Fair value accounting and the financial crisis : literature review of the media’s conjectures about the relationship between fair value accounting and the financial crisis

(1)

1

Fair value accounting and the

financial crisis

Literature review of the media’s conjectures about the relationship between fair value accounting and the financial crisis

Wouter Johannes Kempers 10422676

UNIVERSITEIT VAN AMSTERDAM MS. H. KLOOSTERMAN MSC FINAL DRAFT BACHELOR THESIS BSC ACCOUNTANCY & CONTROL

(2)

2

Statement of Originality

This document is written by Student Wouter Kempers, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

List of Tables

Table 4.1: Argument about the contribution of fair value accounting to the

leverage increase in financial booms………18

Table 4.2: Arguments about the contribution of fair value accounting to

contagion among financial institutions………..21

Table 5.1: General arguments about the contribution of fair value accounting

(5)

5

Abstract

The credit crisis of 2008 was the result of the burst of the housing bubble. Many argue that this was the result of behaviour of financial institutions. However, in the media, there was another cause being brought up: fair value accounting for financial institutions. I.e., the media claimed that fair value accounting contributed to the financial crisis through pro-cyclicality. Pro-cyclicality is a process in which financial institutions that use fair value accounting have to write down their assets in a financial crisis, as a result of dropping market prices. Because they still have to meet certain capital requirements, the financial institutions have to sell their assets in the market. The increased supply that now occurs, pushes the market prices further down, and puts the financial institutions in a negative spiral. This literature review examines the relationship between fair value accounting and the financial crisis by looking at the conjectures that the media made about it.

In the academic literature, two ways are named in which fair value accounting contributed to pro-cyclicality.

1) The increase in leverage that financial institutions take on in times of financial booms when fair value accounting is used, which increases volatility.

2) The contagion among financial institutions that increases when mark to market accounting is used in the financial crisis.

I find evidence that suggests that fair value accounting contributed to pro-cyclicality in the first way, but not in the second way. Nevertheless, I conclude that fair value accounting has contributed the financial crisis as the media suggest it did.

(6)

6

Samenvatting

De kredietcrisis van 2008 is ontstaan na het barsten van de huizenbubbel op de Amerikaanse huizenmarkt. Velen geven aan dat het gedrag van financiële instellingen deze crisis heeft veroorzaakt. De media noemen echter een andere oorzaak, namelijk: “Fair Value Accounting.” Zij stellen dat fair value accounting bijdroeg aan het ontstaan van de crisis vanwege procycliciteit. Dit houdt in dat deze vorm van economische regulering het effect van de economische situatie versterkt. Er is sprake van procycliciteit aangezien financiële instellingen, die verplicht zijn om fair value accounting toe te passen, tijdens een financiële crisis genoodzaakt zijn de activa af te schrijven. Dit wordt veroorzaakt door dalende marktprijzen. Doordat deze instellingen aan de kapitaalvereisten moeten blijven voldoen, worden activa te koop aangeboden. Door een aanbodoverschot dalen de marktprijzen verder, waardoor men in een negatieve spiraal terecht komt. Dit onderzoek probeert inzicht te bieden in de relatie tussen fair value accounting en de financiële crisis, door zich te richten op de vermoedens die in de media zijn uitgesproken.

In de academische literatuur worden twee manieren genoemd, waarop fair value accounting bijdraagt aan procycliteit.

1) Een hoge verhouding van de schulden ten opzichte van het eigen vermogen, dat bedrijven creëren in perioden van economische bloei.

2) Het besmettingsgevaar dat ervoor zorgt wanneer één financiële instelling zijn activa afschrijft, meerdere financiële instellingen zullen volgen.

In de analyse wordt alleen bewijs gevonden dat de eerste factor heeft bijgedragen aan het ontstaan van de financiële crisis. Voor de tweede factor blijft dit bewijs uit. Desondanks wordt de conclusie getrokken dat fair value accounting heeft bijgedragen aan de crisis, zoals de media claimen.

(7)

7

1. Introduction

The most recent financial crisis started with the housing bubble (Achayra, 2009; Diamond, 2008). It brought down the financial system through two types of financial institutions’ behaviour designed to evade the regulatory capital requirements. First, they had placed assets temporarily in off-balance sheet entities, so that they did not have to hold a capital buffer against them. Second, the regulations allowed financial institutions to reduce the amount of capital that financial institutions had to hold against them if they were AAA rated assets. According to the media however, these two reasons were not the only ones causing the financial crisis.

According to the media, fair value accounting for financial institutions was the other cause of the financial crisis (Forbes, 2009; Wallison, 2009). Steve Forbes is editor in chief for the business magazine Forbes. In his opinion the primary reason that there was a financial crisis is fair value accounting. He argues that it put financial institutions in a negative spiral by having them write down their financial assets. When financial institutions have written down their assets, they had to get new capital. This expanded supply of securities pushed the price further down. Financial institutions then had to write down their assets again, and the circle started again. This circle, more commonly known as pro-cyclicality, the media claim is caused by fair value accounting. To escape the vicious circle, Forbes called for a suspension of fair value accounting (Forbes, 2009).

On the other hand, Shaffer (2009) among others claims that there is no link between fair value accounting and pro-cyclicality. Badetscher (2012) adds that there is a minimal effect of fair value losses on regulatory capital and that financial institutions with lower capital ratios actually sold less during the financial crisis than financial institutions with high capital ratios, when using fair value accounting.

The media suggest that fair value accounting contributed to the financial crisis through pro-cyclicality. I look at the academic literature to determine to what extent fair value accounting contributed to the financial crisis.

As is mentioned above, there have been critics in the media that say fair value accounting contributed to the recent financial crisis. Some have claimed that fair value accounting contributed to the financial crisis in a major way. There have been studies to examine the effect of fair value accounting on the financial crisis, which lead to

(8)

8

contradictory beliefs on the subject. These studies however are mostly based on models, so there is not a lot of empirical research on this subject available. Therefore, a lot of the evidence that is found in this paper is based on models, and not on empirical research. Next to that, there have been no studies that look at the relationship between the crisis and fair value accounting by starting with conjectures from the media’s perspective. Therefore, I try to determine if the academic literature and the media agree on the contribution of fair value accounting to pro-cyclicality and therefore to the crisis. This study reviews these different views in the academic literature about the media’s conjectures and tries to determine whether the conjectures that the media make about fair value accounting in the financial crisis are right according to the academic literature. It is important to determine whether the media is right or wrong because they could have an influence in decision making of people. If the media spreads information that claims that fair value accounting contributed to the financial crisis, many people will think it did. The FASB and IASB will be pressured to change accounting rules on fair value accounting even though there might not be a connection between fair value accounting and the financial crisis. Therefore, it is important to know that the conjectures that are made by the media are right according to the academic literature. Hence, the problem of whether or not fair value accounting contributed to the financial crisis in the way that the media claim is examined in this thesis. The purpose of this study is to review theories and evidence to find out in which way fair value accounting contributed to the financial crisis.

In the rest of this thesis, the standpoint of the media will be tested by using the academic literature. In chapter two conjectures about and causes of the financial crisis are described from the viewpoint of the media. The media have one clear argument in the form that fair value accounting contributed to pro-cyclicality, which in turn contributed to the financial crisis. In chapter three the background of fair value accounting is outlined. Chapter three also explains how fair value accounting could have contributed to pro-cyclicality. I find that there are two main ways in which fair value accounting could have contributed to pro- cyclicality and thereby to the financial crisis. These are:

1) The increased leverage which firms take on in financial booms, that makes the financial system more vulnerable

(9)

9

In chapter four I analyse these two ways by looking at the academic literature. I find unilateral evidence about the increased leverage that financial institutions are said to take on in times of financial crisis. Therefore, I draw the conclusion that fair value accounting caused this increased leverage in booms and thereby contributed to pro-cyclicality and the financial crisis. About the criterion of contagion among financial institutions, I find mixed evidence. I draw the conclusion that I have no reason to assume that fair value accounting contributed to pro-cyclicality through contagion. In chapter five I discuss other ways that fair value accounting could have had an influence on the financial crisis. In this chapter I come to the same conclusion as I do in chapter four: fair value accounting did contribute to the financial crisis through pro-cyclicality. This does however not mean that the accounting standards should be changed. In the conclusion, I put the two criteria together and come to the conclusion that fair value accounting has contributed to pro-cyclicality through the increased leverage financial institutions take on because of fair value accounting, but not through the contagion among financial institutions. Therefore, in chapter six I come to the general conclusion that the media were right in their conjectures: fair value accounting did contribute to the financial crisis through pro-cyclicality.

2. Causes of the financial crisis

This chapter explains what the media say about how the financial crisis started and what has contributed to its severity, according to them. Based on the explanations provided in this chapter, the next chapters test whether the outcomes that are reached in this chapter are correct, based on the academic literature.

As described above, the financial crisis started with the housing bubble (Achayra, 2009; Diamond, 2008). It brought down the financial system through two types of financial institutions’ behaviour designed to evade the regulatory capital requirements. First, they had placed assets temporarily in off-balance sheet entities. If assets are placed on the balance sheet, regulations require financial institutions to hold a capital buffer against them. By placing the assets in off-balance sheet entities they evaded

(10)

10

the capital requirements and therefore did not have to hold a buffer against the assets. Second, the regulations allowed financial institutions to reduce the amount of capital that they had to hold against them if they were AAA rated assets. Poorly backed securities such as ninja loans (no income, no asset) where put together in large packages that got a AAA rate assigned. Therefore, even some of the poorest backed securities got a AAA rating. The financial institutions that held them therefore were allowed by regulations to decrease the amount of capital they had to hold against them. These assets were financed by short term debts. In good financial times, short term debt seems relatively cheap to long-term debt, but in bad times it becomes clearer that there is a high risk of illiquidity for financial institutions. At the peak of the crisis, mortgages and securities backed by them had become almost worthless on the open market (Lee, 2014).

As argued by Forbes (2009) and Wallison (2009) in the introduction, the above mentioned causes of the financial crisis are not the only ones. They say fair value accounting causes financial institutions to go into a negative spiral. This negative spiral can be explained as follows. In times of financial crisis, the market value of assets goes down. Because financial institutions use fair value accounting, the assets on their balance sheet have to be marked down to those prices, creating a balance sheet in which the assets have less value. Because the assets have less value, the value of the capital of the firm decreases in value. Financial institutions now have trouble in meeting their capital ratios and other control mechanisms. Therefore, they have to sell assets. Because of the sale of these assets, the supply of assets for sale goes up, pushing the market price further down. This pushes the value of the assets of the firms further down, decreasing the value of the capital of the financial institutions further. The story repeats itself, creating a vicious circle.

The vicious circle that Wallison, DeGrauwe and Forbes speak of is part of a phenomenon that not only describes a vicious circle in financially bad times. It is called pro-cyclicality and it tends to exacerbate financial trends. As market values rise, increasingly more money comes available to carry these assets. In this case, demand is higher than supply and prices rise more. Therefore, the upward spiral, that is also known as a bubble, continues. A vicious circle as described in the upper section is what happens in times of falling market values.

Forbes and Wallison, and Van de Poel however, are not the only ones that have been critical on fair value accounting in the media. DeGrauwe (2008) is one of the other

(11)

11

media critics of fair value accounting. Namely, he claims that mark to market accounting did not merely cause the downward spiral, but also accelerated it. Then fair value accounting does not push the prices further down, so the negative effect of pro-cyclicality is eliminated.

Lee (2014) is another critic that says that the Enron case provides with an example of how fair value accounting is hardly free of manipulation. He claims that one of the results of the aforementioned pro-cyclicality is that mortgages and securities backed by assets had almost become worthless in the open market. He also says that fair value accounting in times of a crisis is even less reliable because nobody wants to buy anything due to panic. Pozen (2009) is yet another critic that claims that fair value accounting drives financial institutions to be almost insolvent because it erodes their capital base. He furthermore says that fair value accounting has forced an artificial decline in asset values.

Wallison (2009) furthermore says that with the exception of assets that are held for trading purposes, the true value of an asset is more likely to be the value of the future cash flows that it produces than the value of the market. This especially holds in a situation where asset prices are driven by panic. Moreover, he argues that the standards for keeping assets as held to maturity should change, so that it is easier for financial institutions to hold assets in this category. Van de Poel (2008) says this is because long term assets need other policy instruments to guide the asset. These assets are not marked to market in this way. Nonetheless, this only holds for long term assets, because short term assets should still be valued at market value. Assets held for sale should be held at the value of their future cash flows to make them more veracious. Next to that, Van de Poel (2008) says that fair value accounting should be temporarily suspended, so that in the meantime, fair value accounting can be fine-tuned and upgraded. Updating and using fair value accounting at the same time takes too long he says.

Putting this together, the media claim that fair value accounting caused pro-cyclicality, which exacerbated the financial crisis. The next chapter looks at how fair value accounting works and how it could have contributed to pro-cyclicality by looking at the academic literature.

(12)

12

3. Fair value accounting

This chapter starts by outlining the background of fair value accounting. Thereafter, I look at what the academic literature says about the conjectures that the media have on fair value accounting. This is done by looking at two ways in which fair value accounting contributes to pro-cyclicality according to the academic literature.

3.1 Background

As described by both IFRS 13 and IAS 157, fair value is the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants on the measurement date. Fair value accounting means that quarterly, the price of an asset is revalued according to the aforementioned definition of the fair value of that asset. Transaction costs should be excluded in the fair value measurement. SFAS 159, about the fair value option for financial assets and financial liabilities, and SFAS 157, on fair value measurements, followed IFRS 13 in a move towards more application of fair value accounting.

Only after the crisis of the 1930s in the USA, there came standards for reporting asset values as a direct response of the lack of stability that characterizes this crisis(Georgiou & Jack, 2011). For a long time the standard has been historical cost accounting, but from 1970 until the start of the crisis, fair value accounting emerged and started to be the standard for increasingly more securities.

Before the emergence of fair value accounting, historical cost accounting was more commonly used (Georgiou & Jack, 2011). Under historical cost accounting, assets are recorded at the amount of cash or cash equivalents paid or the fair value of the consideration given to acquire them at the time of their acquisition. Liabilities under historical cost accounting are recorded at the amount or proceeds received in exchange for the obligation, or at the amount of cash or cash equivalents expected to be paid to satisfy the liability in the normal course of business (Stolowy & Lebas, 2006).

SFAS 159 describes the rationale of the change towards fair value accounting: “The board considers fair value measurements of financial instruments to be more relevant to financial statement users than cost-based measurements because fair value reflects the current cash equivalent of the entity’s financial instruments rather than the price of a past transaction.” Also empirical research from the past twenty years has found that a firm’s stock price is more closely associated with the market value of

(13)

13

its underlying financial or real assets than with their historical cost (Barth, Beaver, and Landsman, 2001; Landsman 2006).

In 2008 the FASB and IASB temporarily refrained from some aspects of fair value accounting for financial institutions by introducing some provisions. In 2009, they provided additional guidance on how to account for securities in illiquid markets. This was being done so that companies in illiquid markets could from then on value the assets on their books as if they were orderly transactions, rather than being dumped in a fire sale.

By moving towards a larger application of fair value accounting, the two frameworks (IASB and FASB) distinguish a hierarchy in inputs, from most reliable to least reliable. Level one inputs are quoted prices in active markets for identical assets or liabilities that the entity can access at the measurement date. Quoted prices in active markets provide the most reliable evidence of fair value and is used without adjustment to measure fair value whenever available, with limited exceptions. This holds even if the market’s normal trading volume is not sufficient to absorb the quantity held by the firm. However, in case of distress, the market values go down and therefore, the values of the balance sheets of firms that are marked to market go down in theory.

If there are no quoted market prices available, level two inputs are used. These are inputs other than quoted market prices included within level one that are observable for the asset or liability, either directly or indirectly. There are two subclasses of these inputs. The first one is quoted market prices in active markets for similar items or in inactive markets for identical items. The reliability depends on the nature and magnitude of the adjustments. The second subclass is other observable inputs such as yield curves and exchange rates. The reliability of these inputs, that are also called mark-to-model inputs, are only as reliable as the models deployed. According to Ryan (2008), during the crisis, the price transparency offered by these sources substantially evaporated. Dealers had been reluctant to provide bid and ask quotes for subprime positions, and when they did the bid-ask spread was very wide. Only few orderly transactions occurred, and the ones that did typically were privately negotiated (Ryan, 2008).

Level three inputs are unobservable inputs for the asset or liability. These inputs are supplied by the firm and are supposed to reflect the assumptions that market participants would use. These inputs are often derived from statistical or other analysis of historical data (Ryan, 2008). In a crisis however, one cannot reasonably expect that

(14)

14

the future will be the same as the historical data. These inputs therefore are easier manipulated by management, and therefore less reliable.

As discussed in chapter two, the media speculated on the influence that fair value accounting has on the financial crisis. The argument they put forth is that of pro-cyclicality. A lot has been written on this subject in the academic literature. To test whether pro-cyclicality caused or contributed to the financial crisis, as the media say it did, in the next two chapters, the arguments on pro-cyclicality found in the academic literature are being explained. I start with giving some general arguments about the pros and cons of fair value accounting.

3.2 How could fair value accounting contribute to the crisis

Even before the recent financial crisis, there was critique of fair value accounting. Barth, Landsman, and Wahlen (1995) for example claim that fair value earnings are more volatile than historical cost earnings. They also claim that regulatory capital requirements for financial institutions are more often violated under fair value accounting than under historical cost accounting. Nelson (1996) argues in another critique before the financial crisis that the fair value measures for investment securities, loans, deposits, long-term debt and off-balance sheet financial instruments are not relevant in value for book value and the financial statement proxies for future profitability. This means that fair value has no incremental explanatory power in relation to book value. Thus, according to Nelson, for these securities, there is no reason that fair value accounting instead of other measures should be used.

During the financial crisis, critique of fair value accounting increased. Benston (2008) names four shortcomings of fair value accounting. He first says that fair values that are based on level two or three inputs are costly to determine and verify. Managers must determine the best use to which a group of assets can be put. This must be verified by auditors. The process of determining the use and verifying it is often a major undertaking in an informational way. Second, because managers are the ones that determine the value, these other than level one inputs are easily manipulated. Third, transaction costs should be excluded in the fair value measurement, but often are not. Finally, assets and liabilities that are stated at their exit prices have little value to investors that want to know the value of the ongoing firm. Whalen (2008) agrees with this view and adds that fair value accounting is driving and increasing investor fears about the solvency of many financial institutions.

(15)

15

Magnan (2009) claims in another critique on fair value accounting that it has severely undermined critical foundations of financial reporting in four ways. First, fair value accounting encompasses different measurement modes (level one, two and three). The validity of these alternatives is conditional on market efficiency. Second, fair value based figures are not necessarily optimal from a societal perspective. It is focused too much on shareholders and therefore does not take the regulator’s objectives of stability in the banking system into account. Third, even though the rationale for concepts such as verifiability, reliability and conservatism still exists, fair value accounting does challenge these traditional accounting concepts. Finally, the volatility in reported earnings is potentially increased through the implementation of fair value accounting. Critique has not only been focussed on fair value accounting in general however. As mentioned above, the largest critique of fair value accounting is that it caused the financial crisis. The media claim this is being done by the principal of pro-cyclicality. The pro-cyclicality can be linked to fair value accounting in two ways (Laux & Leutz, 2009a):

1) Leverage increase in financial booms

The first way is that fair value accounting and asset write-ups allow financial institutions to increase their leverage in financial booms. In financial booms, financial institutions want to increase their leverage because it allows them to invest more. Financial institutions’ balance sheet values rise under fair value accounting in financial booms, in comparison to other accounting measures. Because of this, banks are willing to lend them more money, because the financial institutions have more assets to back their loans. However, in case of a financial downturn, the chance that financial institutions pay back their loans decreases if they have increased their leverage, because the value of their capital in comparison to debt decreases. Now, it is harder to pay back their loans in case of a default of one of their securities (Berk & De Marzo, 2011). This means that financial institutions are subject to increased vulnerability. The financial market has therefore increased vulnerability, which makes financial crises more severe if they occur (Plantin et al. 2008).

As Adrian and Shin (2008) point out, when marking to market accounting is used, changes in balance sheet size are strongly correlated to changes in leverage. They say that financial intermediaries adjust their balance sheets actively. This means that the leverage is high during booms and low during busts, thus leverage is

(16)

pro-16

cyclical. From a firm’s perspective it can be understood as follows: when asset prices increase, the balance sheets of firms get stronger. The firms hold surplus capacity. They will try to find a way to exploit this surplus capacity by expanding their balance sheets. They take on more short term debt on the liabilities side, because debt is relatively cheaper for firms with a high enough capital buffer. On the asset side, firms try to find borrowers to expand their capacity surplus.

2) Contagion among financial institutions

The second argument is that fair value accounting can provoke contagion in financial markets. The idea of this argument is that financial institutions in a crisis have to sell assets for a price that is lower than the fundamental value of the assets in order to meet their capital requirements. Because they dump the assets on the market, the market price of the asset drops. All other financial institutions that use mark to market accounting and have assets that trade in the same market now have to mark down the value of their assets to the market price. In this way these other financial institutions are ‘infected’ by the first institutions and the contagion spreads (Allen and Carletti, 2008; Plantin et al., 2008).

Thus, there are two ways in which fair value accounting could have contributed to pro-cyclicality and thereby to the financial crisis, being: 1) through the increased leverage financial institutions can take on in financial booms, and through 2) contagion among financial institutions because of fair value accounting. In the next chapter these two ways are analysed.

4. Analysis

The media’s opinion on what caused the financial crisis as is described in the second chapter is analysed in this chapter. The analysis is done based on the ways fair value accounting could have contributed to pro-cyclicality as described in chapter three. Thus, first the leverage increase in financially good times is analysed, and second the contagion among financial institutions is analysed.

(17)

17

4.1 Leverage increase in financial booms

Because of its proclaimed contribution to pro-cyclicality, the connection between fair value accounting and the leverage increase in financially good times is analysed. The problem with this subject is that there has not been a lot of empirical research been done on the subject. Fair value accounting is said to contribute to increased leverage

in the following way.

In financial booms, financial institutions want to increase their leverage because it allows them to invest more. Financial institutions’ balance sheet values rise under fair value accounting in financial booms, in comparison to other accounting measures. Because of this, banks are willing to lend them more money, because the financial institutions have more assets to back up their loans. However, in case of a financial downturn, the chance that financial institutions pay back their loans decreases if they have increased their leverage, because the value of their capital in comparison to debt decreases. Now, it is harder to pay back their loans in case of a default of one of their securities (Berk & De Marzo, 2011). This means that financial institutions are subject to increased vulnerability. The financial market has therefore increased vulnerability, which makes financial crises more severe if they occur (Plantin et al. 2008).

Adrian and Shin (2008) made a model to point out, when marking to market accounting is used, changes in balance sheet size are strongly correlated to changes in leverage. They say that financial intermediaries adjust their balance sheets actively. This means that the leverage is high during booms and low during busts, thus leverage is pro-cyclical. From a firm’s perspective it can be understood as follows: when asset prices increase, the balance sheets of firms get stronger. The firms hold surplus capacity. They will try to find a way to exploit this surplus capacity by expanding their balance sheets. They take on more short term debt on the liabilities side, because debt is relatively cheaper for firms with a high enough capital buffer. On the asset side, firms try to find borrowers to expand their capacity. In the years leading up to the crisis this has been done to a substantial extent by even giving homeowners mortgages to buy a house that did not have proof of income and down payment at all. Because housing prices had always been rising, financial institutions thought that if the homeowners defaulted, they could sell the house at a profit and not get into trouble. The homeowners started defaulting on their loans, and banks sold their houses. However, as more houses were available on the market, the housing prices started to drop, and financial institutions started to lose money on mortgages. Because financial institutions

(18)

18

had given out mortgages to a substantial extent, the losses were high. The vulnerability of the increased leverage, was therefore higher than the financial institutions had recognized beforehand. Financial institutions had to sell assets to keep their capital ratios, which put them into the negative spiral as described before (Achayra, 2009; Diamond, 2008). This was one of the things that led up to the financial crisis. Furthermore, Adrian and Shin find that this pro-cyclicality does affect volatility and the price of risk. Furthermore, dealer balance sheet changes forecast changes in the volatility risk premium.

This evidence nonetheless has the implication that it has been done previously to the financial crisis. It can say something about the beginning of it, and only make conjectures about the rest of it. Still, Adrian and Shin argue that financial institutions increased their leverage prior to the financial crisis, and this is also what happened in real life. As is described above, financial institutions gave out mortgages which led to increased vulnerability, even though they did not recognize it until it was too late. The vulnerability in the end led to pro-cyclicality.

The following table summarizes the arguments that are made above to give an overview by author.

Table 4.1: Argument about the contribution of fair value accounting to the

leverage increase in financial booms

Author(s) Year of publication Kind of study Main argument

Berk, De Marzo 2012 Model Fair value

accounting leads to increased leverage.

Plantin et al. 2008 Model Increased

vulnerability leads to more severe crisis.

Adrian, Shin 2008 Model Fair value

accounting led to increased volatility which led to the crisis.

(19)

19

Putting these arguments together, fair value accounting is said to cause the leverage to increase in times of financial boosts and thereby increase volatility. This increased volatility is leads to a crisis that is more severe. There is however no empirical evidence that can support this model. Thus, I find no evidence to indicate that fair value accounting did not contribute to increased vulnerability and thereby increased volatility, which made the crisis more severe. Because of the models that say fair value accounting would contribute to a severe crisis, I am lead to conclude that it did.

4.2 Contagion among financial institutions

Fair value accounting is said to have a contribution to contagion among financial institutions, which leads to pro-cyclicality. In this chapter this is tested.

As argued by Allen and Carletti (2008), when mark to market accounting is used, there can be contagion that causes financial institutions to liquidate unnecessarily. The sales that are related to this are called fire sales. They use a model to explain that when the markets are illiquid and mark to market accounting is used, the volatility of asset prices directly affects the value of financial institutions’ assets. If the financial institutions were allowed to continue their assets until they are mature, there would not be a problem, because they could use mark to model accounting and did not have to mark down the asset values. However, because they are using mark to market accounting they have to mark down the prices of the assets to their market values. Hence, some financial institutions become insolvent. The model that Allen and Carletti use, does not take into account that central banks can provide liquidity and only investigates a market for long term assets, and therefore in other markets different outcomes could occur.

Khan (2010) also studies whether fair value accounting is associated with additional contagion in the banking industry by doing a cross sectional analysis. He finds, as Allen and Carletti did, that the increase in contagion is most severe during periods of market illiquidity. Therefore quoted prices (level one) should not be used when markets are disorderly or distressed. Namely, level two or three should then be used. This does not hold in a period of good financial times. Furthermore banks that are poorly capitalized or have a high use of fair value accounting have more chance of contagion. However, he names the importance of the fact that fair value accounting alone may not increase contagion among banks. As Plantin et al, (2008) argue: in an economy with more than one imperfection (fair value accounting and others), the

(20)

20

removal of fair value accounting alone may not be welfare improving. Namely, it could increase the negative effect of the other imperfections. Nevertheless, this is merely a suggestion, and it is not proven by research that when fair value accounting is removed from the economy, that this would increase the negative effect of other imperfections.

Plantin, et al. (2008) furthermore made a model to examine the influence of fair value accounting on financial stability. They show that the usage of mark to market accounting shortens the decision horizon of management, which is mostly because of the agency problem. The agency problem is based on the agency theory that consists of an agent and a principal. The agent has information about the conditions of the company, whereas the principal does not. The problem is that the agent, who is supposed to make the best choices for the principal, is self-interested. A self-interested manager does not always make the decisions that are best for the principal (Eisenhardt, 1989). When other financial institutions sell, management that is focused on short term accounting earnings foresees the falling asset prices, and starts selling themselves. This will only amplify the price fall. The agency problem relates to this, because the agent (management in this case) is self-interested and only wants to do good in the short run, because that is what he is paid for. Thus, the falling asset prices have spread to other institutions because of the agency problem, which means that management is only interested in the short run. This means that contagion takes place.

On the other hand, Badetscher (2012) did an empirical study to examine the relation between fair value accounting and contagion. He finds no evidence of contagion: The interrelated selling activity did not increase, it even decreased during the financial crisis in his studies. A weakness of this study is however that it makes no distinction between risky and riskless securities. Therefore it is possible that financial institutions sold more risky securities than riskless ones. This could imply that interrelated selling activity for risky assets did not decrease. Hence, there could be a relationship between fair value accounting and contagion for risky securities. There is however no evidence that says it did, so I cannot assume so.

Shaffer (2010) says that there is no relationship between contagion and fair value accounting. He finds no indication that financial institutions were encouraged by fair value accounting to sell assets at distressed prices that could indicate a pro-cyclical effect. The problem with his article is however that it only contains a small sample of only fourteen of large US based banks, making it less credible.

(21)

21

The following table summarizes the arguments that are made in this paragraph to give an overview by author.

Table 4.2: Arguments about the contribution of fair value accounting to

contagion among financial institutions

Author(s) Year Kind of study Argument

Allen and Carletti 2008 Model Fire sales.

Khan 2010 Analytical Most severe in

periods of market illiquidity.

Fair value accounting alone may not increase contagion.

Plantin et al 2008 Model The removal may not

be welfare improving. Fair value accounting and agency problem shorten decision horizon. Shorter decision horizon spreads contagion. Badetscher 2012 Empirical evidence There is no relation. Shaffer 2010 Empirical evidence There is no relation.

Small sample used.

Putting all of these arguments together, Allen and Carletti say that when fair value accounting is used, there can be contagion that leads financial institutions to liquidate unnecessarily. Khan finds in a cross sectional analysis that contagion is most severe during periods of market illiquidity. Plantin et al. made a model that shows that the usage of mark to market accounting and the existence of the agency problem shortens

(22)

22

the decision horizon of management. Management with a shorter decision horizon starts selling when other institutions start selling. Hence, contagion spreads. On the other hand, Khan argues that fair value accounting alone may not increase contagion among banks. Badetscher and Shaffer find in empirical studies that fair value accounting did not increase contagion. However, Badetscher finds it both for risky and riskless securities and Shaffer uses only a small sample. Furthermore, Khan argues that it may not be just fair value accounting alone that increases contagion. Thus, there are two empirical studies that indicate that fair value did not contribute to contagion in the recent financial crisis, and two models and an empirical analysis that claim it does in general. The empirical studies overrule the models, because the models are only general theories and the empirical studies are specifically about the last crisis. Therefore, I draw the conclusion that I have no reason to assume that fair value accounting contributed to pro-cyclicality through contagion.

Concluding, there are two main arguments in which way fair value accounting potentially has contributed to the financial crisis: 1) through increased leverage in financial booms, and 2) through the financial contagion it can provoke. I find no evidence to assume that fair value accounting did not contribute to increased leverage in financial booms and therefore to pro-cyclicality. I do find evidence to assume that fair value accounting did not contribute to pro-cyclicality through contagion.

5. Discussion

In chapter four, the contribution of fair value accounting to pro-cyclicality is analysed by looking at two ways it could have contributed. These ways are:

1) The increase in leverage that financial institutions take on in times of financial booms when fair value accounting is used, which increases volatility.

There are however other ways in which fair value might have impaired or enhanced the financial crisis. These ways will be discussed in this chapter.

(23)

23

As Badetscher (2012) shows with his empirical paper, fair value accounting did not affect the banking industry in the way that is commonly alleged by critics. In his empirical paper he finds mixed evidence of pro-cyclical activity at firm level by examining a large sample of 150 banks. Sales of securities are correlated with other than temporary investments and bad debt expenses and with decreases in capital ratios. This implies under these circumstances there is a relationship between fair value accounting and pro-cyclicality. However, banks with low capital ratios engaged in less selling than banks with high capital ratios. This implies that under these circumstances there is no relation between fair value accounting and pro-cyclicality. Plantin et al (2008) on the other hand argue that in liquid markets fair value accounting has advantages of more timely and relevant information. Because of this more timely and relevant information Allen and Carletti (2008) say that fair value accounting can give an early warning signal in case of a crisis. Therefore, financial institutions can be earlier in taking appropriate measures. In this case fair value accounting can actually help to reduce the severity of a financial crisis.

Furthermore, it is not clear that the pro-cyclical effect of lending caused by financial institutions taking increased leverage in booms, means that the accounting standards should change. Namely, also the regulations could for instance be changed in a way that financial institutions are not allowed to take on extra leverage in financial booms, or that they have to build up extra capital reserves in financially good times. In addition to that, Laux and Leutz (2009) argue that they do not think that the tendency of financial institutions to expand leverage in booms is an issue that merely arises under fair value accounting. They however have no evidence, but this could be an issue for further research. They say that a financial institution can also increase its leverage in a boom under historical cost accounting. They can do this by selling an asset and retaining only a small claim in it.

Even though I do not examine the differences between fair value accounting and historical cost accounting, this statement can be of importance. Namely, if historical cost accounting would have made financial institutions take on the same leverage, it is not fair value accounting that contributed to pro-cyclicality. Then it is probable that it is something else that contributed to pro-cyclicality. Yet, it is important to add that the only evidence that financial institutions would have taken on the same leverage under historical cost accounting as they did under fair value accounting, are merely the thoughts of Laux and Leutz. Therefore, this is an interesting subject that

(24)

24

can be studied in the future, as I will come back to in the conclusion.

Laux and Leutz (2009) say a way to avoid pro-cyclicality is to deviate from market prices in situations where contagion is likely to occur. The accounting standards (IFRS and US GAAP) allow for such deviations to be made. Market prices from forced sales should not be used, which protects against deviation. The standards also allow for deviations from mark to market accounting to mark to model accounting to be made under level two and three. Furthermore, the standards introduced softening of fair value accounting at the end of the crisis to re-classify fair value assets into a category to which HCA and less stringent impairment tests apply. However, one should note that this has been introduced at the end of the crisis and therefore could not be used from the beginning on. Nevertheless, IFRS and US GAAP both have a mechanism that protects negative spill overs, and thereby contagion, from happening in distressed markets.

However, deviations from market prices require substantial judgment by auditors and preparers. The managers, directors and auditors face severe legal penalties if the deviation from the market price cannot be justified. This can be seen when looking at the Sarbanes-Oxley act of 2002.

The problem when deviating from mark to market to mark to model accounting is that it is easy to game the system for managers. It is difficult for standard setters to distinguish between a situation in which a manager claims that he has to use mark to model accounting and a situation in which a market price is indeed misleading.

Managers have an information advantage over the regulators and therefore it is difficult to write fair value accounting standards when it is needed and constrain managers’ behaviour when it is not needed. This can also be seen as the trade-off between relevance and reliability, where mark to market prices are easier to verify, but mark to model prices may be more relevant. Thus, in a world that has an information advantage for managers, the optimal fair value standards supposedly contain deviations from market prices, that would have been permitted without an information asymmetry (Laux & Leutz, 2009).

The following table summarizes the arguments that are made in this paragraph to give an overview by author.

(25)

25

Table 5.1: General arguments about the contribution of fair value accounting to the

financial crisis

Author(s) Year Kind of study Main argument

Badetscher 2012 Empirical Mixed evidence.

Plantin et al. 2008 Model Timely and

relevant information.

Allen and Carletti 2008 Model Early warning

signal in case of crisis.

Laux & Leutz 2009 Literature review Accounting

standards should not change.

Accounting

standards allow for deviation in an illiquid market. Optimal fair value standard contains deviations from the market price.

Sarbaines & Oxley

2002 Law Deviation from

mark to market accounting requires substantial judgment, otherwise penalty.

Putting all of these arguments together, Badetscher argues that fair value accounting contributed to pro-cyclicality under some circumstances. In other circumstances he says that it did not contribute. Plantin et al. argue that a benefit of fair value accounting is that it provides timely and relevant information. Allen and Carletti add that it could

(26)

26

be an early warning signal in case of crisis due to the timely and relevant information. Laux and Leutz say that the fact that financial institutions take on extra leverage in financial booms, does not mean that accounting standards need to change, because the same could have happened under historical cost accounting. However, these are merely conjectures, and therefore cannot serve as evidence. They furthermore say that accounting standards allow for a deviation from mark to market accounting when markets are in distress. A lot of managers did not deviate however because it requires severe judgment by auditors to do so and the penalties of unjustified deviation are severe according to SOx 2002. Thus, looking at the discussion, fair value accounting contributed to pro-cyclicality under some circumstances. However, this does not mean that it should be suspended. Namely, it could be an early warning signal in case of a crisis due to timely and relevant information. Managers furthermore could have deviated from mark to market accounting, but a lot of them did not do so.

Thus, fair value accounting did contribute to pro-cyclicality and therefore to the financial crisis, as the media said it did. I find the same in the analysis chapter. Furthermore, I find that this does not mean that it should be suspended.

6. Conclusion

During the financial crisis, a lot of critique has been given on financial accounting. This critique primarily came from the media that claimed that fair value accounting caused pro-cyclicality that, in its turn, contributed to the financial crisis. There have been studies to examine the effect of fair value accounting on the financial crisis, but there have been no studies that look at the relationship between the crisis and the fair value accounting by starting with conjectures from the media’s perspective. Furthermore, these studies led to contradictory belief on the subject. This study reviews these different views in the academic literature about the media’s conjectures that fair value accounting contributed to pro-cyclicality and thereby to the financial crisis. It tries to determine whether the conjectures that the media make about fair value accounting in the financial crisis are right according to the academic literature. There is little empirical evidence available, so a lot of the evidence that is found is based on models, and not based on empirical evidence.

(27)

27

In the academic literature I find that fair value accounting could have contributed to pro-cyclicality in two ways. These two are used to analyse whether fair value accounting contributed to the financial crisis in the way that the media claim. The first of those two is that fair value accounting has contributed to the increase in leverage that financial institutions take on in times of financial booms. The second is that fair value accounting contributes to contagion among financial institutions in a financial crisis.

I find unilateral evidence about the increased leverage that financial institutions are said to take on in times of a financial crisis and that thereby increases the volatility. Adrian and Shin (2008) find that fair value accounting causes this increased leverage in financial booms. Therefore I draw the conclusion that fair value accounting has contributed to pro-cyclicality through the extra financial leverage and the accompanying extra volatility that financial institutions take on in times of financial booms.

With respect to the criterion of contagion among financial institutions, I find mixed evidence. Mark to market accounting can cause contagion that leads financial institutions to liquidate unnecessarily in case of long term securities. In periods of market illiquidity, as the financial crisis was, this contagion is the most severe. This evidence is based on models however. There is other evidence that is based on an empirical study. This evidence says that fair value accounting did not contribute to contagion among financial institutions. The evidence that claims that fair value accounting did not contribute to pro-cyclicality through contagion is an empirical paper, and the other evidence is based on model. The empirical studies overrule the models, because the models are only general theories and the empirical studies are specifically about the last crisis. Therefore, I draw the conclusion that I have no reason to assume that fair value accounting contributed to pro-cyclicality through contagion.

In the discussion I discuss other ways in which fair value might have impaired or enhanced the financial crisis. I find that fair value accounting did contribute to pro-cyclicality and therefore to the financial crisis, as the media said it did. I find the same in the analysis chapter. This does however not mean that it should be suspended. The analysis of the two aforementioned ways that fair value accounting contributed to cyclicality indicates that fair value accounting contributed to pro-cyclicality and thereby to the financial crisis through the increased leverage that financial institutions take on in financial booms. Fair value accounting does not

(28)

28

contribute through contagion. Therefore, I can answer my research question of whether or not fair value accounting contributed to the financial crisis in the way that the media say it did. The media say that fair value accounting contributed to the financial crisis through pro-cyclicality. I find the two abovementioned ways in the academic literature. I find that fair value accounting contributed to pro-cyclicality through the two ways and therefore I come to the conclusion that fair value accounting contributed to the financial crisis in the way that the media claim it did.

If the same would have happened if the financial institutions had used historical cost accounting could be something for future research. Another issue that can be investigated in future research is whether the regulations instead of accounting standards should have been adapted so that financial institutions cannot take on more leverage in financial booms. Furthermore, there is very little empirical evidence about the contribution of fair value accounting to pro-cyclicality and therefore, this could be examined in future research.

7. References

Adrian, T., & Shin, H. (2008). Liquidity and Leverage. SSRN Journal. doi:10.2139/ssrn.1139857

Allen, F., & Carletti, E. (2008). Mark-to-market accounting and liquidity pricing. Journal

Of Accounting And Economics, 45(2-3), 358-378. doi:10.1016/j.jacceco.2007.02.005

Barth, M., Beaver, W., & Landsman, W. The Relevance of the Value Relevance Literature For Financial Accounting Standard Setting: Another View. SSRN

Journal. doi:10.2139/ssrn.246861

Barth, M., Landsman, W., & Wahlen, J. (1995). Fair value accounting: Effects on banks' earnings volatility, regulatory capital, and value of contractual cash flows.

Journal Of Banking & Finance, 19(3-4), 577-605.

(29)

29

Badertscher, B., Burks, J., & Easton, P. (2012). A Convenient Scapegoat: Fair Value Accounting by Commercial Banks during the Financial Crisis. The Accounting

Review, 87(1), 59-90. doi:10.2308/accr-10166

Benston, G. (2008). The shortcomings of fair-value accounting described in SFAS 157.

Journal Of Accounting And Public Policy, 27(2), 101-114. doi:10.1016/j.jaccpubpol.2008.01.001

Berk, J., & De Marzo, P. (2011). Corporate finance. Boston [etc.]: Pearson.

Eisenhardt, K. (1989). Agency Theory: An Assessment and Review. Academy Of

Management Review, 14(1), 57-74. doi:10.5465/amr.1989.4279003

FASB (Financial Accounting Standards Board), 2006. Fair value measurements. Statement of Financial Accounting Standards No. 157, September.

FASB (Financial Accounting Standards Board), 2007. The fair value option for financial assets and financial liabilities – including an amendment of FASB Statement 115. Statement of Financial Accounting Standards No. 19, February.

Forbes, S. (2009). Steve: End Mark-To-Market. Forbes. Retrieved 8 June 2015, from http://www.forbes.com/2009/03/20/steve-forbes-mark-to-market-intelligent-investing-market.html

Georgiou, O., & Jack, L. (2011). In pursuit of legitimacy: A history behind fair value accounting. The British Accounting Review, 43(4), 311-323. doi:10.1016/j.bar.2011.08.001

Grauwe, De, P. (2015). Act now to stop the markets' vicious circle - FT.com. Retrieved 8 June 2015, from http://www.ft.com/intl/cms/s/0/12d5e436-f61e-11dc-8d3d-000077b07658.html#axzz3ZN09zSzQ.

International Accounting Standards Board (IASB). 2006. International Financial

Reporting Standards No. 13: Fair value measurement. London: IASB.

Khan, U. (2010). Does Fair Value Accounting Contribute to Systemic Risk in the Banking Industry?. SSRN Journal. doi:10.2139/ssrn.1911895

Landsman, W. Fair Value Accounting for Financial Instruments: Some Implications for Bank Regulation. SSRN Journal. doi:10.2139/ssrn.947569

(30)

30

Laux, C., & Leuz, C. (2009a). The crisis of fair-value accounting: Making sense of the recent debate. Accounting, Organizations And Society, 34(6-7), 826-834. doi:10.1016/j.aos.2009.04.003

Lee, C. (2015). Why Fair-Value Accounting Isn’t Fair. Retrieved 8 June 2015, from https://www.gsb.stanford.edu/insights/charles-lee-why-fair-value-accounting-isnt-fair

Magnan, M. (2009). Fair Value Accounting and the Financial Crisis: Messenger or Contributor?. Accounting Perspectives, 8(3), 189-213. doi:10.1506/ap.8.3.1 Nelson, K. K. (1996). Fair value accounting for commercial banks: An empirical

analysis of SFAS No. 107. Accounting Review, 161-182.

Poel, v. (2008). Fair value: nood breekt wet. Accountant.nl. Retrieved 15 May 2015, from https://www.accountant.nl/opinie/2008/10/fair-value-nood-breekt-wet/ Pozen, R. (2009). Is It Fair to Blame Fair Value Accounting for the Financial Crisis?.

Retrieved 21 May 2015, from https://hbr.org/2009/11/is-it-fair-to-blame-fair-value-accounting-for-the-financial-crisis

Ryan, S. (2008). Accounting in and for the Subprime Crisis. The Accounting Review,

83(6), 1605-1638. doi:10.2308/accr.2008.83.6.1605

Sarbanes, P. & Oxley, M. (2002). Public Company Accounting Reform and Investor

Protection Act. Washington.

Shaffer, S. (2010). Fair Value Accounting: Villain or Innocent Victim - Exploring the Links Between Fair Value Accounting, Bank Regulatory Capital and the Recent Financial Crisis. SSRN Journal. doi:10.2139/ssrn.1543210

Stolowy, H., & Lebas, M. (2006). Financial accounting and reporting. London: Thomson.

Wallison, P. (2009). EconoMonitor : EconoMonitor » Fair Value Accounting – A

Critique. Retrieved 8 June 2015, from http://www.economonitor.com/blog/2009/01/fair-value-accounting-a-critique/ Whalen, R. The Subprime Crisis: Cause, Effect and Consequences. SSRN Journal.

Fair value accounting and the financial crisis : literature review of the media’s conjectures about the relationship between fair value accounting and the financial crisis