at least one article with negative news about the financial crisis in 2007. Over the last quarter of 2007 this was 17%. In 2008 40% of the editions included at least one article with negative news about the financial crisis. Over the last half year of 2008 this increased to 60%. In 2007 and the first half year of 2008 the general belief in these articles showed that the crisis was expected to be only temporary and that the EU would not suffer that much from the crisis. In the second half of 2008 the focus of the articles shifted to the banks and showed that the EU also would suffer and that the crisis will remain until at least 2010.
The products that caused the financial turmoil for banks were financial instruments based upon mortgages. With hindsight it became clear that mortgage providers in the United States had taken too much risk in making mortgages to borrowers who are less creditworthy than prime borrowers. These mortgages were heavily securitized and subsequently packed into collateralized debt obligations (CDO’s) by investment banks. Retail banks around the world, including banks in the EU, bought these financial instruments. Which were prior to 2007 seen as low risk products, mainly due to their high credit ratings. Dwyer and Tkac (2009) argue that when the subprime mortgages began to experience higher default rates than predicted, financial institutions found it difficult to value these financial instruments. This quality uncertainty led to more counterparty risk, which led to a decrease in trading that was not confined to the mortgage-backed CDO market and spilled over into other asset markets and into funding markets (Dwyer and Tkac, 2009).
Banks, which assets and liabilities are for 90% classified as financial instruments (Bisschof, 2009), can reduce the information asymmetry by providing higher quality financial instrument disclosure in their financial statements. On the other side, banks have two main incentives to not increase their disclosure quality about financial instruments. First they can face proprietary costs (Healy and Palepu, 2001), these are costs which arise if competitors use the disclosed information in a way that is harmful to the reporting bank. Secondly, as banks have interests which are not in line with the interest of stakeholders, banks want to take more risk than stakeholders (e.g. Verrecchia, 1983; Healy and Palepu, 2001).
That banks do not optimally comply with mandatory disclosure standards is supported by prior compliance research (e.g. Street et. al., 1999; Glaum and Street, 2003; Hodgeon and Wallace, 2008; AFM, 2009; CESR, 2009; Wohlmannstetter et. al., 2009). This non-compliance by banks is possible because of weak regulatory enforcement (e.g. FEE, 2002; Brown and Tarca, 2005; Dao, 2005; Larosière, 2009; Preiato and Brown, 2009). More specifically Ball et. al. (2001) make a distinction between areas of high regulatory enforcement and areas with low regulatory enforcement, respectively common-law and code-law countries. Most EU countries can be seen as code-law. Relative low regulatory enforcement in the EU, for example, means that managers and auditors in the EU face less litigation risk than managers and auditors in common-law countries. This means that especially in the EU (above
for example the United States and Canada) non-compliance with mandatory disclosure standards is possible.
How information about financial instruments has to be disclosed is stated in IFRS 7 Financial Instruments: Disclosure. As of January 1, 2007 this standard has become mandatory for EU stock listed firms which are engaged in financial instruments. The presence of both this mandatory standard and the financial crisis gives a good opportunity to research the relationship between information asymmetry and disclosure in the EU banking sector.
To test if banks increase their financial statement disclosure quality to reduce information asymmetry, and in this way research the effect of information asymmetry changes on disclosure quality provided by EU banks, I use a proxy which measures the compliance with IFRS 7. As IFRS disclosure requirements are produced by knowledgeable parties and are adopted by many countries across the world, all EU countries included, IFRS disclosure requirements are generally seen as one of the highest quality standards available. Furthermore, measuring compliance with IFRS 7 allows examining 90% of banks’ assets and liabilities disclosures which are prepared under an identical mandatory standard. An increase in compliance with IFRS 7 can thus be seen as an increase in disclosure quality, and a decrease in compliance with IFRS 7 can be seen as a decrease in disclosure quality. IFRS 7 compliance is measured by using an IFRS 7 checklist generally available on the IAS Plus website, and is attached in Appendix A and Appendix B. The checklist covers the full range of the IFRS 7 disclosure standard.
This proxy is preferred above a word/page count proxy which is used by Bisschof (2009) who examined the effect of IFRS 7 adoption on European banks disclosure, mainly because the method of Bisschof (2009) captures a change in disclosure quantity and not necessarily a change in disclosure quality. An increase in quantity can even result in a decrease of quality as it can overload the user with irrelevant information. This overload will decrease understandability and relevance of the disclosed information while the proxy will measure the opposite.
The IFRS 7 compliance proxy used in this thesis is also preferred above a self-constructed disclosure index which is used by Gallery et. al. (2008). They examined the effect of adopting the Australian equivalent of IFRS (AIFRS) on financial statement disclosure quality of large Australian corporations. The method of Gallery et. al. (2008) has the risk of being highly subjective and in that way does not
necessarily capture the generally accepted quality standards: understandability, relevance, reliability and comparability.
The IFRS 7 proxy is used on a sample of 30 EU stock listed retail banks for the years 2007 and 2008. This sample allows to control as much as possible for factors that, in accordance to prior research, may have an influence on the researched effect of information asymmetry on disclosure quality. As the research involves reviewing the scores for the same banks in 2007 and in 2008, I will conduct a paired-sample t-test to compare the 2007 disclosure score and the 2008 disclosure score. Furthermore I used a two-sample t-test to research the effect of government bail-outs and early adopting IFRS 7 on disclosure quality of bank in the EU.
To my knowledge no prior detailed IFRS 7 research using the full IFRS 7 scope has been conducted which examines if and how banks in the EU change the disclosure quality of their financial statements in a time when they face enormous negative consequences of information asymmetry between them and their stakeholders. By researching the relation between disclosure quality and information asymmetry this paper addresses an issue which deserves greater attention in the accounting literature (Verrecchia, 2001), because it is a relatively less developed topic (Frolov, 2007) and of which our understanding is limited (Healy and Palepu, 2001).A more practical contribution concerns the value of this research for investors. As information contained in the disclosures is key to understand a company’s financial position and performance and its omission might consequently affect the ability of investors to make decisions regarding their investment (CESR, 2009). Furthermore, the research can provide insight in the functioning of the EU enforcement environment which is seen as of low quality in prior enforcement and compliance literature.
The remainder of this paper is organized as follows. The next section provides a theoretical background of IFRS 7. The third section discusses the relationship between information asymmetry and disclosure. The fourth section addresses the effects of the financial crisis on disclosures by banks and builds hypotheses. In the fifth section the research method is discussed and in section six the research design. In the seventh section I will present and discuss the results. My thesis ends with a conclusion.
2 Theoretical background, disclosure requirements of IFRS 7
In August 2005, the IASB issued IFRS 7 Financial Instruments: Disclosures. IFRS 7 is effective for annual periods beginning on or after 1 January 2007, and supersedes both IAS 30 Disclosures in the Financial Statement of Banks and Similar Financial Institutions and the disclosure requirements of IAS 32 Financial Instruments: Disclosure and Presentation. After the introduction of IFRS 7, IAS 32 was renamed IAS 32 Financial Instruments: Presentation. Because the IASB already issued IFRS 7 in 2005 it was possible for firms to apply the standard in the financial statements of 2005 and later. Firms which choose to apply the standard in the 2005 and/or 2006 financial statements are called ‘early-adopters’.
IFRS 7 is applicable to all entities and to all risks arising from all financial instruments3, whether recognized or unrecognized. Recognized financial instruments include financial assets and financial liabilities that are within the scope of IAS 39 Financial Instruments: Recognition and Measurement. Unrecognized financial instruments include some financial instruments that, although outside the scope of IAS 39, are within the scope of IFRS 7 (IFRS 7.4).
Following IFRS 7.6, these financial instruments have to be divided in ‘classes of financial instruments’. The reporting entity is required to group financial instruments into classes that are appropriate to the nature of the information disclosed and that take into account the characteristics of those financial instruments. The classes are determined by the entity and are distinct from the categories of financial instruments specified in IAS 39 (IFRS 7.B1). The classes are distinguished between instruments measured at amortized cost from those measured at fair value, financial instruments outside the scope of IFRS 7 have to be treated separately (IFRS 7.B2). And an entity has to decide, in light of its circumstances, how much detail it
3 Except for the following financial instruments: ‘Interest in subsidiaries, associates and joint ventures
accounted for in accordance with IAS 27, IAS 28 or IAS 31 respectively’, ‘Construction contract receivables accounted for under IAS 11 Construction Contracts’, ‘Own equity shares’ (exept for the requirements on derecognition and embedded derivatives), ‘Derivatives over own equity shares that are classified as equity, including derivatives over shares in non-controlling interest in the consolidated financial statements’, ‘Employer’s rights and obligations under employee benefit plans covered by IAS 19’, ‘Share based payment arrangements covered by IFRS 2’.
provides to satisfy the requirements of IFRS 7, how much emphasis it places on different aspects of the requirements and how it aggregates information with different characteristics. In order to keep the disclosures understandable and relevant it is necessary to strike a balance between overburdening financial statements with excessive detail that may not assist users of financial statements and obscuring important information as a result of too much aggregation (IFRS 7.B3). These additions show that although the requirements are mandatory, parts of these requirements are subject to the discretion of management and thus also are subject to disclosure incentives. These incentives will be discussed in section 3.2.
There are two key objectives of IFRS 7, the first one is to require entities to provide disclosures for the statement of financial position, statement of comprehensive income and equity that enable users to evaluate the significance of financial instruments for the entity’s financial performance (IFRS 7.1(a)). More specifically, these disclosures are intended to assist users in understanding the extent to which accounting policies affect the amounts at which financial assets and financial liabilities are recognized (Poole and Spooner, 2009), such as assumptions4 made in determining fair values (Bischof, 2009). Together with the disclosures of gains and losses by category of financial instrument5, the disclosure of the carrying amounts for each category of financial instrument allows users to appraise management on its decisions to buy, sell or hold financial assets and to incur, maintain or discharge financial liabilities (Poole and Spooner, 2009). These disclosures thus also serve the principals’ monitor needs which will be discussed in section 3.1.
4 An entity shall disclose whether fair values are determined, (a) in whole or in part, directly by
reference to published price quotations in an active market or are estimated using a valuation technique, (b) whether the fair values recognized or disclosed in the financial statements are determined in whole or in part using a valuation technique based on assumptions that are not supported by prices from observable current market transactions in the same instrument (i.e. without modification or repackaging) and not based available observable market data (IFRS 7.27).
5 The categories of financial instruments are: (a) financial assets at fair value through profit or loss,
showing separately (i) those designated as such upon initial recognition and (ii) those classified as held for trading, (b) held-to-maturity investments, (c) loans and receivables, (d) available-for-sale financial assets, (e) financial liabilities at fair value through profit and loss, showing separately (i) those designated as such upon initial recognition and (ii) those classified as held for trading, (f) financial liabilities measured at amortized cost (IFRS 7.8).
The second objective is to require entities to provide disclosures in their financial statements that enable users to evaluate the nature and extent of risks arising from financial instruments to which the entity is exposed during the period and at the end of the reporting period, and how the entity manages those risks (IFRS 7.1(b)). Both qualitative and quantitative disclosures are required about the risks that arise from financial instruments and how these risks have been managed. These risks typically include, but are not limited to, credit risk, liquidity risk and market risk (IFRS 7.32). Bischof (2009) explains these three risks as follows. Credit risk is the risk of defaults in payments to be received by customers and is thus a function of the customers’ credit quality. Liquidity risk arises from maturity gaps in an entity’s asset and liability management if obligations to be serviced exceed the entity’s current liquidity. Market risk refers to the entity’s exposure to fluctuations in market prices and comprises three types of risk: currency risk, interest rate risk and other price risk. All these risks played a major role in the financial crisis. The financial products based on mortgages provided to home owners were sensitive to credit risks, especially because many mortgages were provided to less creditworthy people. These mortgages could be provided by the mortgage agencies as they could sell these products with high profits to investment banks. The investment banks could pack the mortgages in a way they received high credit ratings by rating agencies (e.g. Standard & Poors and Moody’s), and subsequently sell them to the commercial banks which then received a seamless risk free stream of mortgage payments. When the credit risk of these products appeared to be higher than disclosed by all relevant parties, the value of these products decreased. This sudden decrease led to the breaching of covenants and freezing credit markets which resulted in liquidity risks for the banks. As the demand for liquidity increased due to the liquidity risk and the supply of liquidity decreased due to mistrust that resulted from information asymmetry, the interest rates on liquidity increased. This caused the market risk to increase. The latter directly influences the future profitability of banks and therefore has a direct negative influence on the stock value of banks. Credit risk, liquidity risk and market risk are all interrelated, and were influenced by the financial crisis. Proper disclosure of the risks will lower the information asymmetry which will in turn lower the actual risks.
An entity has to disclose the following qualitative information for each type of risk arising from financial instruments: (a) the exposures to risk and how they arise, (b) its objectives, policies and processes for managing the risk and the methods used
to measure the risk, (c) any changes in (a) or (b) from the previous period (IFRS 7.33). Disclosure may be provided on a gross basis or net of any risk transfer and other risk mitigating transactions. Such information is useful since it highlights the relationship between financial instruments and provides the user of the financial statements with information to understand the effect that those relationships have on the nature, timing and uncertainty of future cash flows (Poole and Spooner, 2009).
An entity has to disclose the following quantitative information for each type of risk arising from financial instruments. A summary about its exposure to that risk at the end of the reporting period6, based on information reported internally to key management personnel7 (IFRS 7.34(a)). Where more than one method is used to manage and report information about risk exposures, the disclosure of the method that provides the most relevant and reliable information shall be disclosed (IFRS 7.B7). Disclosure based on management information provides a useful insight into how risk is viewed and managed by the entity, is based on information that has a more predictive value than information based on assumptions and methods that management does not use, provides disclosures which adapt to changes in the manner in which risk is measured and managed and allows users to use the same data that management uses to measure and manage risk (Poole and Spooner, 2009). Furthermore, disclosures of material credit, liquidity and market risk which are not covered by disclosures based on information provided to key management have to be provided (IFRS 7.34(b)). And disclosure of concentrations of risk8 are mandatory. The identification of concentration of risk requires judgment taking into account the circumstances of the entity (IFRS 7.B8).
6 If the quantitative data disclosed as at the end of the reporting period are unrepresentative of an
entity’s exposure to risk during the period, an entity shall provide further information that is representative (IFRS 7.35).
7 Key management personnel are defined as: those persons having authority and responsibility for
planning, directing and controlling the activities of the entity, directly or indirectly, including any director (whether executive or otherwise) of that entity (IAS 24.9).
8 Concentrations of risk arise from financial instruments that have similar characteristics and are
In sum, the objectives and corresponding requirements of IFRS 7 are in place to reduce the information asymmetry between firms and users of the financial statements. Although the requirements are mandatory there is much implementation discretion left for management in disclosure of financial information, this can lead to non-compliance and fluctuations in compliance over time.
3 Relationship between information asymmetry and disclosure
This chapter describes the relationship between information asymmetry and disclosure. The prior sections already indicated that disclosures can decrease information asymmetry and that full compliance with mandatory standards is not always the case. By addressing ‘adverse selection’, ‘moral hazard’ and ‘agency problems’ this chapter in the first place describes in depth why and how disclosure can decrease information asymmetry. Secondly, this chapter describes the incentives managers have to not disclose all private information and thus not fully comply with IFRS 7. This is important as without knowledge about these incentives one would expect that within the legal environment and with audited financial statements banks fully comply with mandatory audit standards. Again, this is not the case as banks do not necessarily disclose all relevant items.
By addressing why and how disclosure can decrease information asymmetry, and by addressing incentives managers have to not fully comply with IFRS 7, a theoretical basis is provided for the reasoning behind the research question.
3.1 Information asymmetry
Information asymmetry is described by Mishkin (2006) as a situation that arises when one party’s insufficient knowledge about the other party involved in a transaction makes it impossible to make accurate decisions when conducting the transaction. This problem therefore impedes the efficient allocation of resources in a capital market economy (Healy and Palepu, 2001). Information asymmetry introduces adverse selection, moral hazard and agency problems.
The next three sections describe the theory behind information asymmetry by addressing adverse selection, moral hazard and agency problems in relation to the banking sector.
3.1.1 Adverse selection
Adverse selection occurs before the transaction and describes a situation that arises when an uninformed party fears a loss when conducting a transaction with an informed party (Mishkin, 2006). A loss is feared because the uninformed party is uncertain about the quality offered by the informed party. Hail and Leuz (2007)
explain that as a result the uninformed party lowers the price at which he is willing to buy, and increases the price at which he is willing to sell to protect against the losses from trading with the informed counter party. This mechanism of price protection will lead to a bid-ask spread which decreases the number of transactions which can cause a market to malfunction and even fail.
When banks initiate debt contracts (i.e. borrowers pay a fixed amount of money to the banks at periodic intervals) in the private market they suffer from adverse selection. Parties who are the most likely to produce an undesirable outcome (i.e. potential bad credit risks for banks) are the ones most likely to want to engage in a transaction because they know they are most likely unable to pay back the loan (Mishkin, 2006). Therefore, banks have to be informed about the quality offered by the borrower to separate bad credit risk from good credit risk and by that maximizing utility. In the next paragraph I will use the reasoning of Akerlof (1970) from his article about quality uncertainty in the used-car market to see what the market consequences are if the bank stays uninformed about the quality offered by the borrower.
If banks cannot distinguish between low and high credit risk they demand an interest rate that is the average of the interest rate demanded for low credit risk and the interest rate demanded for high credit risk. Parties with low credit risks are unwilling to borrow at that average interest rate and thus do not participate in these transactions, while parties with high credit risks are still willing to borrow. Because there are now less low credit risks in the market the average interest rate demanded by banks will increase, and again parties who know that this increased interest rate does not suite their credit risk would not enter into these transactions. The average interest rate demanded will again increase and the previous discussed steps will repeat. In this case there would always remain a gap between the interest rate asked by banks and the interest rate bid by low credit risk borrowers (i.e. sufficient borrowers). As uninformed banks, prior to the transaction to be entered into, are aware that the interest rate they would ask only attracts insufficient borrowers, no transactions will be made. To mitigate the adverse selection problem that banks are suffering in this private debt market, banks have become experts in producing information about borrowers so that they can distinguish low credit risks from high ones (Mishkin, 2006).
Lenders who initiate debt contracts and investors who initiate equity contracts (i.e. claims to a share in the profits and assets of a bank) with banks in the public market also suffer from the adverse selection problem. Just as in the private market the adverse selection problem will prevent the establishment of contracts, and thus transactions, due to the bid-ask gap between parties. The difference of the adverse selection problem between the private market and the public market lies in the mitigation of this problem.
Mishkin (2006) points out that in the public market lenders and investors are not willing to pay for quality information about banks (e.g. become an expert in producing quality information or buy information from experts) to mitigate the adverse selection problem because in this market they face the free-rider problem. This problem occurs when parties which do not pay for information take advantage of the information that other parties have paid for (Mishkin, 2006). So in the case where a lender or investor would research the quality of a security and subsequently invests in it because they believe the security is underpriced, other investors will follow immediately which drives up the price of the security to the security’s true value, then the profits from producing quality information disappears.
If banks want to raise money in the public markets they are therefore the ones which have to mitigate the adverse selection problem. Banks can do this by disclosing honest information about themselves so lenders and investors can determine the quality of the banks (i.e. value of investments, credit risk, liquidity risk, market risk). Investors and lenders need this information about the quality of banks to determine proper transaction prices and in this way increase the value of their investment. This higher investment value makes is easier for banks to raise money.
3.1.2 Moral hazard
Moral hazard describes a situation in which the uninformed party fears a loss due to undesirable actions by the informed party after the transaction is made. It results from the fact that both parties are utility maximizers (Jensen and Meckling, 1976) while having different interests. Mishkin (2006) points out that this problem plays a role for equity contracts and for debt contracts, both in private and public markets. Just as with adverse selection this problem is faced by banks when making loans and buying financial instruments and by lenders and investors who provide capital to banks. Frolov (2007) describes that especially in the banking business managers are able to
increase risks without “outside” parties noticing because the value of the bank assets is not transparent for other banks which provide short-term liquidity and shareholders. Two reasons exist for this quality uncertainty of banks’ assets. First because the banks’ assets are mostly financial assets which for one part allow quick and easy trading and thus enable banks to silently shift risk to the lenders and investors. Second because another part of the financial assets are non-transparent non-tradable loans.
3.1.2.1 Moral hazard in private markets
To mitigate the moral hazard problem in private markets with debt contracts, banks have to align the interest of the borrower with the interest of the bank. Because in this case the borrower has to pay a fixed amount at periodic intervals, receiving this amount is in the banks’ interest. Banks align interests as far as possible by using restrictive covenants to increase the borrower’s ability to pay the fixed amount. These covenants can demand several actions from the borrower by discouraging undesirable behavior (e.g. restrict engaging in predetermined risky businesses) and/or encourage desirable behavior (e.g. keep a certain equity/liability ratio, keep collateral in good condition) (Mishkin, 2006). An important part of aligning interests is monitoring the borrowers’ compliance with the covenants and a subsequent implementation of enforcement actions when the borrower does not comply. For monitoring debtors in private debt markets banks can use their expertise and/or can write a covenant which mandates the borrower to disclose information.
Although banks collect this information there always remains a certain risk of default of the borrower. In the case of mortgages made to subprime borrowers this risk of default was spread among market participants by securitizing these mortgages. Subsequently these residential mortgage-backed securities (RMBS’s) were packaged into collateralized debt obligations (CDO’s), which were intended to spread the risk even further (Dwyer and Tkac, 2009).
Furthermore, commercial banks did not provide the mortgages themselves, but bought these mortgages from other banks and investment banks. At the same time banks trusted the credit ratings of these financial products, instead of performing their classical work of detailed monitoring of debtors. In other words, the moral hazard risks that mortgage providers faced from less creditworthy home owners in private markets was shifted via private markets to investment banks. These banks spread this
risk by “slicing and re-packing” the debt into new financial products in such a way these products were receiving high credit ratings from the rating agencies. Subsequently the risk was spread again as the investment banks sold these financial instruments in the private market to commercial banks. But when subprime mortgages began experiencing higher delinquency and default rates than predicted it became clear that the values of these securities where hard to determine. And thus the spreading of risks resulted in more uncertainty.
As such, the classical way to mitigate the moral hazard problem in private markets was not followed by commercial banks which resulted in bad investments and unexpected losses. Therefore, one can make the assumption that commercial banks did not have proper insight in the risks they faced prior to the crisis. This could be one of the causes of low disclosure quality in 2007.
3.1.2.2 Moral hazard in public markets
To mitigate the moral hazard problem in public markets which involve debt and equity contracts, lenders and investors have to align the interests of the banks with their interests. The financial instruments traded in private markets are on the balance sheet of publicly traded commercial banks, which links the financial instruments to the public market.
When banks participate in public markets and engage in debt contracts moral hazard can be mitigated by writing covenants, followed by monitoring and enforcement. When participating in public markets and engaging in equity contracts Jensen and Meckling (1976) describe the relationship between banks and borrower as the ‘principal-agent’ relationship. They define this ‘principal-agent’ relationship as follows:
‘The owners of equity (principals) engage another party (the agent) to perform some service on their behalf which involves delegating some decision making authority to the agent. In this principal-agent relationship the principal (i.e. uninformed party) can partly align its interest with the interest of the agent (i.e. informed party) by establishing appropriate incentives for the agent (Jensen and Meckling, 1976).’
Monitoring and enforcement are crucial for proper alignment. But due to the free-rider problem that exists in public markets, for both debt and equity contracts, lenders and principals do not want to pay for monitoring the borrowers and agents. Therefore, in public markets borrowers and agents can reduce the moral hazard problem by disclosing honest information about their activities and risk exposure. This includes providing information about their activities in the private market such as received collateral, credit risk exposure and how this exposure is being decreased, the markets in which the loans are provided, the maturity and quality of loans provided and allowances made for these loans.
As banks lend out money for long periods they have borrowed for short periods, a sudden drawback of the borrowed money could lead to the bankruptcy of a bank. When a “big” bank (i.e. system bank) fails this could have a far reaching impact on the economy as a whole. Therefore, governments reduce the moral hazard problem that lenders to banks face. Governments bail-out banks in financial distress when a bankruptcy of the bank has more impact on the economy then the bail-out. This on the other hand results in banks taking higher risks when they know they are “too big to fail” and thus will be bailed out by the government. When there is no threat of bankruptcy there is also less incentive to disclose private information, this can result in lower disclosure quality in periods were banks are not bailed-out, thus in 2007. However, this effect can be offset by the fact that banks which are “too big to fail” are more visible. Higher visibility can increase incentives banks have to provide higher quality of disclosure. The relationship between visibility and disclosure incentives, as well as the incentives themselves, will be discussed in section 3.2.
Reducing the moral hazard (i.e. risk after the transaction) that lenders and investors face when doing business with banks in the public markets will also reduce the adverse selection (i.e. risk prior to the transaction) that lenders and investors face, this allows banks to raise money in the future. When banks continuously provide high quality disclosure in their financial statements, they are signaling to future lenders and investors that they will face less moral hazard and thus those lenders and investors are more willing to make transactions with those banks, which lowers the costs of capital for the banks and increases stock prices.
3.1.3 Agency problems
The previous parts showed that information asymmetry becomes a problem when parties which are willing to engage or are engaged in a transaction have conflicting interests. Once parties are engaged in a transaction in which one of them provides capital to the other in return for fixed payments or in return for a share of the profits, a separation between the decision and risk bearing function is created. In this relationship the capital provider faces the risk and the capital receiver makes the decisions about what to do with the capital. For debt contracts it is in the interest of the capital provider (i.e. lender) to receive fixed payments in accordance with the contract. For equity contracts it is in the capital providers’ (i.e. principal) interest that the capital receiver (i.e. agent) maximizes share value. Typically borrowers and agents have less at stake than lenders and principles, therefore borrowers and agents are willing to take more risks than lenders and principles. The problems of aligning these interests are called agency problems. The costs that result from this alignment include costs of structuring, monitoring and bonding (Fama and Jensen, 1983) as well as the value of output lost because the costs of full enforcement of contracts exceed the benefits (Jensen and Meckling, 1976) and are called agency costs.
In the situation of the financial crisis the agency problem lies in decreasing the excessive risk taking behavior of the banks. As earlier mentioned, lenders and investors have to monitor banks before they can take proper enforcement actions to decrease this risk taking by banks. In public markets banks can let themselves be monitored by disclosing their private information. But as risk taking is in the interest of the banks, they are reluctant to provide an optimal level of disclosure from the lenders and principals point of view, in terms of quantity and quality.
Thus, banks have incentives to provide voluntary disclosure which is dishonest, or to provide no disclosure at all about certain private information. The agency problem here concerns giving banks incentives9 to provide disclosures which are in the principals’ interest, and subsequently assess the credibility of these disclosures.
9Although without these incentives banks would not voluntary provide certain disclosures, the
disclosures provided after these incentives are in place are called ‘voluntary disclosure’ in accounting literature.
Healy and Palepu (2001) identify two mechanisms for increasing the credibility of voluntary disclosures. First, third-party intermediaries can provide assurance about the quality of management’s disclosures. In the case of publicly traded commercial banks third-parties are the auditors, more specifically the “Big Four” (i.e. Deloitte, Ernst & Young, KPMG, PwC). Secondly, there can be validation of prior disclosures through required financial reporting itself. This would only be effective when managers and auditors are penalized for disclosures that are subsequently proven false. Penalization can come in different forms. One could think of litigation and jail-punishment, forms of punishment that are more frequently seen in common-law countries (e.g. United States and Canada) than in code-law countries (e.g. EU countries). One also could think of reputation loss for audit firms, which in most cases is more severe than monetary fines. And job loss for managers, which is in most cases less severe than jail-punishment. But as both jail punishment and litigation threat is less in code-law countries, the assurance by auditors and validation of prior disclosures do not have to result in full compliance with IFRS 7. However, the risk of reputation loss for auditors increased when it became clear they had provided assurance to the overvalued financial statements of banks in 2007. Therefore, in 2008 much public attention was given to the banks and the auditors, to decrease reputation loss auditors had to pay more attention to the compliance of their clients with IFRS 7 in 2008. This can result in higher disclosure quality in 2008 compared to 2007. Besides reputation loss, litigation costs and formal enforcement banks have incentives to provide voluntary disclosure. The incentives banks have to voluntary provide disclosures as well as the incentives banks have to not voluntary provide disclosures are discussed in section 2.2.1. Voluntary disclosure.
3.2 Disclosure
By partly mitigating the information asymmetry problem and in that way increasing the efficient allocation of capital, disclosure of banks’ private information in public markets is socially desirable. The next section about voluntary disclosure discusses the incentives that banks have to voluntary provide disclosures as well as the incentives that banks have to not voluntary provide disclosures about private information. Section 2 showed that although IFRS 7 requirements are mandatory, parts of these requirements are subject to the discretion of management and thus also are subject to disclosure incentives.
Frolov (2007) argues that in the case of banking the incentives to voluntary disclose private information are not sufficient enough (i.e. voluntary disclosure falls short of the socially-optimal level of disclosure), and that this lack of incentives thus brings the attention to mandatory disclosure. It has to be said that as a result of agency costs not all managerial accounting manipulation will be eliminated, so the risk of fraud always remains. Section 3.2.2 addresses mandatory disclosure in banking.
3.2.1 Voluntary disclosure
3.2.1.1 The effect of visibility on voluntary disclosure
As outlined by Wallace et al. (1994) and Gallery et al. (2008) incentives to provide voluntary disclosures are stronger for firms which are more visible, this visibility concerns banks and audit firms especially during the financial crisis in 2008.
Besides the fact that it is generally known that the visibility of banks and audit firms increased in 2008 I quantify this visibility of banks by researching the 2007 and 2008 quantity of articles about the financial crisis in the leading Dutch financial newspaper, Financieel Dagblad. This financial newspaper has 6 editions a week. The full editions of 2007 and 2008 were digitally available on the website of the Financieel Dagblad. By counting the articles in 2007 I note that 10% of the Financieel Dagblad editions included at least one article with negative news about the financial crisis in 2007. Over the last quarter of 2007 this was 17%. In 2008 40% of the papers included at least one article with negative news about the financial crisis. Over the last half year of 2008 this was 60%. One can see a strong increase in editions with articles related to the financial crisis between 2007 and 2008. Articles in the Financieel
Dagblad also have an impact on other news in the Netherlands as it is a much quoted newspaper. This increase in editions with articles related to the financial crisis therefore also increases visibility of banks in 2008.
Furthermore I reviewed the content of the articles and noted the following. In 2007 and the first half year of 2008 the general belief in these articles showed that the crisis was expected to be only temporary and that the EU would not suffer that much from the crisis. In the second half of 2008 the focus of the articles shifted to the banks and showed that the EU also would suffer and that the crisis will remain until at least 2010. So not only the quantity of the articles but also the substance of the articles increased the visibility of banks. Visibility will increase the incentives which will now be discussed.
3.2.1.2 Incentives for voluntary disclosure
Healy and Palepu (2001) reviewed previous research on disclosure and identified six forces that effect managers’ voluntary disclosure decisions for capital market reasons: capital market transactions, corporate control contests, stock compensation, litigation, proprietary costs, and management talent signaling10. As discussed earlier not only managers are involved in the disclosure of financial information, also auditors are involved in this process. Therefore, the incentives that auditors have to make their client comply with IFRS 7 are also discussed. All these forces can induce both positive as negative incentives on disclosure and will be discussed next.
The first incentive is the capital market transaction theory, it suggests that managers who anticipate making capital market transactions (e.g. issue public debt or equity) have incentives to provide voluntary disclosure to reduce the information asymmetry, thereby reducing the firms’ information asymmetry component of the cost of capital11. The information asymmetry component of the cost of capital is the discount that capital receivers provide as a means of accommodating the adverse selection problem (Verrecchia, 2001). Verrecchia (1983) explains that even
10 Management talent signaling will not be discussed further because there is no evidence that supports
or refutes this hypothesis.
11 Hail and Leuz (2007) reviewed empirical evidence of capital market effects of disclosure and
conclude that there is a negative relationship between disclosure and the information asymmetry component of the cost of capital.
disclosure of “bad” information12 is favorable for the capital receiver, but only in the case where the capital provider is expecting that “bad” information. In that case no disclosure of good or bad information will result in a discount of the information to “bad” by the capital provider, in other words, no disclosure is seen as disclosure of “bad” information. But there remains the case where the capital provider does not expect “bad” information, in this case there is no punishment when capital receivers do not provide sufficient disclosure. The capital receiver would then delay disclosure of “bad” news while hoping that during the interim period some “good” news will occur to offset the “bad” news (Verrecchia, 1983). Another reason why banks are reluctant to disclose “bad”news is provided by Frolov (2007). He argues that because non-aggregated information about bank assets is noisy and therefore its disclosure may lead to excessive funding costs for banks. Since the banking business relies on the diversification of loan exposure across many borrowers, it is difficult for bank outsiders to judge whether a single loss event they observe in a bank’s loans indicates a generally mispriced portfolio or just an extreme realization in a correctly priced portfolio (Frolov, 2007).
The corporate control contest theory suggests that, given the risk of job loss accompanying poor stock and earnings performance, managers use corporate disclosures to reduce the likelihood of undervaluation and to explain away poor earnings performance (Healy and Palepu, 2001).
Healy and Palepu (2001) identified two main reasons why stock compensation schemes provide incentives for managers to engage in voluntary disclosure. First, managers interested in trading their stock have incentives to disclose private information to meet restrictions imposed by insider trading rules and to increase liquidity of the firm its stock (Healy and Palepu, 2001). Second, managers acting in the interest of existing shareholders have incentives to provide voluntary disclosures to reduce contracting costs associated with stock compensation for new employees (Healy and Palepu, 2001). However Core (2001) provides a critical review on the paper of Healy and Palepu (2001) and brings to the attention that managers not always
12 On May 15, 2008 the Financieel Dagblad reported a case where an impairment of subprime securities
by ING made its share price rise: ING made an impairment of EUR 55 million on their subprime mortgage securities, but as investors were expecting worse impairments ING stock rose with 3.9% after the announcement while stocks of other banks in the EU remained on their original level.
act in the interest of the principals. In that situation there remains a case that managers do not follow the law and selectively disclose at the expense of shareholders.
The litigation cost theory suggests that capital providers can litigate managers when they discover inadequate or untimely disclosure by these managers. Healy and Palepu (2001) identify both an incentive for management to voluntary disclosure as well as an incentive to reduce disclosure. Under the threat of litigation management would voluntary provide disclosure about information of which they are more certain. Management would provide less disclosure about more uncertain (e.g. forward-looking) information. This would occur if managers believe that the legal system penalizes incorrect (with hindsight) forecasts made in good faith because it cannot effectively distinguish between unexpected forecast errors due to chance and those due to deliberate management bias.
Verrecchia (1983) argues that for both favorable as for unfavorable information proprietary costs exist, which lead to less disclosure. Disclosure of favorable information can lead to proprietary costs because this information can be useful to competitors, shareholders, or employees in a way which is harmful to a firm’s prospective. Frolov (2007) argues the disclosure of proprietary information about borrowers can endanger the very existence of the banking business. In this case the information about borrowers that banks produce in the private debt market becomes public which results in the free-rider problem. Competitors would be able to economize on the production cost and leave the disclosing bank either with no profit or with net loss on the production of the information (Frolov, 2007). Thus the disclosure of certain proprietary information is not favorable for principals because it is a threat to the existence of the bank, though without this information principals cannot monitor a part of the agents’ risk behavior.
Disclosure of unfavorable firm information can lead to bank runs (i.e. depositors withdraw their savings en masse). Because banks operate on a “sequential service constraint”, depositors have a very strong incentive to show up at the bank first. These bank runs also can occur when no disclosure is given. Miskin (2006) provides the next example to explain this. Due to an adverse shock, 5% of the banks have such large losses on loans that they become insolvent. Because of asymmetric information, depositors are unable to tell whether their bank is a good bank or one of the 5% that are insolvent. Depositors at bad and good banks recognize that they may not get back all of what they deposited and will want to withdraw the funds.
As discussed by Beatty (1993) incentives auditors face to comply with regulation are losses directly from litigation (i.e. legal judgments, costs of defense) and reputation loss. Audit firms will offer higher quality services when there is a litigation threat (Venkataraman et al., 2008), and thus higher quality of disclosure by their clients. Of course when it became clear that financial instruments as included in audited financial statements were mispriced the litigation threat of auditors increased. However, as EU countries are seen as code-law the threat of litigation in the EU is not significant. Therefore the litigation threat is not a big incentive on audit firms in the EU. The threat of reputation loss on the other hand is real, as it is not related to the legal environment. The incentive of reputation loss in 2008 has increased audit quality and thus compliance with IFRS 7.
In summary, banks have incentives to disclose private information that is in their interest. It depends on the situation and information if voluntary disclosure will lead towards a socially optimal level. In a situation where outsiders do not expect certain unfavorable information, banks have incentives to not disclose this information or disclose dishonest information. In 2007, the year prior to the financial crisis, outsiders of banks did not expect the unfavorable information about banks that was revealed in 2008. Thus in 2007 banks had incentives to not disclose that unfavorable information. This will lead to a disclosure level that is not socially optimal. Therefore, only voluntary disclosure is not sufficient enough. Furthermore, audit firms also faced less strong incentives to get their clients to comply in times without significant financial distress. The next section therefore looks at how mandatory disclosure may lead towards a more socially optimal disclosure level.
3.2.2 Mandatory disclosure
Hail and Leuz (2007) provide three arguments from the regulatory literature which can justify mandatory disclosure: the existence of externalities, economy-wide cost savings from regulation, strict sanctions serve firms as a commitment device.
The first argument concerns the existence of externalities, these arise whenever the social and private values of information differ. For example in the case where information disclosed by one firm can be useful in valuing other firms and increase investors’ willingness to hold position in other firms. In such a case, without mandatory disclosure firms trading off the private costs and benefits do not provide the socially optimal level of disclosure (Hail and Leuz, 2007). In theory, disclosure
regulation can mitigate this problem by mandating the socially optimal level of disclosure. However, in practice, it is likely to be difficult for regulators to determine the socially optimal level of disclosure and whether markets produce too little or too much information (Leuz and Wysocki, 2007).
The second argument that is provided by Hail and Leuz (2007) states that privately imposed disclosure regulation on agents is very expensive and in many cases even impossible because the penalties that private contracts can impose are generally limited. In other words, the agency problem cannot be decreased enough because the agency costs are too high. Thus, if contractual and monetary penalties are not sufficient, a mandatory disclosure system with a public enforcer and criminal penalties can offer advantages. This threat of penalties then serves as a mechanism to improve the credibility of disclosures.
The third argument is closely related to the previous argument. It states that a mandatory disclosure regime serves as a commitment device. As mentioned under the voluntary disclosure section, firms (banks included) have incentives to withhold or manipulate information in certain situations. Mandatory disclosure requirements forces firms to reveal information in both good and bad times, i.e. they provide some form of commitment, which in turn should mitigate information asymmetries and uncertainty (Hail and Leuz, 2007).
However, mandatory disclosure can only provide these benefits when it is sufficiently enforced (e.g. FEE, 2002; Brown and Tarca, 2005; Dao, 2005; Larosière, 2009). LaPorta (1997), Kaufmann et. al. (2009) and Preiato et. al. (2009) all provide different proxies which measure enforcement across countries, all these proxies indicate that enforcement is far from optimal. This would in theory indicate that the compliance with mandatory disclosure requirements is also not optimal. This theory is strengthened by empirical evidence which finds significant non-compliance with mandatory disclosure requirements (e.g. Street et. al., 1999; Glaum and Street, 2003; Hodgeon and Wallace, 2008; AFM, 2009; CESR, 2009; Wohlmannstetter, 2009).
In summary, mandatory disclosure can lead towards a more socially desirable level of disclosure than voluntary disclosure mainly because it can force sufficient commitment upon the agents through penalties. This serves as an incentive to reveal proper information both in good times and bad times. However, due to the lack of sufficient enforcement of mandatory disclosure requirements compliance with mandatory disclosure regulation is not optimal. This is especially the case in EU
countries which are in general code-law countries. Thus also when banks have to comply with mandatory requirements they are able to not comply. Compliance with mandatory requirements as IFRS 7 can therefore vary between banks and between years.
4 Financial crisis effects on disclosure
The previous section described how disclosure of private information can decrease information asymmetry by decreasing adverse selection and moral hazard in public markets with equity contracts. It also showed that only voluntary disclosure is not sufficient enough to reach a socially desirable disclosure level and that therefore mandatory disclosure requirements need be put in place to reach a higher level of disclosure. Compliance with these mandatory disclosure requirements is not optimal because the requirements are not enforced strongly enough, which is especially the case in code-law countries. Therefore, voluntary disclosure incentives still play an important role in what is disclosed under mandatory disclose requirements. As previously described, banks have an incentive towards socially undesirable high risk taking behavior and also have the opportunity to increase risks because the value of the bank assets is not transparent to outsiders. This is due to parts of financial assets that are quick and easy tradable. And because parts of the financial assets are opaque non-tradable loans for which the value is hard to determine. Therefore, compliance with mandatory disclosure requirements can change depending on the circumstances banks face. The financial crisis that became visible in 2008 provided such a change in the environment in which banks operate.
Because 90% of total assets and liabilities of banks are classified as financial instruments13 (Bischof, 2009), disclosure requirements which EU listed banks have to comply with most are the mandatory requirements of IFRS 7 Financial Instruments: Disclosure. To research the effect of the financial crisis on banks’ disclosure quality, the compliance with this standard will be used as a proxy for disclosure quality.
Avgouleas (2009) points out that miss-aligned incentives and inadequate disclosure have been widely cited as almost the sole cause of the global financial crisis. In line with this theory one can argue that the information asymmetry provides an opportunity for banks, which have other incentives than their stakeholders, to provide
13 A financial instrument is defined as any contract that gives rise to a financial asset of one entity and a
financial liability or equity instrument of another entity (e.g. trade receivables and payables, bank loans and overdrafts, issued debt, ordinary and preference shares, investments in securities, various derivatives) (IAS 32.11).
socially sub-optimal disclosures. Under IAS 39 financial instruments are recognized and measured differently based upon their characteristics. The choice of how to measure and recognize financial instruments has a major impact on their value and profit and loss figures. Banks are therefore able to manage their earnings by changing their portfolio characteristics. Disclosures under IFRS 7 are thus crucial for investors to understand and interpret the balance sheet and profit and loss figures.
In 2007 the market did not show its awareness of the credit risks that eventually would lead to the crisis. In this case the theory of Verrecchia (1983) which predicts that banks will delay disclosure of the “bad” information while hoping “good” information in the future will offset the “bad” information was applicable. Banks are reluctant to disclose “bad” information because this can have a negative effect on stock value, they fear bank runs which can cause the bank to default, and in this case disclosure would have revealed their risk-taking behavior. Banks are also reluctant to disclose information about which they are uncertain. In this case information about the credit risks they faced, because of the litigation threat in the case they provide false disclosures. Taken this into consideration, in 2007 bank managers had opportunities to not fully comply with mandatory disclosure requirements due to the unawareness of the “bad” situation by outsiders, and even had incentives to not fully comply with IFRS 7 due to the litigation threat that disclosing uncertain information brings along. Furthermore, in periods of low visibility, audit firms also have less strong incentives to force banks to disclose items against their will as this can lead to a bad client relationship which can result in losing the client. With hindsight it also became clear that in 2007 audit firms probably were not aware of the credit risks that banks faced.
A year later in 2008 both the awareness of the crisis and the visibility of banks increased dramatically. The awareness of the financial crisis by markets is especially evident from the equity market deterioration around mid-2008. This equity market deterioration resulted from the quality uncertainty that investors faced (Bartram and Bodnar, 2009). The awareness by banks is especially evident from the poor functioning credit markets in 2008 and the TED spread which peaked in mid-2008. That banks were the source of a public debate and thus became more visible is evident from the increased public attention for banks in newspapers as of mid-2008. Because banks have incentives to increase the stock value of underpriced stocks and to recover the poorly functioning credit markets, banks have to provide higher disclosure quality
