EU membership, financial services and stability.

Armstrong, Angus

This paper examines whether EU membership enhances or diminishes the UK's financial sector stability, and therefore its prominence in global finance. The UK is host to the largest share of financial services in the EU, despite being outside of the Eurozone. An important reason is that, as a member of the EU, the UK has direct access to the Eurozone's financial infrastructure. If the UK leaves the EU (and EEA) banks and other financial services firms may continue to have access to the Single Market, but they are unlikely to have direct access to the Eurozone's infrastructure. Banks in the UK will no longer be direct members the Eurozone's payments system. The swap arrangement between the European Central Bank and Bank of England would have no legal enforcement mechanism. Resolution of cross-border banks would be more challenging with less incentive for a cooperative outcome. While some may welcome the reduced size of the financial system, not without reason, this could be achieved more effectively with domestic regulation than by leaving the EU. Given the uncertainty that would follow a vote to leave, there is a risk of capital flight.

Keywords: governance: financial infrastructure

JEL Classifications: E52; E58; F33; F36; G00; G001

"Respecting the powers of the central bank in the performance of their tasks, including the provision of central bank liquidity within their respective jurisdictions"

From the Agreement of the European Council Meeting, 18 and 19 February, 2016

The UK joined the European Economic Community in 1973, the same year as the collapse of Bretton Woods. This was a system of closed capital accounts and fixed exchange rates designed to overcome the financial instability of the interwar years.

The demise of Bretton Woods led to a gradual opening up of capital accounts. The UK led the way in 1979 by abolishing exchange controls. In the same year, other members of the EEC created the European Exchange Rate Mechanism of pegged exchange rates from the remains of the 'Snake in the Tunnel' system. Despite the movement of capital being a fundamental freedom in the Treaty of the Functioning of the European Union (TFEU, 1958) progress was slow due to excess capacity in many countries and the new pegged exchange rate regime. The rise of cross-border capital flows coincided with a revolution in IT that transformed financial intermediation. Batch processing allowed bank assets, long considered unsuitable for transfer, to be bundled up and traded. International financial borders were torn down, enabling financial institutions to merge into conglomerates operating around the globe. The regulatory framework was designed in such a way as to support financial sector growth. (1)

Continental Europe began to catch up. EU leaders either recognised the need for more efficient financial markets, or accepted that they could no longer defend domestic financial behemoths. Europe's largest banks made strategic decisions to become global players. (2) The EU Financial Services Action Plan (1998) accelerated the creation of a single wholesale market allowing regulated financial firms of any Member State to operate in the markets of any other Member State without any further restrictions (so-called 'passporting'). (3)

With repeated financial crises culminating in the global financial crisis, it is clear that global finance is a challenge to traditional governance structures. (4) Lord King's famous remark that banks are "global in life and national in death" indicates a profound misalignment between incentives and responsibility. The lack of congruence in traditional governance structures is expressed in many ways e.g., Rodrik (2000), Schoenmaker (2013) and Avdjiev et al. (2015).

The common thread is an inconsistency between internationalism and the policy domain of nation-states.

In the financial context, this inconsistency is between the global financial institutions operating in multiple jurisdictions and with multiple currencies, and the domestic governments which stand behind them with their taxpayers' funds and domestic currencies. Our exam question is therefore whether EU membership, and access to the Eurozone's financial infrastructure, makes this global financial governance problem more or less congruent with our nation state.

UK financial services

The UK, and the City of London in particular, has been a leading financial centre for two centuries. Finance is one of the UK's most important areas of economic activity. The Office of National Statistics estimates that financial sector output is 8 per cent of national output, although others suggest if relevant business services are included this would be 12 per cent of output. (5) The UK is by far the largest net exporter of financial services in the world with a trade surplus of over 3 per cent of GDP. (6) HM Government (2014) reports that financial services contributed 12 per cent to PAYE Income Tax and 15 per cent to corporation tax receipts in 2012-13. The distinguishing features of the UK finance system are that it is: (a) large; (b) international, a conduit for other EU nations; (c) multi-currency; and, (d) in large part funded by foreign direct investment.

The first defining feature of UK finance is that the sector is big with a wide variety of services provided. Table 1 shows the size of the financial services on a residency basis (i.e. operating within the national border) relative to national output. (7) Derivatives are excluded. Banking is roughly split between domestic and foreign-owned institutions. On a national basis, UK banking sector assets, including overseas operations, are approximately 450 per cent of GDP. Stability of domestic institutions matters for the national public sector balance sheet. This is an important distinction between the UK and Luxembourg that has a large but almost entirely foreign owned banking system. (8) The diversity of financial services indicates that not all specialities will be equally influenced by EU membership.

[FIGURE 1 OMITTED]

The second defining feature of UK finance is the large presence of overseas firms. According to the Bank of England (2015), around half of the world's largest financial firms, commercial and investment banks, insurers, asset managers and hedge funds have their European headquarters in the UK. The City has the largest share of EU investment banking, wholesale finance, insurance, asset management and hedge funds despite being outside of the Eurozone. (9) Having 'passporting' rights to the Single Market and access to the financial infrastructure of the Eurozone is vital to hosting Euro financial markets.

With so many international firms, the UK banking system is a multi-currency area. This matters for the governance issues raised in section 1. Panel (a) in figure 1 shows that less than half of UK banking system assets are denominated in sterling. (10) Around 28 per cent of assets are denominated in dollars and 20 per cent in euros. Yet panel (b) shows that a greater proportion of loans and advances to UK residents are from European banks compared to US and banks from other nations. This suggests that the UK is a conduit for non-EEA firms (especially from the US) carrying out banking in the rest of the EU via the 'passporting' arrangement. (11)

The fourth defining feature is the amount of foreign direct investment (FDI) related to the financial sector. Delivering financial services usually requires a physical presence or ownership of a local company. This implies a large amount of FDI in financial services. The UK is the second largest hub (inward and outward) of FDI in the world and by far the largest in the EU. Of the 937bn [pounds sterling] stock of FDI in the UK in 2012, 43 per cent is from the EU and 40 per cent of the total is related to financial services. (12) According to Ernst & Young, 72 per cent of foreign investors consider access to the Single Market as most important to the UK's attractiveness as a destination for FDI. (13) FDI is traditionally seen as a committed form of foreign capital. An important question is whether this is still true today.

EU membership and governance

International financial regulation is guided by G20, the Financial Stability Board and Global Standard Setting Bodies. This is an appropriate governance structure, given the global nature of finance. Representatives from the UK are influential voices in these institutions. The implementation of the guidelines to the Single Market is the responsibility of the European Commission (EC). Before the crisis the EC issued directives about opening markets, setting minimum regulatory standards, mutual recognition and consumer protection. After the crisis the EC has relied increasingly on regulations which apply directly to Member States and provisions for maximum harmonisation of regulations across the EU. Some regulations are seen as intrusive and inappropriate. (14)

The decision to establish a European Banking Union (EBU) will have profound implications for the whole of the EU. The scope of the EBU is well beyond anything which has been achieved between sovereign states in the past. The UK was one of its strongest proponents as a means of resolving the underlying tensions of the Eurozone crisis. However, there is a clear concern that to deliver the EBU involves ever greater integration. Financial regulation may be cast in the interests of members of the EBU and not be in the interests of Member States outside of the EU.

Since the UK has such a large financial sector, this is important to our economic sovereignty. Moreover, because the Eurozone countries have a qualified majority at the Council of Ministers they can, in theory, caucus or act as a block or pass legislation in the group's interest. (15) Prime Minister Cameron's renegotiation goes some way to allay these fears. First, it reinforces the principle of non-discrimination on the basis of currency. The UK has challenged three areas of financial regulation before the European Court of Justice (ECJ) and has had some important successes. (16) Second, the UK is permitted to be a non-voting member of the Eurogroup and raise matters of serious concern to the Council. However, there is no full answer to the threat of caucusing by Eurozone nations. Further concessions to protect British interests would come to be more like a veto as more Member States join the EBU. (17)

Regulatory sovereignty

The important issue for the referendum is whether the UK would gain more regulatory control by leaving the EU. Assuming that the UK intends to continue as the hub of Eurozone finance, pursuing a separate regulatory agenda is unlikely to be permitted and would undermine financial stability. First, if the UK is to have access to the Single Market for financial services it must be accepted as 'equivalent' in regulation terms. This judgement is in the gift of the EU. Second, the UK would have no direct say in future EU regulations. Third, there is limited scope for regulatory discretion for any international financial centre. Shoenmaker (2013) shows that if institutions operate across borders, regulatory decisions at a national level tend to have spillovers to other nations creating potential instability. (18)

Loss of access to the Single Market for financial services would be problematic. If the UK leaves the EU and does not join the EEA (an inferior outcome for the EU and the UK) then from a regulatory perspective the UK would be a 'third country'. The UK could be granted access to EU markets if its regulatory regime is considered equivalent. Firms which sell to retail markets in Member States would be required to establish an entity in each market. Firms which sell to professional investors may be permitted if they are equivalent and registered with the European Securities and Markets Authority. Banks must also meet equivalence criteria of the Capital Requirements Directive. Whether this proves to be a stable equilibrium over time depends on whether the EU would seek 'line by line' equivalence. Loss of access to the ECJ would make it difficult to appeal any contentious ruling.

Because of the regulatory 'sword of Damocles' over market access, non-EU financial institutions located in the UK are likely to establish a subsidiary in a Member State of the rest of the EU. They could then set up branches from that new subsidiary. As well adding to, rather than reducing, their regulatory burden, the risk is that the UK subsidiary becomes less important. EU banks operating in the UK may also seek to establish a subsidiary if the UK leaves the EU. Because subsidiaries in the UK are individually capitalised and do not allow consolidated internal cash management they are a more expensive form of banking. And the UK would have gained little, if any, regulatory sovereignty.

EU membership and financial infrastructure

A necessary ingredient for stability is to make sure that the financial infrastructure operates smoothly, in particular, that the payments and settlements systems function efficiently at all times so that non-cash transactions are settled promptly and with full confidence. It is the threat of disruption to the financial infrastructure which often forces a government's hand to intervene to prop up failing firms.

Payments system

The payments system for the Eurozone is TARGET2. (19) The Eurosystem operates as a 'hub-and-spoke' structure with the ECB as the hub and Eurozone national central banks (NCBs) as the spokes. (20) Commercial banks facilitate cross-border payments on behalf of clients by registering debits and credits at their respective NCB. At the end of each day the net amount is reported to the ECB and the NCB's reserve account is adjusted accordingly.

A unique feature of TARGET2 is that participants are not always nations in the Eurosystem. NCBs, banks and designated financial institutions within the EEA are invited to join TARGET2 even if they are not in the Eurozone. However, the risk to the Eurosystem is tightly controlled as intraday day credit is usually limited for each counterparty and there is no access to overnight credit for NCBs outside the Eurosystem. The Bank of England (and NCBs of all nine non-Eurozone countries) is a shareholder of the ECB but has declined to be a member of TARGET2. (21)

Given the amount of euro-denominated finance carried out in the UK, which must be settled and cleared each day, access to a reliable payment services is critical. UK banks and other designated financial institutions are permitted to be direct participants in TARGET2 even though the Bank of England does not participate. Where there are imbalances, they often use the NCB of a Eurozone Member State. For example, Lloyds Bank London is reported to hold its euro reserves account with the Dutch Central Bank. (22) Any use of central bank finance is provided against collateral delivered by the Lloyds local office.

How would this change if the UK were to leave the EU? If the UK were to join the EEA there is no technical difference in access to TARGET2. If the UK were to leave the EU and not join the EEA then banks in the UK could no longer be direct members of TARGET2. They would have to operate through subsidiaries (or perhaps branches assuming the UK is deemed 'equivalent' in terms of regulation) within the EEA. This would make euro banking via the UK more expensive. It would also erode the attraction of the UK as a destination for non-EU institutions to establish their EU headquarters. It would also be unusual to extend direct access to institutions in countries outside the legal jurisdiction of the EU.

Central counterparties

Another important part of financial infrastructure is central counterparties (CCPs). In response to the global financial crisis, G20 committed to standardising derivatives and moving trading from opaque over-the-counter bilateral transactions to CCPs. An example is shown in figure 2. The idea is that the CCP is 'buyer to every seller and seller to every buyer': they stand between all counterparties. This creates greater transparency, security and enhances risk management. The figure shows how complex webs of bilateral transactions can be greatly simplified by a CCP. Each counterparty is required to post collateral at the CCP. In 2015 the value of pre-placed margin and default funds for the four CCPs supervised by the Bank of England was 91 bn [pounds sterling]. (23) While CCPs may seem a highly technical part of the underpinning infrastructure of the financial system, there is no doubt about the importance they play to London. The UK is the global centre of derivatives and securities trading. These activities complement international banking activities. Over half of interest rate derivatives (the largest asset class) and one third of credit derivative are traded on CCPs. However, the more that trade is concentrated on CCPs the more they become systemically important financial institutions. CCPs in the UK have access to the Sterling Monetary Framework including Discount Window Facilities. In the Eurozone CCPs have access to TARGET2.

[FIGURE 2 OMITTED]

A critical issue known as 'Location Policy' emerged as a result of the ECB's Eurosystem Oversight Policy Framework. According to the ECB Article 127 (2) of TFEU, one of the functions of the Eurosystem is to "promote the smooth operation of payment systems" and the means by which this is assigned is specified in Article 22 of the Statute of the European System of Central Banks and the ECB: "the ECB and NCBs may provide facilities, and the ECB may make regulations, to ensure efficient and sound clearing and payment systems within the Union and with other countries". (24) As it is acknowledged that CCPs can be a source of systemic risk, as focal points for credit and liquidity risk, an important question is where they should be geographically located.

The quotation at the start of this paper suggests that if CCPs may require euro liquidity, which is exclusively in the gift of the ECB, they should be located in the jurisdiction of the Eurosystem. Indeed, the ECB stated that CCPs with more than 5bn [euro] daily net credit exposure (in euros) per main product would be located within the Eurosystem jurisdiction. This is consistent with aligning the functions of the state, central bank and currency discussed in section 1. However, the UK requested a judicial review by the European Court of Justice (Case T-49611). The court annulled the ECB's policy on the basis that the TFEU granted competence for payment systems and not CCPs.

Swaps and incomplete contracts

How this issue was resolved is important for the EU referendum. The Bank of England and ECB agreed to joint oversight of CCPs and reciprocal currency swaps to facilitate multi-currency liquidity support. (25) This extends the existing swap agreements between the Federal Reserve and five major central banks. (26) Note that this is not a pre-commitment and there is no notion of amount. (27) However, Prime Minister Cameron re-affirmed the soundness of this liquidity support by negotiating a 'no discrimination' for Member States which do not use the euro in an important clause in the European Council Agreement in February 2016.

If the UK leaves the EU it would no longer be a shareholder of the ECB, it would not have access to the ECJ and the principle of 'no discrimination' would not come into force. It is difficult to see why the ECB would be interested in renewing an arrangement it had not wanted in the first place. In the words of Former Governor Noyer of the Banque de France, "if Britain left the EU, the Euro Area authorities could no longer tolerate such a high proportion of financial activities involving their currency taking place abroad". (28)

Why should the ECB object to the swap agreement if the UK were to leave the EU? The Fed maintains swap lines with other nations, but they do not have anything like one-third of the US dollar wholesale markets. The size of currency injections could complicate the euro reserves market and obstruct the setting of monetary policy. It is therefore unlikely that a major central bank would permit a swap line to support such a volume of transactions to take place offshore and outside the legal jurisdiction of the EU.

Even if the ECB were to commit to maintaining a swap line, there is no way to commit credibly to the agreement. In a crisis the optimal response by the ECB may be to act in the interests of EU citizens. Outside of the EU the UK would have no capacity to call for a review. Issues about the interpretation and application of the TFEU are outside the jurisdiction of the EFTA Court (the Court of the EEA) and a Memorandum of Understanding has no basis in international law. Large financial institutions which may require access to emergency liquidity will be aware of the legal uncertainty around this ongoing support.

EU membership and resolution

While a large and international financial system can bring economic rewards, it also carries risks if there are failures. Chancellor Osborne describes this as the 'British dilemma', where we enjoy the benefits of the largest global financial centre but the inherent risks are too big for our relatively small tax base. The dilemma is sharper today given our high level of public debt and limited capacity for future support.

Whatever cooperation agreements may exist between governments to cooperate in resolving cross-border institutions, history shows that they act in their taxpayers' interests, often irrespective of the spillover to other countries. For example, even among founding members of the Eurozone, cooperation proved lacking in the Fortis Bank failure, whose losses were finally divided according to national boundaries. (29)

The obvious way to resolve this incongruence is to remove the dependence on national taxpayers. In 2009 the G20 set out a requirement to reform financial regulation, including new resolution regimes which would allow institutions to fail without disrupting key economic functions and without recourse to taxpayers. The UK introduced its Special Resolution Regime (SRR) in 2009, which has been enhanced by transposing the EU-wide Bank Resolution and Recovery Directive (BRRD). (30) The SRR covers UK incorporated banks, building societies, subsidiaries of foreign firms, some investment firms and central counterparties. Branches of foreign firms may be resolved under the UK regime in some circumstances, although the UK government is currently consulting on this issue.

The UK's resolution authority is the Bank of England's Prudential Regulatory Authority (PRA). The PRA has the authority to write-down existing equity or convert ('bailin') existing unsecured debt into equity to re-capitalise an ongoing firm or cover losses in a failed firm. Of course, bail-in implies a loss for the asset holders. Given the size of UK financial institutions, it is important that shareholders and unsecured creditors are geographically spread otherwise the losses would simply be passed within the economy. The key assumption is that contagion from 'bail-in' is less than from 'bail-out'.

If 'bail-in' proves ineffective in a systemic event, or there are residual losses, then the UK taxpayer remains the fiscal backstop. If in the next systemic crisis 'bail-in' is at least partially successful, then this will reduce the losses facing the taxpayer. In a systemic event it would be highly improbable the government balance sheet would be unused. (31) Therefore, the 'British Dilemma' may, in fact, be unresolved.

Resolution and coordination

Eurozone countries and other Member States which choose to join the Banking Union implement the BRRD through the Single Resolution Mechanism (SRM) for significant and cross-border banks. This has two advantages. First, the SRM, while a complex decision process, requires that resolution decisions are taken with regard to all participants and not national lines and has a non-discrimination requirement. This in theory removes the risk of inefficient spillovers from the national resolution of cross-border banks. Second, a pre-financed Single Resolution Fund (SRF) will be created to mutualise any residual losses. This risk-sharing mechanism is designed to reduce transferring from large failed banks to the sovereign. This in principle is an attractive solution to the 'British Dilemma'.

This raises the question of whether the UK should join the Banking Union. Denmark, the only other EU nation along with the UK to have a permanent opt-out of the Eurozone, has signalled its intention to join. This involves transferring responsibility for financial regulation of most large UK institutions from the Bank to the ECB under the Single Supervisory Mechanism. This would entail some loss of discretion over macro and micro-prudential policy. Moreover, the UK would not be able to join the Governing Council which is the highest decision making authority of the ECB. This would be a large transfer of power without the UK having a seat at the decision table.

Resolution of large EU-wide institutions would require close cooperation between the ECB and the Bank of England. Each institution would be representing their own constituents. There may be disagreements about the optimal approach and ultimately loss sharing. However, within the EU it is reasonable to expect the ECB and UK authorities would have a more collegiate relationship. Moreover, the UK would have the right of access to the ECJ to challenge any perceived discriminatory behaviour on the basis of currency. Being inside the EU does not solve the 'British Dilemma' but it may make any loss sharing more equitable.

Capital flight

The question set at the start of this paper is whether EU membership strengthens or diminishes financial stability and therefore enhances the UK's role as a global financial centre. If the UK leaves the EU (and does not join the EEA) it is likely that it will no longer have direct access to the EU's financial infrastructure. This will erode prospects for financial stability and diminish the UK's current status as the global financial centre.

Well functioning financial markets are public goods: the benefits extend well beyond the home nation. In the event of a decision to leave the EU, we expect this would be recognised by UK and rest of the EU negotiators. Rather than reaching for the 'mutually assured destruction' button, negotiators have every incentive to maintain the status quo as far as possible. This means that the UK would transpose and adopt EU regulation with no meaningful gain in 'regulatory sovereignty'. The EU would reciprocate by accepting UK regulations as equivalent and permit full 'passporting' to the Single Market. Whether this is a stable equilibrium is an open question.

The UK would be likely to lose access to the EU's financial infrastructure. This is always the responsibility of the central bank. The quotation at the start of this paper (repeated in Governor Carney's evidence to the Treasury Select Committee) clearly respects the legal authority of central banks. UK incorporated banks would require a branch in the EU, and non-EU banks resident in the UK would require an EU subsidiary. CCPs are also likely to migrate. Any replacement swap agreement would have no legal basis and therefore be of uncertain value. (32) Finally, the UK would no longer be a shareholder or member of the ECB. This may reduce the incentive for cooperation in the event of resolving a substantial cross-border financial institution.

Some financial institutions are likely to move elsewhere in the EU. This is a form of capital flight, although it is difficult to judge as there are no EU cities with a similar expertise and support in financial services. The outcome is likely to be less efficient intermediation and a higher cost of capital. Bank of England Governor Carney has suggested that the referendum poses a risk to domestic financial stability. (33) There are a number of factors which suggest that this is indeed a risk. First, there is uncertainty around the financial arrangements in the event of a vote to leave the EU. Both sides may like to maintain the status quo, but this will be complicated by legal and political uncertainty. Second, the UK has a current account deficit of almost 7 per cent of GDP even with a healthy surplus in financial services trade. Third, the stock of FDI in the UK related to the financial sector is likely to be near [pounds sterling]250bn. (34) Uncertainty over financial services may make funding of the current account deficit more challenging.

NOTES

(1) The original Basle capital regulations were essentially an instrument of industrial policy designed to slow down the growth of Japanese banks.

(2) For example, the attempted takeover by Pirelli, advised by Goldman Sachs, of Continental AG encouraged Deutsche Bank to become a global investment bank.

(3) The Single European Act (1986) was also key to the creation of a Single Market.

(4) Jorda et al. (2011) present a dataset of banking crises for the fourteen wealthiest nations. Between 1950 and 1980 there were two crises and between 1980 and 2008 there were seventeen crises.

(5) TheCityUK (2016).

(6) UNCTAD database and ONS Pink Book.

(7) These measures exclude interbank assets and liabilities so avoids double counting.

(8) This was discussed at an LSE roundtable (Schelkle and Lokdam, 2015). Unlike Luxembourg, the UK has a large domestic banking system. As long as the government may act as a backstop, the state must supervise and be the lender of last resort. Transferring oversight and decision making for such a large contingent liability would be a considerable diminution of sovereign power.

(9) See TheCityUK (2016).

(10) MFIs are banks and non-bank financial institutions excluding the central bank.

(11) In 2014 there were 348 banking companies in the UK; 248 are incorporated overseas, of which 170 are from outside of the European Economic Area (EEA).

(12) OECD.stat FDI series https://stats.oecd.org.

(13) Reported in Bank of England (2015).

(14) For example, the Bank of England willl not extend the application of the EU's bonus cap to smaller companies, and the Alternative Investment Fund Managers Directive (AIFMD) is widely thought to introduce unnecessary regulations on hedge funds.

(15) The European Financial Stability Mechanism (which the UK contributes to) was used to make an urgent loan to Greece despite an earlier decision not to use the fund to support Eurozone countries.

(16) These are over the location of CCPs (upheld), bankers bonuses (rejected) and a financial transaction tax (postponed).

(17) The UK has a 'double-lock' on voting at the European Banking Authority. This requires a majority of both Eurozone and non-Eurozone members to agree a decision.

(18) An exception is tighter rules on domestic markets such as property.

(19) Second generation of Trans-European Automated Real-time Gross settlement Express Transfer system.

(20) The Eurosystem is both the ECB and Eurozone NCBs.

(21) While the Bank has the third largest shareholding of the ECB in notional terms, the paid-up capital is much smaller and covers only operating costs.

(22) See Jones (2014).

(23) Bank of England (2016), Supervision of Financial Market Infrastructure Report. The four CCPs supervised by the Bank are CH Clearnet Ltd, ICE Clear Europe, LME Clear and CME Clearing Europe.

(24) Quotes from the TFEU and Eurosystem Statutes are from ECB (2011).

(25) Swap agreements are necessary for long dated securities, otherwise central banks could use the spot market.

(26) The big-6 of the international monetary system are the Federal Reserve, Bank of Japan, Bank of England, European Central Bank, Bank of Canada and Swiss National Bank.

(27) Pre-commitment would violate sovereign control of a central bank.

(28) See Noyer (2016).

(29) See for example Schoenmaker (2013). Another example of closure of a globally integrated bank having cross-border consequences is Lehman Brothers. Immediately before filing for bankruptcy its global cash management system withdrew cash from its systemically important branch in London which then had to close on the same day.

(30) EU Directives set out the end result that each Member State must adhere to, but the national authority has discretion over how this is achieved.

(31) In the last crisis the largest direct intervention was the Treasury's Asset Protection Scheme negotiated in February 2009 and the nadir of the crisis. This 'insured' UK banks against tail-risk on certain hard to price assets. The initial insurance cover was around [pounds sterling]250bn of market value assets.

(32) The Bank of England famously had a swap agreement with the Bundesbank which failed so spectacularly in 1992.

(33) Evidence to Treasury Select Committee 8th March, 2016.

(34) The stock of EU FDI in the UK is [pounds sterling]469bn and 30 per cent of recent flow is in the financial sector.

REFERENCES

Avdjiev, S., McCauley, R. and Shin, H. (2015), 'Breaking free of the triple coincidence in international finance', BIS Working Papers No 524.

Bank of England (2015), EU Membership and the Bank of England, http://www.bankofengland.co.uk/publications/Documents/ speeches/2015/euboe211015.pdf.

--Resolution Framework, http://www.bankofengland.co.uk/ financialstability/Pages/role/risk_reduction/srr/framework.aspx.

Burrows, O. and Low, K. (2015), 'Mapping the UK financial system'. Bank of England Quarterly Bulletin, Q2.

European Central Bank (2011), Eurosysem Oversight Policy Framework, http://www.ecb.europa.eu/pub/pdf/other/ eurosystemoversightpolicyframework2011 en.pdf.

HM Government (2014), Review of the Balance of Competences between the United Kingdom and the European Union: The Single Market: Financial Services and the Free Movement of Capital.

Jones, R. (2014), 'Who said payment systems are simple?', European Banking Federation, http://enews.ebf-fbe.eu/2014/01 /who-saidpayment-systems-are-simple-legal-issues-arising-from-the- structure-of-target-2/.

Jorda, O., Schularick, M. and Taylor, A. (2011), 'Financial crises, credit booms, and external imbalances: 140 years of lessons', IMF Economic Review, 59, 2, pp. 340-78.

Noyer, C. (2016), Benefits of Britain's Membership, OMFIF, The Bulletin, http://www.omfif.org/analysis/commentary/2016/march/ benefits-of-britains-membership/.

Rodrik, D (2000), 'How far will international economic integration go?', Journal of Economic Perspectives, 14(1), pp. 177-86.

Schelkle, W. and Lokdam, H. (Rapporteurs) (2015), 'Financial regulation and the protection of eurozone outs', London School of Economics.

Schoenmaker, D. (201 3), Governance of International Banking, Oxford: Oxford University Press.

TheCityUK (2016), A Practioner's guide to Brexit, www.thecityu k.com.

Angus Armstrong, National Institute of Economic and Social Research and Centre for Macroeconomics. E-mail: a.armstrong@niesr.ac.uk.

Table 1. UK financial sector by residency (2014)

                                      [pounds          per
                                    sterling] bn   cent of GDP

UK banks                                3,631          200
Foreign banks                           3,374          186
Finance companies and SPVs                480           26
Pension funds                           1,430           79
Insurance                               1,830          101
Unit, investment trusts and ETFs          880           48
Hedge funds & private equity              760           42

                                       12,385          682

Source: Table B1 Bank Stats and Burrows and Low (2015).