What shapes the distribution of the cost of a financial sector rescue plan? : policy choices of the Dutch government to rescue the financial sector in 2008

(1)

What shapes the distribution of the cost of a Financial Sector Rescue Plan?

Policy choices of the Dutch Government to rescue the Financial Sector in 2008

Adriaan den Engelse – 11259078

Master Thesis in Political Economy at the University of Amsterdam

Supervisors: Dr. Jasper Blom and Dr. Farid Boussaid

Word count: approximately 18,227

(2)

2

(3)

3

Abstract

The financial sector and policy makers were unprepared for the 2008 financial crisis. When the crisis struck policy makers had little time to reflect and had to act quickly to keep their national financial sector afloat. Some countries were more successful in containing the financial crisis than others. Some countries were also more effective in having the financial sector contribute towards the cost of their own rescue. Financial stability could be considered a public good. Financial institutions cannot be excluded from benefitting from financial stability without having contributed towards it. This provides an incentive for strong financial institutions to remain inactive and therefore forcing the government to bail out the weaker financial institutions at the cost of the public budget. Some countries had the norms and routines for the financial sector and government to coordinate their response to the financial crisis. Other countries forced their national financial sector to act collectively as they had the ability to coerce all major banks. Using archival research this study reveals that the Netherlands did not have required norms and routines for the government and the financial sector to

coordinate a response to the financial crisis. As a result, most of the cost of the financial sector rescue plan went to the public budget.

(4)

4

Acknowledgements

A special thanks to my supervisor Jasper Blom for proving guidance and inspiration. I am grateful to all my fellow students for the good times and challenging debates.

(5)

5

1. Introduction

The 2008 financial crisis was worst crisis in eighty years.1_{Nearly a decade later its economic, social and political}

consequences still have a significant effect on society. Both the financial sector and policy makers did not foresee the financial crisis and were unprepared. All developed economies need a stable and well-functioning financial sector. When the crisis started, policy makers had little time to reflect and debate on the most effective approach to keep its financial sector afloat. In many cases the negotiation processes took place over a chaotic weekend to establish a financial sector rescue plan. The financial crisis forced policy makers to

intervene in their financial sector to prevent it from failing. These rescue packages were often a mixture of guarantees, capital injections and the state taking over troubled assets. National governments sometimes particularly hindered by their insufficient legal powers to intervene. Some states were more successful in providing financial stability at a lower cost to the public budget than other states. The intuitive explanation is that the size and exposure of the financial sector are the main factors which determine the cost of a financial sector rescue programme. The countries were the financial sector and national governments worked together were considerable more likely to provide financial stability at a lower cost (see Grossman and Woll, 2014; Culpepper and Reinke, 2013; Kluth and Lynggaard, 2013).

The financial sector enjoys a privileged position within society due to its power to control production; all developed economies need credit creation to function properly. The cost of non-intervention in the financial sector would therefore most likely by far exceed the cost of an intervention. The cost of the financial sector rescue plans would often exceed the cost of any other major government expenditure such as national defence and healthcare. Financial stability should be considered as a public good as it is rivalrous and non-excludable (Buchanan, 1965). Figure one below depicts the theory of clubs as developed by Buchanan in 1965. There is one major complication in the provision of public goods: free-riding.

1_{The definition of a banking crisis is taken from Reinhart and Rogoff (2009) when one or both of the following}

events occur: (i) bank runs that lead to the closure, merging, or takeover by the public sector of one or more financial institutions; or (ii) if there are no runs, the closure, merging, takeover, or large-scale government assistance of an important financial institution (or group of institutions) that marks the start of a string of similar outcomes for other financial institutions.

(7)

7

Figure one – The theory of clubs.

Free-riding is a complicated problem to overcome. An example of free-riding on a public good is street lighting, which is provided by the government. If citizens do no pay their fair share in taxes they cannot be excluded from using the street lighting. This results in a collective action problem. When all parties believe it is in their interest to contribute towards a collective good they will contribute towards it. Such conditions are almost exclusively confined to conditions of existing close cooperation in a small group or conditions in which a single actor receives more utility from the public good than it costs that actor to provide the public good in its entirety (Olson 1965:33).

When a financial sector faces severe distress investors and depositors no longer seek a return on capital but a return of capital. This results in capital to flow to places which are deemed to be the safest: the financially strongest banks with the most stellar reputation. Alternatively, depositors withdraw their funds and hide under their matrass. Business should be Darwinian and promotes the survival of the fittest. Failure should therefore be allowed. Bank failures could severely impair an economy as depositors lose their money, credit creation grinds to halt and confidence is lost. Especially in modern finance were often the ten largest financial institutions often control over half of the financial sector. A financial crisis is therefore likely to make the strong financial institutions stronger and could force the weaker ones into insolvency. This also further incentivises the stronger financial institutions to (openly) speculate against the weaker ones. The national government has than little option than to bail out the weaker institutions at the cost of the taxpayer. As a result, the healthy financial institutions not only free-ride; they actually receive a free-riders bonus.

The failure of the giant hedge fund Long Term capital management is another good example. When the fund was about the fail in 1998 it would send financial markets into a further tailspin following the default of Russia at the time. The US Federal Reserve Bank orchestrated a bailout involving all the large US financial institutions. The rescue plan did not require any public capital or guarantee. Yet the rescue plan nearly fell through as Bear Stearns declined the participate in the rescue scheme while all other banks did (Lowenstein,

Excludable Non-excludable

Rivalrous Private goods

Cars, shoes, clothing

Common pool resources Fish, timber, coal

Non-rivalrous Club goods

Cinemas, swimming pools

Public goods Street lighting, national defence, financial stability

(8)

8

2001). Other financial institutions had to commit 500 million US Dollar to the rescue plan and in effect got the ‘suckers’ pay-off as Bear Stearns was able to free-ride.

The theory of collective action (Olson, 1965) has been widely empirically tested by Ostrom (2000). Ostrom (2000) argues the cooperation is more likely when; first, participants have faith in the likelihood of cooperation of other participants to contribute towards collective solutions. Second, personal communication considerably increases the likelihood of cooperation. Third, learning about the game increases cooperation. Fourth, participants are willing to pay in order to punish other participants which make a below average contribution. Woll (2014:59) therefore argues that collective action in the financial sector is unlikely because; first, the believe that other financial institutions will not contribute. Second, there is no personal

communication between the financial institutions. Third, financial institutions have not experienced a similar situation before. Fourth, it is expected that individual financial institutions to be willing to accept expenditures to have control mechanisms over the behaviour of other financial institutions.

The objective of this research is to determine what shaped the distribution of the cost of the financial sector rescue plan in the Netherlands in 2008?

This research uses a qualitative evidence to improve our understanding of what shaped the distribution of the cost of the financial sector rescue package which was shaped in the Autumn of 2008. This paper this divided into three main chapters. Chapter one discusses the need to study financial sector rescue plans. In addition, chapter one provides a theoretical framework and research methodology. Chapter two discusses the existing literature on the causes of the 2008 financial crisis, the difficulties of implementing a financial sector rescue plan and the factors which shape the distribution of the cost of a financial sector rescue plan. Chapter three provides a case study of the financial sector rescue plan which was shaped in Autumn 2008 in the Netherlands. The case study provides qualitative evidence of what shaped the distribution of the cost of the financial sector rescue plan.

Why study Financial Sector Bailouts and Bailout Cost

Relatively little research has been performed into solving the 2008 financial crisis in comparison to what caused it. There is little consensus on what an effective financial sector rescue plan is. The government actions which follow a financial crisis should minimise the cost to the public budget and subsequently the taxpayer. The public budget is very likely to suffer because of a financial crisis as this leads to a recession or worse depression. A recession or depression results in lower government tax revenue and higher spending due to increased welfare payments such as unemployment insurance. The financial sector often disproportionally benefits from economic growth while the cost of an economic downturn is often distributed via the public budget to all taxpayers.

(9)

9

Denmark and Ireland stand out when comparing their policy responses to the financial crisis. Both Denmark and Ireland are of similar size and economic development. Both countries had limited sovereign debt, sound fiscal balances and a bloated financial sector which was heavily exposed to domestic real-estate lending. Both countries allowed their financial sector to draw several multiples of annual GDP in terms capital and guarantees; the Danish financial sector required 67% while Irish financial sector drew capital exceeding 260% of GDP. This resulted in fiscal crisis and Ireland lost access to international capital markets and was forced to go cap in hand to the International Monetary Fund and European Central Bank. Ireland paid a horrific price and had its sovereign debt nearly triple within a very short time span. This resulted in a prolonged and deep recession, unemployment and caused many young Irish to look for employment overseas. Figure two below depicts a ranking of the cost of the financial sector rescue plans per country as a percentage of GDP. Figure two – Financial sector rescue plan cost 2007 – 2010 as percentage of 2010 annual GDP

Source: Woll (2014:32)

The US spent 5.7% of its annual GDP on its financial sector rescue plan. The Vietnam War cost 6.8% of US GDP and the entire Marshall Plan a ‘modest’ 5.2% of GDP. The financial crisis started in the US as the result of a real estate bubble which popped. The US spend a modest amount as a percentage of GDP in comparison to some European countries. Unfortunately, financial sector rescue plans still matter today. In November 2014 at the G20 meeting, world leaders agreed to end bank bailouts. Any future bailout will have to be bail-in, as what happened in Cyprus in 2012. Various countries adopted legislation in support of ending bank bailouts following the G20 meeting. Italy went in the opposite direction by providing 6.6 billion euro of public money to Monte dei Paschi di Sienna in June 2017.

327.5% 256.0% 92.2% 53.0% 46.9% 53.0% 17.6% 24.8% 18.2% 24.9% 29.9% 268.5% 67.2% 20.5% 16.1% 16.9% 16.1% 11.9% 10.1% _6.0% 5.7% 0.1% 0% 50% 100% 150% 200% 250% 300% 350%

Ireland Denmark Belgium United Kingdom

Greece Netherlands Luxembourgh Germany France United States Finland

Committed Actual

(10)

10 Theoretical Framework and Research Propositions

The objective of this paper is to explain what shaped the distribution of the cost of a financial sector bailout which was designed and implemented in the Netherlands in Autumn 2008. Financial sector rescue plans often involve huge cost; it is the question whether (some) of these costs are distributed back to financial sector or whether the taxpayer is left with the bill. This paper will therefore analyse the policy choices of the

Netherlands which were made in autumn 2008, the peak of the financial crisis in Europe.

In 2014 Cornelia Woll published her book titled the theory of inaction: bank bailouts in comparison. This book is primarily based on three previous studies: Grossman and Woll, 2012; Kluth and Lynggaard, 2013 and Culpepper and Reinke, 2014. All three papers have a clear consensus; the cost of the financial rescue packages depended on more than economic circumstances and the exposure of the financial sector alone. The three papers also agree that the cost of the financial rescue plan to taxpayers was lower when the financial sector and the government worked together to coordinate their response.

During times of severe financial distress in the financial sector, depositors flee to financial institution which they consider to be safest. This creates a ‘winner-take-all’ effect for the strongest financial institutions while the other institutions get weaker. In modern finance the cost on a non-intervention is very likely to far exceed the cost of an intervention. The government intervention, or financial sector rescue plan, often comes at the cost of the taxpayer. Both the financial sector and the national government have an urgent need to restore financial stability, both pay a very high price for possible failure. The question remains with the

provision of a public good is the distribution of the cost. When the financial sector and policy makers organised a coordinated response the cost of the overall intervention tended to be lower (Woll, 2014).

Coordinated interventions tend to have overall lower cost due to a variety of reasons. It enables the healthier financial institutions to cross-subsidise the less healthy financial institutions. This enables financial institutions to recapitalise cheaply (Culpepper and Reinke, 2014). Perhaps even more important is the fact that it provides a disincentive for the stronger financial institutions to (openly) speculate against the weaker ones. This in turn prevents a ‘winner-take-all’ effect were the strong financial institutions get stronger and the weaker ones need to be bailed out at the cost of the taxpayer. Grossman and Woll (2012) and Woll (2014) do not provide any counter arguments when the cross-subsidised funding of financial institutions could do more harm than good. In the absence of severe financial distress the weak financial institutions can drain the healthier ones. Italy created a financial stability for its financial sector. The existence of this fund is subject to criticism as it harms the healthy financial institutions and impairs the competitive working of the market.

Grossman and Woll (2012) argue that the distribution of the cost of a financial sector rescue plan is determined by game theory, or a game of chicken. Woll (2014) turns the theory of collective action by from Olson (1965) on its head. If the financial sector would not act collectively it forced the national government to pick up the broken pieces. The cost of the financial sector rescue package would then shift to the taxpayer.

(11)

11

Financial institutions need to be able to recapitalise to continue to lend money to boost economic activity. Healthy financial institutions benefit from a stable financial system, yet they also have an incentive not to collaborate in a coordinated response as they can free-ride. Woll (2014:62) defines three outcomes of how the relationship between the financial sector and the government influence the success a financial sector rescue plan:

1. When the financial sector is capable of organising collectively and contributing to its own rescue, this will contain cost and manage a financial crisis in the least painful way to the public budget.

2. When a collective commitment from the financial sector is lacking it is best for a government to unilaterally impose a government solution.

3. A combination of a very unbalanced regulatory approach, collective action of the industry may contribute little to resolving a financial crisis.

Woll (2014) deemed the dynamics of the bailout negotiation a giant game of chicken. Please note that a game of chicken is different from the prisoner’s dilemma as the mutual defection option has a far worse pay-out for each participant.A game of chicken is a game-theoretical ideal type which takes its name from a challenge of daring in which two challengers drive cars towards each other in the same lane. Both face disaster if they have a frontal collusion but the winner is the driver who does not swerve to avoid the collision. The game takes its name from the act of calling the loser a “chicken”, which implies coward. One winning strategy is to make your opponent think you have no control over your car, which forces them to steer clear to save their life. (Rapoport and Chammah, 1966). In the 2008 financial crisis, both the financial sector and government engaged in brinkmanship surrounding the crisis interventions in an attempt to make the other party pay a larger share of the bailout costs. Table one below depicts the pay-out matrix for a game of chicken.

Table one - Standard Pay-out Matrix for a game of chicken

Financial sector, State State cooperates State defects

Financial sector cooperates 0, 0 -5, 5

Financial sector defects 5, -5 -10, -10

The pay-off table of a standard game of chicken does not reflect one major limitation; if a very large financial institution is in serious trouble it cannot be saved by the financial sector. Two good examples would be the Irish lender Anglo Irish Bank. At the time of its failure the balance sheet of Anglo Irish Bank exceeded the annual output of the entire Celtic economy. Another example is the failure of Lehman Brothers. At the time of its failure the financial sector could not really act any more. JPMorgan had absorbed two major competitors. Many of the other financial institutions had recapitalise from private sources at a high cost. Bank of America was the only player left which could still act and used the opportunity to acquire Merrill Lynch instead of Lehman Brothers.

(12)

12

This demonstrates an important point, it is not the strength but the weakness of private actors which policymakers must fear. Rodrik and Zeckhauser (1988) called this “the dilemma of government

responsiveness”. Private actors can exploit their national government if these private actors are more valuable as a going concern. The concern was the perceived or real – importance of the financial sector in the economy.

The pay-off table can be adjusted when both the financial sector and policy makers cooperate. The possible pay-off depends on the situation. Both the financial sector and the national government benefit from financial stability. However, a healthy financial institution could benefit more from a failed or bailed out competitor than financial stability. The exact pay-off value when both parties’ corporate is debatable (Snyder, 1971).

Table two - Bailout game of chicken pay-off table

Financial sector, State State cooperates State defects

Financial sector cooperates 2, 4 -7, -7

Financial sector defects 5, -5 -10, -10

Denmark and France are both examples were the financial sector and their respective states collaborated (Grossman and Woll, 2012). In both cases all major participants were required to provide capital for a financial stability fund. This fund would absorb the first losses when weaker participants of the financial sector would require bailout capital. Denmark still paid a high price for the financial rescue plan but this could have been a lot worse considering its bloated financial sector which was heavily exposed to a domestic real estate bubble. France spend relatively little on its financial sector rescue plan in comparison with other European countries.

The United States originally achieved a similar position as Denmark and France except it allowed Lehman Brothers to fail. The financial authorities tried the strategy that they had no control in the game of chicken, like throwing the steering wheel out of the window of a driving car. Policy makers had been very clear to the financial sector that they would and could not intervene any more. The Chairman of the US Federal Reserve Bank at the time, Ben Bernanke, is adamant that the government did not have the legal means to bailout Lehman Brothers. The US Treasury could not spend without congressional approval and the Federal Reserve Bank could only lend to solvent businesses, and, claims Bernanke, Lehman Brothers was insolvent at the time (Bernanke, 2015). The financial sector also defected and both participants lost.

The United Kingdom and Germany could not get its financial sector to act collectively. Both countries had one or more healthy financial institutions which did not require any aid from the government. In both cases the healthy institutions wanted to avoid the stigma of being bailed out by the government. It is worth noting that the financial sector can do whatever it takes in not having to accept government money. The British bank Barclays avoided having to accept money from the British government and opted to tap into a

(13)

13

fund from Qatar at a 14% annual interest charge. This was two percent higher than the already penal rate of twelve percent charged by the British government.

The game of chicken could explain the pay-off of the game, but it does not explain what drove the financial sector to defect. How could financial institutions defy their national regulators? Culpepper and Reinke (2013:3) define structural power as the ability to defy national regulators because of the internationalisation of their markets. In other words, a financial institution is in a strong position to defy a national regulator when the majority of its assets under management and revenue comes from abroad. The financial institution can then credibly threaten to scale back operations in its domestic market is less attractive. Scaling back operations could imply that less credit is created and jobs could be lost. Culpepper and Reinke (2013) argue that the major financial institutions in the United States could not escape the long arm of the financial regulators. As a result, all major financial institutions were forced to accept a government bailout, including the financially strong ones which did not needed it. On the contrary, in the United Kingdom this strategy failed as some of the London based banks could outmanoeuvre the British financial regulators.

Kluth and Lynggaard (2013) provide a different argument of why the financial sector and the government coordinate their response to a financial crisis; norms and routines. The authors argue that the Scandinavian financial crisis which happened during the 1990s created the norms and routines for the financial sector and government to coordinate a response against a financial crisis. These norms and routines constitute institutionalised legacies of collective action informing behavioural patterns which re-emerged during the 2008 financial crisis.

The contribution of this paper is to test both the proposition of structural power and the institutional explanation in one case study. Kluth and Lynggaard (2013) use the institutional explanation while Culpeper and Reinke (2013) use the structural power approach. Woll (2014) mainly applies the structural power of the financial sector to explain the distribution of the cost of a financial sector bailout. Woll (2014) acknowledges the argument of the relevant norms and routines developed by Kluth and Lynggaard but does not consistently apply this framework on other case studies other than the one performed by Kluth and Lynggaard (2013).

Figure three below provides an explanatory framework of what shapes the distribution cost of a financial sector rescue plan. Financial stability can be considered a public good. Using the theory of clubs from Buchanan (1965) this adds the complication of free-riding. During times of severe financial distress capital tends to flow to the strongest financial institutions with the best reputation. This creates a ‘winner-take-all’ effect while the government is forced to bailout the weaker institutions at the cost of the public budget. The overall cost of a financial sector rescue plan tends to be lower when the financial sector and the government coordinate their response. This implies that all major financial institutions receive government provided capital injections and in some cases, were all major financial institutions provide capital towards a government fund to promote financial stability. This provides a disincentive for financial institutions to (openly) speculates against each other. Strong financial institutions have an incentive to defect as they have the ability to free-ride.

(14)

14

This could be prevented when financial institutions cannot defy their national regulator, or when the norms and routines are in place for the financial sector and government to coordinate their response.

Figure three - Explanatory framework were the structural power of the financial sector and the norms and routines determine whether a financial rescue plan is coordinated or not.

The objective of this research is to explain what shaped the distribution of the cost of the financial sector rescue plan which was shaped and implemented in the Netherlands in autumn 2008. In the

Netherlands, there had not been a coordination between the financial sector and the government. Financial institutions were not required to provide capital to a financial stability fund, neither did all major financial institutions receive a capital injection whether they needed this or not. This resulted in the ‘winner-take-all’ effect; one financial institution became stronger while the cost of bailing out the weaker ones went to the public budget.

This paper has the objective to test the following two propositions:

Proposition one. The high level of structural power of the financial sector in the Netherlands prevented a coordinated response to the financial crisis.

Proposition two. The absence of the norms and routines prevented a coordinated response to the crisis. The next section of this paper provides an argument why the Netherlands was selected as a case study.

Banking crisis and financial instability

Financial sector and government can coordinate

Financial sector and government cannot coordinate

Problem of collective action is solved

Problem of collective action is not solved

Coordinated intervention which is

cheaper overall

Uncoordinated intervention which is more expensive overall

Banking crisis solved and financial stability

Financial instability, currency crisis, sovereign debt crisis Financial sector and government can coordinate due to lack of structural power

(15)

15 Case Selection

There has not been a coordinated response between the financial sector and the government in the

Netherlands. Financial institutions were not required to provide capital to a financial stability fund, neither did all major financial institutions receive a capital injection. The result was that the strongest financial institution became stronger while the cost of rescuing the weaker ones ended up in the public budget. A case study of what shaped the distribution of the cost of the financial sector rescue plan of autumn 2008 in the Netherlands could add significant value due to several reasons discussed below.

Policy makers in the Netherlands were in a very difficult position due to a variety of reasons. The Dutch financial sector was the tenth largest financial sector in the world. This made the Netherlands very vulnerable in case of a financial crisis as the country might not have the fiscal base to provide a rescue plan. Ireland suffered a sovereign debt crisis when it could not shoulder the cost of its financial sector rescue plan. Iceland suffered a currency crisis and complete meltdown of its financial system.

The Netherlands has one of the most concentrated financial sectors in the world; the five biggest financial institutions control 85% of the financial sector. At the time of the peak of the financial crisis in Autumn 2008 there were three major financial institution left: Rabobank, ING and Fortis / ABN Amro. At the time of the financial crisis in 2008 Fortis / ABN Amro was effectively insolvent and had to throw itself at the mercy of the Dutch government. ING also needed significant government support but enjoyed a very strong position in terms of structural power. Most of ING’s revenue came from abroad and this put ING in a strong position to defy the regulator as it could credibly threaten to scale back operations in the Netherlands if the climate would become unfavourable. Finally, the mutual bank Rabobank did not require a capital injection as it is mainly funded through deposits. As a mutual bank its shares were not traded on a stock exchange.

Rabobank benefitted hugely from the financial turmoil and experienced a significant inflow from depositors looking for a safe place. This would result in leaving the Dutch taxpayer with the cost of the rescue of ING and Fortis / ABN Amro.

Financial sector rescue plans (fortunately) are not implemented often. The last time the Netherlands had a systemic banking crisis was in the 1920s during the Economic boom which followed the First World War.2_{As a result, there was no institutional memory in the financial sector or with policy makers to counter a}

systemic financial crisis. Previous research into government policy choices from Kluth and Lynggaard (2013), Culpepper and Reinke (2014) and Woll (2014) compare the policy choices of six countries in three comparative case studies: United Kingdom and the United States, Germany and France and finally Denmark and Ireland. This study combines both propositions of the structural power of the financial sector and by analysis of the norms of routines of the financial sector and the government to coordinate a response to a financial crisis.

(16)

16 Research Methodology and Sources

It can take several decades before a bailed-out financial institution is fully privatised again. At the time writing nearly a decade after the peak of the financial crisis in 2008 the Dutch state still owns substantial parts of the financial sector. Most of the financial sector rescue plan had been repaid by 2014. The literature review in section two of this paper will discuss previous quantitative research of the financial sector rescue plans. Quantitative overviews help to clarify the most striking differences and glean the evolution of the bank support, the do not provide a detailed understanding of the stakes and realm of options governments faced (Woll, 2014:12).

It is difficult to draw reliable conclusions from the use of statistics when analysing financial sector rescue programmes due to three reasons. First, the statistical overviews prepared by international organisations such as the International Monetary Fund or the European Commission are preceded by

extensive bargaining over categorisation and accounting methods (Woll, 2014:12). Second, the data published on the bailouts are continuously updated or corrected. Two good examples are the bailout of the troubled German lender Hypo Real Estate. In the fall of 2011 the German finance minister Wolfgang Schäuble

announced that an accounting error had overstated the liabilities of unwinding Hypo Real Estate by 55 billion Euro (Wiesmann, 2011). Another example is the British Lloyds Banking Group which was bailed out in 2008 by receiving approximately 20.2 billion Pound. In Spring 2017 the British government received a total of 20.3 billion Pound and there made a ‘profit’ of approximately 900 million Pound. This amount of money does not consider any interest charges. Third, not all data on the bailouts is publicly accessible. The European financial authorities publicly announced all taken measures to keep the financial sector afloat. However, it remains confidential who the indirect beneficiaries of the bailouts were. This data is not confidential in the United States. As a result, it is known which financial institutions benefited from the bailout of US financial

conglomerate American International Group. This type of data is confidential in the European financial sector rescue programme. A good example is the much-criticised bailout of the Irish lender Anglo Irish Bank. Anglo Irish Bank was nationalised by the Irish government and paid all liabilities including junior bondholders. This issue is somewhat painfully captured in a working paper from the Bank of England (Rose and Wieladek, 2012). The authors used confidential data which is only available within the Bank of England to evaluate the bailout programme. As the authors were bound by the issue of confidentiality it cannot offer any significant

conclusion except the positive correlation between a bank size and bailout funds received.

This research therefore uses qualitative evidence to discover what shaped the distribution of the bailout cost of the financial sector rescue plan in the Netherlands. The term ‘sample’ means “the case, or cases, chosen for study, referred to collectively” (Gerring 2004). The case study of the policy response to cope with the financial crisis in the Netherlands in 2008 will be based on the history, structure and exposure of the Dutch financial sector. This followed by an overview of the size and bailout measures to a qualitative analysis

(17)

17

of the political decision-making process (Woll, 2014: 13). This analysis is based on extensive archival research of official policy documents, Parliamentary inquiries and special investigations.

The case studies performed by both Woll (2014) and Culpepper and Reinke (2014) about the financial sector rescue plans implemented in in the United Kingdom and the United States had a wealth of sources. Many politicians and key policy makers of that time published personal accounts. These personal accounts have added a wealth of information. Unfortunately, no personal accounts have been published in the

Netherlands. Woll (2014) conducted 30 interviews with key personal from financial authorities, central banks, politicians and banking sector officials to add additional information to the other sources. Due to the very high importance of the financial sector rescue programmes, these were shaped by heads of government, cabinet ministers and central bank governors. All these people are incredibly hard to contact for research interviews.

Case studies are subject to various research limitations. One is the credibility of the generalisation of the findings. I will make sure to demonstrate the extent to which this case is similar or different with others of its type (Denscombe, 2010:62). The most important data source for this case study is a Parliamentary Enquiry (in Dutch: Parlementaire Enquête Financial Stelsel). The enquiry started in June 2009 and was completed in April 2012. The committee provided an interim report in May 2010. The final report exceeded 700 pages. This excludes the public hearings which were taken by the commission (in Dutch: Verslagen Openbare Verhoren). The hearings were taken under oath. The interviewed included the former Prime Minister, Finance Minister, the President of the Dutch Central Bank and several high ranking civil servants from the Ministry of Finance and regulatory authorities which were tasked to manage the fall-out of the financial crisis. Board members of the largest financial institutions were also interviewed. All interviews had been recorded and are available in one document, a total of approximately 1900 pages. Other sources include in internal audit / evaluation report from the Dutch Ministry of finance. This report does not contain evaluate the policy choices but method of working in a crisis. The audit report concluded that mistakes could have been avoided had information been shared appropriately.

Concluding Remarks

This chapter has provided an explanatory framework of this research. The objective of this research is to determine what shaped the distribution of the cost of the financial sector rescue plan in the Netherlands in 2008. The argument is that when the financial sector coordinates its response with the government during a financial crisis the overall cost of the financial sector rescue plan are lower. This implies that all major financial institutions receive government provided capital injections and in some cases, were all major financial

institutions provide capital towards a government fund to promote financial stability. This enables the stronger financial institutions to cross-subsidise the weaker financial institutions. This enables financial institutions to recapitalise cheaply. This also provides a disincentive for the stronger financial institutions to (openly)

(18)

18

speculate against the weaker ones. This in turn prevents a ‘winner-take-all’ effect were the strong financial institutions get stronger and the weaker ones need to be bailed out at the cost of the taxpayer. Financial stability can be considered a public good. Participants cannot be excluded from benefitting from a public good even without having made contribution towards it. This could be prevented when financial institutions cannot defy their national regulator, or when the norms and routines are in place for the financial sector and

(19)

19

2. Literature Review

This chapter is divided into three sections. Section one provides a synopsis of the causes of the 2008 financial crisis. The point is to demonstrate that many of the causes of the 2008 financial crisis remain unaddressed; therefore, we could see another financial crisis in the future. Section two discusses the key literature on the policy responses which followed the financial crisis across the world. The objective is to demonstrate the difficulties in evaluating the success or failure of a financial sector rescue plan. Section three discusses the literature on what shapes the distribution of the cost of a financial sector rescue plan. In other words; do the cost go to the public budget or does the financial sector also significantly contribute? The argument of section three is that the distribution of the cost of a financial sector rescue plan are determined by the structural power of the financial sector and the norms and routines of the financial sector and the government to work together.

The Financial Crisis: A Synopsis

The financial crisis started in the United States when the first signals of financial distress were seen in early 2007. It would take to 2008 before the crisis spread to Europe and other parts of the world. The failure of the American investment bank Lehman Brothers on 15th_{of September 2008 is often considered to be the peak of}

the financial crisis. Mono-causal explanations fail to explain what caused the financial crisis as there were a variety of interdependent causes. In the years prior the financial crisis the majority of advanced economies had an uninterrupted upward trend of real estate prices. Real estate prices kept appreciating due to a variety of reasons; financial innovation in the form of derivatives and asset securitisation, government policies encouraging home-ownership, short term incentives in the financial sector, an expansionary monetary and fiscal policies, excessive reliance of debt financing, weak regulatory oversight and global imbalances (Obstfeld and Rogoff 2009).

Financial innovation and lax regulatory oversight allowed banks and other financial institutions to benefit from forms of regulatory arbitrage. This enabled banks to continue to meet regulatory capital requirements while increasing their leverage by using off-balance special purpose vehicles (Keys, 2009). This trend was further encouraged by an abundance of capital which was caused by persistent global payment imbalances. In 2007 China’s foreign currency reserves exceeded one trillion US Dollar. This is the result of devaluating the Chinese currency, the Renminbi, against the US Dollar and other major currencies to promote exports. This excess capital in China also had to find an economic return and played in key role in driving up real estate prices in the United States and other developed economies. Increased capital mobility also resulted in increased financial instability according to Bordo et al. (2001).

(20)

20

An increasing reliance was placed on mathematical risk modelling, particularly the value-at-risk (VAR) method by financial institutions. Some of the models used would only apply two years of historical asset price data to predict future pricing patterns. There is evidence that some of the risk models used by financial institutions and rating agencies could not be adjusted for a decreasing price (Lewis, 2011). The financial sector would often enjoy very short-term incentives such as sales commission and bonusses which would often encourage excessive risk taking. The combination of these factors put in motion a positive loop of rising asset prices and credit creation. Homeowners could swap the positive equity for cash by refinancing their homes which in turn creates a positive cycle of economic activity and asset prices (e.g. Gorton, 2008; Brunnermeier, 2009)

The Federal Reserve Bank of the United States lowered interest rates in 2001 following the dot-com crash and the September eleven terrorist attacks. It would take until 2005 before the Federal Reserve would start to rise the interest rates again to a substantial level. Taylor (2010) created a counterfactual scenario of what the monetary and fiscal policies should have been to prevent real estate prices from overheating. In this counterfactual scenario, the Federal Reserve Bank raised the interest much quicker after the dot-com bubble. Many subprime mortgage products charged a variable interest rate to borrowers. These products often had a ‘teaser’ rate in the first two years of the contract and would than switch over to substantial interest rate payments. These payments would soon trigger many mortgage defaults and subsequent foreclosures. In early 2007 the troubled mortgage portfolio was estimated to be around half a billion US Dollar. This estimate proved out to be too optimistic as this was soon revised upwards to two trillion US Dollar.

Two trillion US Dollar of distressed assets in the United States would likely to cause a recession but should not cause a meltdown of the financial systems. Why did these subprime loans had such a detrimental effect on the financial sector? When in 2001 the dot-com bubble popped, the composite share index of technology companies, the NASDAQ lost nearly four fifths of value in the 30 months that followed. Trillions of Dollar of share value vaporised. The financial sector suffered substantial losses but was not near a meltdown. The effect of the subprime crisis was so severe as many subprime assets were packaged into Collateral Default Obligation derivatives. These assets were often rated as triple-A by the credit rating agencies. This high credit rating required very little if not no capital requirements for financial institutions holding these assets. When losses occurred, it forced the financial institutions to significantly scale back lending to meet capital

requirements (Roubini and Mihm, 2010).

The financial sector was highly reliant of short term debt financing from money markets. Debt plays a key role as this is the main source of funding to provide liquidity and largely functions on trust without having to spend too much time and effort on the due diligence process (Holmstrom, 2015). This is a very effective and cheap method of financing but has two very substantial downsides: first, money markets can very quickly grind to halt in times of financial distress. Second, debt contracts are very often very rigid. Non-payment often triggers a loan covenant, results in restructuring or default. The British Bank Northern Rock was heavily reliant on money markets instead of traditional funding through retail deposits. The large broker-dealers in the United States such as Lehman Brothers and Bear Stearns had approximately three percent equity. When

(21)

21

money markets dried up Northern Rock and Lehman Brothers could no longer fund themselves and became insolvent.

The most significant lesson for the financial sector and policy makers is to learn from history. Sadly, the 2008 financial was not significantly different than previous financial crises in history. Reinhart and Rogoff (2011) provide a concise overview of eight centuries of financial history in their landmark book titled; this time is different. The essence of the book is that financial crises are not that different at all; confidence is lost following the building up of too much debt in either the private or public sector. The financial instability theory from Minsky (1992) provides an explanation. The financial sector demands ever higher profits and therefore loans out money which create an asset bubble. Once the debt bubble has burst it often results in a recession or worse, depression. Eichengreen (2015) draws a historical comparison between in the 1929 and 2008 financial crisis; a rapid appreciation of asset prices preceded both crises.

Nearly a decade after the financial crisis much has happened yet little has changed. One thing is for sure is that the financial sector has become subject to much more regulatory scrutiny since the financial crisis. Yet many of the other factors which lead the crisis remain largely un-addressed. Since the crisis leverage has increased from 142 trillion US Dollar in 2007 to 199 trillion US Dollar at the end of 2014 (McKinsey Global Institute, 2015). Most of the increasing debt is caused by sovereign debt as governments absorbed non-performing assets on their own balance sheet. Another key issue which is very unlikely to be resolved in the foreseeable future is that of global imbalances. Walter (2009) argues that each country will optimise its own policy whether to have a surplus a deficit. National politics cannot be abolished and therefore global imbalances will remain. The financial sector remains highly leveraged. Turner (2015) argues that capital requirements should be several multiples of what these are today. Eichengreen (2015) argues that policy makers decisions to counteract worst financial crisis in 80 years will lead to another one in less than 80 years. For that reason, we need to increase our understanding of financial sector rescue plan and their subsequent distribution of the cost. The next section will discuss the difficulties policy makers face when having to rescue the financial sector.

Solving a Banking Crisis

To prevent is better than to cure. Preventing a financial crisis is essential for a market economy to properly function. The 2008 financial crisis has been the most severe financial crash in 80 years and triggered a global recession in 2009. Governments were taken by surprise by the financial crash and had the act in order to keep their national financial sectors afloat. The required public expenditures were not modest: in the first three years of the crisis the European Union spent 1.6 trillion Euro (13 percent of GDP) while the United States spend 837 billion Dollar (5.5% of GDP) (Stolz and Wedow, 2010). The difference per individual country is significant. For example, Italy and Poland needed to do very little except to provide guarantees which were hardly

(22)

22

required by the financial sector. In the United States, United Kingdom and Ireland the governments had to put in force various rescue packages and other forms of economic stimulus and bank nationalisations.

It is estimated that the 2008 financial crisis resulted in a loss of output of 25 percent of GDP and a median increase in public debt of 24 percent of GDP in the three years which followed the crisis (Laeven and Valencia, 2010). Public debt-to-GDP will increase significantly as the denominator gets smaller. The aftermath of the 2008 financial crisis has been modest when compared to 1929. The 1929 financial crisis had more severe effects; it would take to 1942 before the output value in the United States exceeded that of 1929 (Galbraith, 1954). One could even argue that in absence of the wartime production demand this could have taken even longer. The beggar-thy-neighbour policies which followed the 1929 crisis only made things worse. Banks which became insolvent in 1929 and following years were allowed to fail which in turn caused

depositors to lose their money. This also severely impaired credit creation. Some lessons had been learned from the 1930s and politicians and policy makers pulled every possible lever to keep the financial system afloat; recession which are caused by a financial crisis are more severe and tend to last longer than non-financial crisis recessions (Jordà et al, 2011).

Policy makers were plagued by a severe limitation to be able to effectively cope with the financial crisis and that was insufficient legal powers. The insufficient legal powers were often compounded with limited institutional mechanisms to address the problem encountered. Many studies and government evaluation reports have been written about this shortcomings and areas for future improvement. (i.e. Treasury Committee 2009; Financial Crisis Inquiry Commission 2011; McLean and Nocera 2011; Bernanke 2013). The limited legal power prevented policy makers from acting effectively. The evaluation from the British Treasury committee and financial inquiry commission both mentioned the lack of legal powers to force financial institutions to accept government provided capital. The Chairman of the US Federal Reserve Bank Bernanke was troubled by the fact that the US government could only save Lehman Brothers with congressional approval. Time had simply run out.

Once it is evident that a banking crisis is inevitable policy makers have various choices to act. Landier and Ueda (2009) provide an overview of the various methods to rescue an ailing bank. The authors conclude that there is not a one-size fits-all approach to rescue a bank; it all depends on the circumstances at the time. Landier and Ueda (2009) consider it to be best option to have a voluntarily debt restructuring. This can be realised by converting debt into equity, as this reliefs the balance sheet. The authors clearly acknowledge that in practice this is very difficult to achieve. The theory of Modigliani-Miller (1958) plays an important role. According to this theory the value of a firm is not affected by its capital structure regardless whether the firm is financed solely of debt or equity. In practice, this rather different due to three reasons, first, the interest expense paid on debt is a tax allowable expense while a dividend paid on equity is not. Second, the

management of a financial institution and existing shareholders loathe the option of having to dilute equity as any earned profits will have to be distributed over a larger pool. Third, issuing any new equity requires the permission of existing shareholders. Most likely this will have to done by calling a general shareholders meeting. When in financial distress there is simply no time for this. During the 2008 financial crisis

(23)

23

bondholders were in a very good position. There were only two notable exceptions; bondholders of Lehman Brothers were wiped out and haircuts for Washington Mutual’ bondholders were imposed by the Federal Deposit Insurance Committee (Bernanke 2015).

All other options discussed by Landier and Ueda (2009) require a government guarantee, capital injection or asset swap; this implies that taxpayer’s money is at risk. When taxpayer’s money is at risk it adds a dimension to the problem: moral hazard. When a government steps in to pick up the broken pieces every time a financial institution runs into trouble it is likely to result excessive risk taking in the financial sector. Farhi and Tirole (2009) argue that the issue of moral hazard is particularly difficult to address when banks have access to the same kind of government provided resources within the same timeframe. Haldane and Scheibe (2004) also provide evidence that bailouts can lead to moral hazard. Haldane and Scheibe (2004) argue that the cost of moral hazard are invisible and long-lasting and therefore difficult to analyse. The authors overcome this limitation by analysing the market capitalisation of banks which are based in countries which received a bailout from the International Monetary Fund. The market capitalisation of these banks significantly increased after being bailed out and therefore indicate moral hazard.

It is not easy to answer a relatively straightforward question as; how would you define success in rescuing a bank. The issue is that policy makers face a trade off when saving a deposit taking financial institution:

- To create financial stability - To protect depositors

- To maintain new credit creation - To create a positive economic return

New credit creation is essential for any market economy to function. Policy makers could be faced having to trade-of some financial stability against new credit creation. Another difficulty to overcome by policy makers is to prevent financial institutions which have received bailout capital to only use this to strengthen their own balance sheet without lending any money. Financial institutions develop an appetite for liquid assets, such as high quality corporate bonds and government bonds during times of financial distress (Ivishina and

Scharfstein, 2010). Bailed-out banks can be pressured by politicians and policy makers to lend money to boost economic activity. An effective method to get banks to lend money has been invented in Japan in the 1990s: quantitative easing. Central banks buy high quality corporate bonds and government debt. This creates a distortion in financial markets as the increased demand for these types of financial assets decreases the yield. The result is that banks earn very little or no return on these assets and are therefore forced to lend out money to make a profit (Fawley and Neely, 2013)

If debt cannot be converted in to equity on a voluntarily basis, or if this option is simply insufficient to restore confidence in the ailing bank, it will require the support from the government. Panetta et al. (2009) performed an assessment study of the various financial sector rescue programmes across the world. There are various conclusions from this study. Credit creation is impaired after a bailout. The announcement of a bank

(24)

24

bailout programme reduced the Credit Default Spread; this is one of the main risk indicators of possible default. Bank share prices respond negatively to bailouts; this is a logical response as any company which requires a bailout is unlikely to produce earnings which exceed earnings prior the bailout.

Hryckiewicz (2014) performed an analysis of 23 bank bailouts across 23 countries between 1994 and 2002. There is no one-size-fits all approach according to Hryckiewicz (2014) when it comes to providing effective bailouts; this all depends on the individual circumstances. Hryckiewicz completed the research in 2014 and excluded the bank bailouts of 2008 as these could not yet be properly evaluated. Some bank bailouts take more than twenty years before they are fully resolved. Bhattacharya and Nyborg (2012) discuss the issues policy makers face when deciding between two bailout options: equity injections or creating an asset

management company, or a ‘bad bank’. The creation of an asset management company is a very effective method to remove non-performing assets from the balance sheet of bank. This action has two downsides. First, this option is most likely to benefit existing shareholders and leaved the problem with the taxpayer. Second, the management of the bailout bank has better information about the quality of the assets than policy makers have. This creates the ‘lemons’ problem (Akerlof, 1970). This can be overcome be running an effective auction programme (e.g. Lawrence and Cramton (2008) and Bebchuk (2009)). An equity injection dilutes the ownership of existing shareholders and therefore involves less moral hazard. The disadvantage of this approach is that it limits the incentive for the bank to issue new credit. Policy makers might have a preference to create an asset management company for a different reason. Gandrud and Hallerberg (2014) argue that the rules laid out by the European Statistics Agency play a key role. Twelve EU member states created a ‘bad bank’ even though this benefits the shareholders of a bank. The reason is that when a European member state creates an asset management company which is at least 51% privately owned it is considered as a temporarily institution. temporarily institution do not have to be integrated in the government’s budget and will therefore not affect the budget deficit or debt-to-GDP ratio. Another key aspect was an amendment in the application of international accounting standards. The International Accounting Standards Board changed the application of IAS39, the recognition and measurement of financial instruments. Financial instruments should be valued at market prices, meaning what they are sold for at that moment in time. The International Accounting Standards Board made an adjustment for the measurement of value of assets for which there is no active market. Financial institutions were enabled to value financial assets at their historical cost. This rule came into force on the 15th_{of October 2008 and had very significant effect on financial institutions as they no longer had to write}

down the value of assets if they did not sell them.

Significantly more research has been conducted on the financial sector rescue programmes in the United States compared to Europe. One possible explanation could be relative transparency of the bailout programme of the United States compared to Europe. When Treasury Secretary, Tim Geithner requested the required funds for the Troubled Asset Relieve Programme (TARP) he received various counter demands from the US Congress. One of those demands was a full transparency of which financial institutions received bailout funds but also which funds were paid out by these bailout financial institutions. In Europe, vast amounts of funds were made available to financial institutions which had to be bailed out but effective nothing is known

(25)

25

who the beneficiaries were. This subject is somewhat painfully captured in a working paper from the Bank of England (Rose and Wieladek, 2012). The authors used confidential data which is only available within the Bank of England to evaluate the bailout programme. As the authors were bound by the issue of confidentiality it cannot offer any significant conclusion except the positive correlation between a bank size and bailout funds received.

Veronesi and Zingales (2010) provide a very helpful overview of the effect of US bailout programme with their paper titled: “Paulson’s gift”, named after the Treasury Secretary Henry Paulson. According to the authors the US financial conglomerate Citigroup was one of the largest beneficiaries of the bailout while JPMorgan gained the least. The authors clearly acknowledge the issue of how to account for success. JPMorgan largely avoided the subprime crisis due to prudence and in return acquired rivals Bear Stearns and Washington Mutual. This had an initial negative effect on the share price of JPMorgan but is gained a very significant market share at a bargain price. JPMorgan’s reputation was also boosted although one cannot account for this in a monetary value. Citigroup on other hand gained financially from the bailout but had its reputation severely impaired and market share reduced. The US financial regulators initially imposed punitive bailout terms on financial institutions in need of capital. This was soon changed as policy makers realised that in effect the financial sector cannot be punished without impairing credit creation. If financial institutions pay a high interest rate on bailout capital provided by the government than they pass on these costs when lending money.

This section has addressed the issues policy makers face when implementing a financial sector rescue plan. Policy makers did not have the legal powers to intervene and face difficult trade-offs when providing bailout capital to the financial sector. This section has not addressed the distribution of the cost of a financial sector rescue plan. This is addressed is section three.

Bailing Out: Distribution of the Cost

When in middle of 2008 it became clear that a full blown financial crisis was coming both the financial sector and governments were taken by surprise. With a few exceptions most ailing financial institutions were bailed out at the cost of the taxpayer. How did the financial sector managed to get away with this? This section will discuss the key literature on the distribution of the cost of a financial sector rescue plan. This section will argue that the distribution of the cost of a financial sector rescue plan are determined by the structural power of the financial sector and the norms and routines of the financial sector and the government to work together.

Finance played a key role in the development of the modern states. Stasavage (2011) argues that Italian and German city states became powerful as they managed to borrow from their citizens. Citizens would lend money to the government of the city state and in turn would have a voice on how the money was spend. This started in the twelfth century in Italy and Germany. The result was that these city states could raise

(26)

26

substantially larger armies than traditional kingdoms as they had more capital. City states remained relatively small, maximum a 50-kilometer radius, which was the maximum area which could be administered at the time. This concept is the start of political representation as we know it today. The Netherlands was the first nation state which borrowed from its citizens during the sixteenth century. At the time, the governing bodies of the large cities were run by merchant oligarchies. These wealthy merchants often invested heavily in public credit. It would take to the eighteenth century when a banker was considered to be more powerful than

governments: Nathan Mayer Rothschild. His position was summarised in 1828 by the British Member of Parliament Thomas Dunscombe as “Master of unbounded wealth, he boasts that he is the arbiter of peace and war, and that the credit of nations depends upon his nod” (Ferguson, 1998). Napoleon Bonaparte clearly mentions the perils of being dependent on banks; “When the money of a government depends on banks, they, not the head of state, control the situation” (Johnson and Kwak, 2010). The head of political campaign of US President Bill Clinton, James Carville (The Economist, 2016), famously stated that he would like to reincarnate as the bond market as this would enable him to intimidate everyone.

Perhaps the most famous guide on how to obtain absolute power is the Prince, written by

Machiavelli. The Prince clearly mentions the power of finance as at the time of its publication in 1525. At the time capital markets had not been developed but wealthy families operated their own banks. Machiavelli states a very important boundary condition for any good prince to avoid grief; a good Prince cannot be dependent on the fortune and goodwill of others (Machiavelli, 1525). Strange (1988) argues that the management of money is too important to be left to the bankers. The Financial sector is in a rather unique position when it comes to requesting a bailout from the government. Research from Smith (2012) suggests that the financial sector is by far the most likely sector to be bailed out compared to other industries, regardless of whether the government at the time is right or left wing. One key factors which stands out in whether an industry gets thrown or lifeline from the government or not is possible job losses. This puts the financial sector in a very powerful position; if it does not issue new credit then production capacity will be impaired, this in turn forces firms to lay off workers. The alternative of not bailing an ailing financial institution would most likely result in a default.

The financial sector rescue plan could be considered as a game of chicken, according to Woll (2014). If one or more major players in the financial sector defect the state is left to two options; provide a bailout at the cost of the taxpayer or allow failure. Both the financial sector and the government pay a grave price if the government and the financial sector defect. There are only two examples in the 2008 financial crisis: the US investment bank Lehman Brothers and the entire Icelandic financial sector. Iceland simply could not rescue its financial sector as it faced a currency crisis. Iceland lost nearly two-thirds of its GDP because of the dual financial and banking crisis. The failure of Lehman Brothers resulted in a painful blow to the interdependent financial sector. At the time of its default it had over 600 billion of issued debt and promissory notes. Current estimates are that between 80 and 100 billion US Dollar will be recovered from the bankruptcy proceedings. Woll (2014) demonstrates that the financial sector rescue plans of United States, United Kingdom, France, Germany and Denmark had a similarity; all countries attempted to coerce their financial sector to act

(27)

27

collectively to share the cost. The United Kingdom and Germany failed to achieve this. All business firms are agents of economic activity in capitalist market economies. The state depends on the investment of business firms to generate economic growth. This puts business firms in a powerful position because if they do not invest they will harm the politicians which are governing (Przeworski and Wallerstein, 1988). Barnet and Duvall (2005) define the power of business firms as productive power and is concerned with constitutive social processes that are not controlled by specific state actors. The state can encourage business firms to invest but it cannot coerce them. Governments are therefore predisposed to adopt policies which promote investment. A negative policy, or even the anticipation of one, from the government towards business firms will therefore result in lower investment. This puts the government in a position of dependence on business firms (Lindblom, 1980). This applies to all business firms but most of all to the financial sector as credit creation is one of the key determents of production.

Culpepper and Reinke (2013) argue that the high structural power of the British and German financial sector prevented them from being coerced by their governments. Culpepper and Reinke (2013) argue that financial institutions can defy their national regulator when they have very significant international exposure. Both the British Bank HSBC and the German Deutsche Bank generated less than one-third of their revenues in their domestic markets. When they defied their national regulator the did not fear a regulatory retaliation. Both banks could credibly threaten to scale down operations in their domestic markets if the climate would become unfavourable. Culpepper and Reinke (2013) argue that the 2008 financial crisis was a unique opportunity to demonstrate the structural power of the financial sector. Unusual events provide the

opportunity to test the empirical implication of rival theories, which are often rather close in practice. A good example comes from Albert Einstein. Einstein’s theory predicted that astronomers would be able to observe distant stars located behind the sun, because the sun’s gravity would bend the light around the sun. This could only be observed during a solar eclipse and this proved the validity of Einstein’s theory. (Culpepper and Reinke 2013:13).

Concluding Remarks

The literature review has argued that many of the causes of the 2008 financial crisis remain unaddressed. For this reason, it important to study financial sector rescue plans as we will experience another financial crisis in the future. It is difficult to define what exactly success is when implementing a financial sector rescue plan. Policy makers face trade-offs between financial stability and credit creation. Modern market economies cannot function without sufficient credit creation. This puts the financial sector in a very powerful position. Past research demonstrates that the overall cost of a financial sector rescue plan was lower when the financial sector and government coordinated their response to the financial crisis. Some countries failed in coercing their financial sector to act collectively. As a result, most of the cost of the financial sector rescue plan went to the public budget.

What shapes the distribution of the cost of a financial sector rescue plan? : policy choices of the Dutch government to rescue the financial sector in 2008